asterisk/res/res_pjsip_stir_shaken.c

517 lines
17 KiB
C
Raw Permalink Normal View History

/*
* Asterisk -- An open source telephony toolkit.
*
* Copyright (C) 2020, Sangoma Technologies Corporation
*
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
* Ben Ford <bford@sangoma.com>
*
* See http://www.asterisk.org for more information about
* the Asterisk project. Please do not directly contact
* any of the maintainers of this project for assistance;
* the project provides a web site, mailing lists and IRC
* channels for your use.
*
* This program is free software, distributed under the terms of
* the GNU General Public License Version 2. See the LICENSE file
* at the top of the source tree.
*/
/*** MODULEINFO
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
<depend>pjproject</depend>
<depend>res_pjsip</depend>
<depend>res_pjsip_session</depend>
<depend>res_stir_shaken</depend>
<support_level>core</support_level>
***/
#include "asterisk.h"
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
#define _TRACE_PREFIX_ "pjss",__LINE__, ""
#include "asterisk/callerid.h"
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
#include "asterisk/res_pjsip.h"
#include "asterisk/res_pjsip_session.h"
#include "asterisk/module.h"
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
#include "asterisk/rtp_engine.h"
#include "asterisk/res_stir_shaken.h"
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
static const pj_str_t identity_hdr_str = { "Identity", 8 };
static const pj_str_t date_hdr_str = { "Date", 4 };
/* Response codes from RFC8224 */
enum sip_response_code {
SIP_RESPONSE_CODE_OK = 200,
SIP_RESPONSE_CODE_STALE_DATE = 403,
SIP_RESPONSE_CODE_USE_IDENTITY_HEADER = 428,
SIP_RESPONSE_CODE_BAD_IDENTITY_INFO = 436,
SIP_RESPONSE_CODE_UNSUPPORTED_CREDENTIAL = 437,
SIP_RESPONSE_CODE_INVALID_IDENTITY_HEADER = 438,
SIP_RESPONSE_CODE_USE_SUPPORTED_PASSPORT_FORMAT = 428,
SIP_RESPONSE_CODE_INTERNAL_ERROR = 500,
};
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
#define SIP_RESPONSE_CODE_OK_STR "OK"
/* Response strings from RFC8224 */
#define SIP_RESPONSE_CODE_STALE_DATE_STR "Stale Date"
#define SIP_RESPONSE_CODE_USE_IDENTITY_HEADER_STR "Use Identity Header"
#define SIP_RESPONSE_CODE_USE_SUPPORTED_PASSPORT_FORMAT_STR "Use Supported PASSporT Format"
#define SIP_RESPONSE_CODE_BAD_IDENTITY_INFO_STR "Bad Identity Info"
#define SIP_RESPONSE_CODE_UNSUPPORTED_CREDENTIAL_STR "Unsupported Credential"
#define SIP_RESPONSE_CODE_INVALID_IDENTITY_HEADER_STR "Invalid Identity Header"
#define SIP_RESPONSE_CODE_INTERNAL_ERROR_STR "Internal Error"
#define response_to_str(_code) \
case _code: \
return _code ## _STR;
static const char *sip_response_code_to_str(enum sip_response_code code)
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
{
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
switch (code) {
response_to_str(SIP_RESPONSE_CODE_OK)
response_to_str(SIP_RESPONSE_CODE_STALE_DATE)
response_to_str(SIP_RESPONSE_CODE_USE_IDENTITY_HEADER)
response_to_str(SIP_RESPONSE_CODE_BAD_IDENTITY_INFO)
response_to_str(SIP_RESPONSE_CODE_UNSUPPORTED_CREDENTIAL)
response_to_str(SIP_RESPONSE_CODE_INVALID_IDENTITY_HEADER)
default:
break;
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
}
return "";
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
#define translate_code(_vs_rc, _sip_rc) \
case AST_STIR_SHAKEN_VS_ ## _vs_rc: \
return SIP_RESPONSE_CODE_ ## _sip_rc;
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
static enum sip_response_code vs_code_to_sip_code(
enum ast_stir_shaken_vs_response_code vs_rc)
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
{
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
/*
* We want to use a switch/case statement here because
* it'll spit out an error if VS codes are added to the
* enum but aren't present here.
*/
switch (vs_rc) {
translate_code(SUCCESS, OK)
translate_code(DISABLED, OK)
translate_code(INVALID_ARGUMENTS, INTERNAL_ERROR)
translate_code(INTERNAL_ERROR, INTERNAL_ERROR)
translate_code(NO_IDENTITY_HDR, USE_IDENTITY_HEADER)
translate_code(NO_DATE_HDR, STALE_DATE)
translate_code(DATE_HDR_PARSE_FAILURE, STALE_DATE)
translate_code(DATE_HDR_EXPIRED, STALE_DATE)
translate_code(NO_JWT_HDR, INVALID_IDENTITY_HEADER)
translate_code(INVALID_OR_NO_X5U, INVALID_IDENTITY_HEADER)
translate_code(CERT_CACHE_MISS, INVALID_IDENTITY_HEADER)
translate_code(CERT_CACHE_INVALID, INVALID_IDENTITY_HEADER)
translate_code(CERT_CACHE_EXPIRED, INVALID_IDENTITY_HEADER)
translate_code(CERT_RETRIEVAL_FAILURE, BAD_IDENTITY_INFO)
translate_code(CERT_CONTENTS_INVALID, UNSUPPORTED_CREDENTIAL)
translate_code(CERT_NOT_TRUSTED, UNSUPPORTED_CREDENTIAL)
translate_code(CERT_DATE_INVALID, UNSUPPORTED_CREDENTIAL)
translate_code(CERT_NO_TN_AUTH_EXT, UNSUPPORTED_CREDENTIAL)
translate_code(CERT_NO_SPC_IN_TN_AUTH_EXT, UNSUPPORTED_CREDENTIAL)
translate_code(NO_RAW_KEY, UNSUPPORTED_CREDENTIAL)
translate_code(SIGNATURE_VALIDATION, INVALID_IDENTITY_HEADER)
translate_code(NO_IAT, INVALID_IDENTITY_HEADER)
translate_code(IAT_EXPIRED, STALE_DATE)
translate_code(INVALID_OR_NO_PPT, INVALID_IDENTITY_HEADER)
translate_code(INVALID_OR_NO_ALG, INVALID_IDENTITY_HEADER)
translate_code(INVALID_OR_NO_TYP, INVALID_IDENTITY_HEADER)
translate_code(INVALID_OR_NO_ATTEST, INVALID_IDENTITY_HEADER)
translate_code(NO_ORIGID, INVALID_IDENTITY_HEADER)
translate_code(NO_ORIG_TN, INVALID_IDENTITY_HEADER)
translate_code(NO_DEST_TN, INVALID_IDENTITY_HEADER)
translate_code(INVALID_HEADER, INVALID_IDENTITY_HEADER)
translate_code(INVALID_GRANT, INVALID_IDENTITY_HEADER)
translate_code(INVALID_OR_NO_GRANTS, INVALID_IDENTITY_HEADER)
translate_code(CID_ORIG_TN_MISMATCH, INVALID_IDENTITY_HEADER)
translate_code(RESPONSE_CODE_MAX, INVALID_IDENTITY_HEADER)
}
return 500;
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
enum process_failure_rc {
PROCESS_FAILURE_CONTINUE = 0,
PROCESS_FAILURE_REJECT,
PROCESS_FAILURE_SYSTEM_FAILURE,
};
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
static void reject_incoming_call(struct ast_sip_session *session,
enum sip_response_code response_code)
{
ast_sip_session_terminate(session, response_code);
ast_hangup(session->channel);
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
static enum process_failure_rc process_failure(struct ast_stir_shaken_vs_ctx *ctx,
const char *caller_id, struct ast_sip_session *session,
pjsip_rx_data *rdata, enum ast_stir_shaken_vs_response_code vs_rc)
{
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
enum sip_response_code response_code = vs_code_to_sip_code(vs_rc);
pj_str_t response_str;
const char *response_string =
sip_response_code_to_str(response_code);
enum stir_shaken_failure_action_enum failure_action =
ast_stir_shaken_vs_get_failure_action(ctx);
const char *tag = ast_sip_session_get_name(session);
SCOPE_ENTER(1, "%s: FA: %d RC: %d\n", tag,
failure_action, response_code);
pj_cstr(&response_str, response_string);
if (failure_action == stir_shaken_failure_action_REJECT_REQUEST) {
reject_incoming_call(session, response_code);
SCOPE_EXIT_RTN_VALUE(PROCESS_FAILURE_REJECT,
"%s: Rejecting request and terminating session\n",
tag);
}
ast_stir_shaken_vs_ctx_set_response_code(ctx, vs_rc);
ast_stir_shaken_add_result_to_channel(ctx);
if (failure_action == stir_shaken_failure_action_CONTINUE_RETURN_REASON) {
int rc = ast_sip_session_add_reason_header(session,
ast_stir_shaken_vs_get_use_rfc9410_responses(ctx) ? "STIR" : "SIP",
response_code, response_str.ptr);
if (rc != 0) {
SCOPE_EXIT_RTN_VALUE(PROCESS_FAILURE_SYSTEM_FAILURE,
"%s: Failed to add Reason header\n", tag);
}
SCOPE_EXIT_RTN_VALUE(PROCESS_FAILURE_CONTINUE,
"%s: Attaching reason code to session\n", tag);
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
SCOPE_EXIT_RTN_VALUE(PROCESS_FAILURE_CONTINUE,
"%s: Continuing\n", tag);
}
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
/*!
* \internal
* \brief Session supplement callback on an incoming INVITE request
*
* When we receive an INVITE, check it for STIR/SHAKEN information and
* decide what to do from there
*
* \param session The session that has received an INVITE
* \param rdata The incoming INVITE
*/
static int stir_shaken_incoming_request(struct ast_sip_session *session, pjsip_rx_data *rdata)
{
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
RAII_VAR(struct ast_stir_shaken_vs_ctx *, ctx, NULL, ao2_cleanup);
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
RAII_VAR(char *, header, NULL, ast_free);
RAII_VAR(char *, payload, NULL, ast_free);
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
char *identity_hdr_val;
char *date_hdr_val;
char *caller_id = session->id.number.str;
const char *session_name = ast_sip_session_get_name(session);
struct ast_channel *chan = session->channel;
enum ast_stir_shaken_vs_response_code vs_rc;
enum process_failure_rc p_rc;
SCOPE_ENTER(1, "%s: Enter\n", session_name);
if (!session) {
SCOPE_EXIT_LOG_RTN_VALUE(1, LOG_ERROR, "No session\n");
}
if (!session->channel) {
SCOPE_EXIT_LOG_RTN_VALUE(1, LOG_ERROR, "%s: No channel\n", session_name);
}
if (!rdata) {
SCOPE_EXIT_LOG_RTN_VALUE(1, LOG_ERROR, "%s: No rdata\n", session_name);
}
/* Check if this is a reinvite. If it is, we don't need to do anything */
if (rdata->msg_info.to->tag.slen) {
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
SCOPE_EXIT_RTN_VALUE(0, "%s: Reinvite. No action needed\n", session_name);
}
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
/*
* Shortcut: If there's no callerid or profile name,
* just bail now.
*/
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
if (ast_strlen_zero(caller_id)
|| ast_strlen_zero(session->endpoint->stir_shaken_profile)) {
SCOPE_EXIT_RTN_VALUE(0, "%s: No callerid or profile name. No action needed\n", session_name);
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
vs_rc = ast_stir_shaken_vs_ctx_create(caller_id, chan,
session->endpoint->stir_shaken_profile,
session_name, &ctx);
if (vs_rc == AST_STIR_SHAKEN_VS_DISABLED) {
SCOPE_EXIT_RTN_VALUE(0, "%s: VS Disabled\n", session_name);
} else if (vs_rc != AST_STIR_SHAKEN_VS_SUCCESS) {
reject_incoming_call(session, 500);
SCOPE_EXIT_RTN_VALUE(1, "%s: Unable to create context. Call terminated\n",
session_name);
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
identity_hdr_val = ast_sip_rdata_get_header_value(rdata, identity_hdr_str);
if (ast_strlen_zero(identity_hdr_val)) {
p_rc = process_failure(ctx, caller_id, session, rdata,
AST_STIR_SHAKEN_VS_NO_IDENTITY_HDR);
if (p_rc == PROCESS_FAILURE_CONTINUE) {
SCOPE_EXIT_RTN_VALUE(0, "%s: No Identity header found. Call continuing\n",
session_name);
}
SCOPE_EXIT_LOG_RTN_VALUE(1, LOG_ERROR, "%s: No Identity header found. Call terminated\n",
session_name);
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
vs_rc = ast_stir_shaken_vs_ctx_add_identity_hdr(ctx, identity_hdr_val);
if (vs_rc != AST_STIR_SHAKEN_VS_SUCCESS) {
reject_incoming_call(session, 500);
SCOPE_EXIT_LOG_RTN_VALUE(1, LOG_ERROR, "%s: Unable to add Identity header. Call terminated.\n",
session_name);
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
date_hdr_val = ast_sip_rdata_get_header_value(rdata, date_hdr_str);
if (ast_strlen_zero(date_hdr_val)) {
p_rc = process_failure(ctx, caller_id, session, rdata,
AST_STIR_SHAKEN_VS_NO_DATE_HDR);
if (p_rc == PROCESS_FAILURE_CONTINUE) {
SCOPE_EXIT_RTN_VALUE(0, "%s: No Date header found. Call continuing\n",
session_name);
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
SCOPE_EXIT_LOG_RTN_VALUE(1, LOG_ERROR, "%s: No Date header found. Call terminated\n",
session_name);
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
ast_stir_shaken_vs_ctx_add_date_hdr(ctx, date_hdr_val);
if (vs_rc != AST_STIR_SHAKEN_VS_SUCCESS) {
reject_incoming_call(session, 500);
SCOPE_EXIT_LOG_RTN_VALUE(1, LOG_ERROR, "%s: Unable to add Date header. Call terminated.\n",
session_name);
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
vs_rc = ast_stir_shaken_vs_verify(ctx);
if (vs_rc != AST_STIR_SHAKEN_VS_SUCCESS) {
p_rc = process_failure(ctx, caller_id, session, rdata, vs_rc);
if (p_rc == PROCESS_FAILURE_CONTINUE) {
SCOPE_EXIT_RTN_VALUE(0, "%s: Verification failed. Call continuing\n",
session_name);
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
SCOPE_EXIT_LOG_RTN_VALUE(1, LOG_ERROR, "%s: Verification failed. Call terminated\n",
session_name);
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
ast_stir_shaken_add_result_to_channel(ctx);
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
SCOPE_EXIT_RTN_VALUE(0, "Passed\n");
}
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
static void add_fingerprints_if_present(struct ast_sip_session *session,
struct ast_stir_shaken_as_ctx *ctx)
{
struct ast_sip_session_media_state *ms = session->pending_media_state;
struct ast_sip_session_media *m = NULL;
struct ast_rtp_engine_dtls *d = NULL;
enum ast_rtp_dtls_hash h;
int i;
const char *tag = ast_sip_session_get_name(session);
size_t count = AST_VECTOR_SIZE(&ms->sessions);
SCOPE_ENTER(4, "%s: Check %zu media sessions for fingerprints\n",
tag, count);
if (!ast_stir_shaken_as_ctx_wants_fingerprints(ctx)) {
SCOPE_EXIT_RTN("%s: Fingerprints not needed\n", tag);
}
for (i = 0; i < count; i++) {
const char *f;
m = AST_VECTOR_GET(&ms->sessions, i);
if (!m|| !m->rtp) {
ast_trace(1, "Session: %d: No session or rtp instance\n", i);
continue;
}
d = ast_rtp_instance_get_dtls(m->rtp);
h = d->get_fingerprint_hash(m->rtp);
f = d->get_fingerprint(m->rtp);
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
ast_stir_shaken_as_ctx_add_fingerprint(ctx,
h == AST_RTP_DTLS_HASH_SHA256 ? "sha-256" : "sha-1", f);
}
SCOPE_EXIT_RTN("%s: Done\n", tag);
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
static char *get_dest_tn(pjsip_tx_data *tdata, const char *tag)
{
pjsip_fromto_hdr *to;
pjsip_sip_uri *uri;
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
char *dest_tn = NULL;
SCOPE_ENTER(4, "%s: Enter\n", tag);
to = pjsip_msg_find_hdr(tdata->msg, PJSIP_H_TO, NULL);
if (!to) {
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
SCOPE_EXIT_RTN_VALUE(NULL, "%s: Failed to find To header\n", tag);
}
uri = pjsip_uri_get_uri(to->uri);
if (!uri) {
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
SCOPE_EXIT_RTN_VALUE(NULL,
"%s: Failed to retrieve URI from To header\n", tag);
}
dest_tn = ast_malloc(uri->user.slen + 1);
if (!dest_tn) {
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
SCOPE_EXIT_RTN_VALUE(NULL,
"%s: Failed to allocate memory for dest_tn\n", tag);
}
/* Remove everything except 0-9, *, and # in telephone number according to RFC 8224
* (required by RFC 8225 as part of canonicalization) */
{
int i;
const char *s = uri->user.ptr;
char *new_tn = dest_tn;
/* We're only removing characters, if anything, so the buffer is guaranteed to be large enough */
for (i = 0; i < uri->user.slen; i++) {
if (isdigit(*s) || *s == '#' || *s == '*') { /* Only characters allowed */
*new_tn++ = *s;
}
s++;
}
*new_tn = '\0';
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
ast_trace(2, "Canonicalized telephone number " PJSTR_PRINTF_SPEC " -> %s\n",
PJSTR_PRINTF_VAR(uri->user), dest_tn);
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
SCOPE_EXIT_RTN_VALUE(dest_tn, "%s: Done\n", tag);
}
static void add_date_header(const struct ast_sip_session *session, pjsip_tx_data *tdata)
{
pjsip_fromto_hdr *old_date;
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
const char *session_name = ast_sip_session_get_name(session);
SCOPE_ENTER(1, "%s: Enter\n", session_name);
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
old_date = pjsip_msg_find_hdr_by_name(tdata->msg, &date_hdr_str, NULL);
if (old_date) {
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
SCOPE_EXIT_RTN("Found existing Date header, no need to add one\n");
}
ast_sip_add_date_header(tdata);
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
SCOPE_EXIT_RTN("Done\n");
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
static void stir_shaken_outgoing_request(struct ast_sip_session *session,
pjsip_tx_data *tdata)
{
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
struct ast_party_id effective_id;
struct ast_party_id connected_id;
pjsip_generic_string_hdr *old_identity;
pjsip_generic_string_hdr *identity_hdr;
pj_str_t identity_val;
char *dest_tn;
char *identity_str;
struct ast_stir_shaken_as_ctx *ctx = NULL;
enum ast_stir_shaken_as_response_code as_rc;
const char *session_name = ast_sip_session_get_name(session);
SCOPE_ENTER(1, "%s: Enter\n", session_name);
if (!session) {
SCOPE_EXIT_LOG_RTN(LOG_ERROR, "No session\n");
}
if (!session->channel) {
SCOPE_EXIT_LOG_RTN(LOG_ERROR, "%s: No channel\n", session_name);
}
if (!tdata) {
SCOPE_EXIT_LOG_RTN(LOG_ERROR, "%s: No tdata\n", session_name);
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
old_identity = pjsip_msg_find_hdr_by_name(tdata->msg, &identity_hdr_str, NULL);
if (old_identity) {
SCOPE_EXIT_RTN("Found an existing Identity header\n");
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
dest_tn = get_dest_tn(tdata, session_name);
if (!dest_tn) {
SCOPE_EXIT_LOG_RTN(LOG_ERROR, "%s: Unable to find destination tn\n",
session_name);
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
ast_party_id_init(&connected_id);
ast_channel_lock(session->channel);
effective_id = ast_channel_connected_effective_id(session->channel);
ast_party_id_copy(&connected_id, &effective_id);
ast_channel_unlock(session->channel);
if (!ast_sip_can_present_connected_id(session, &connected_id)) {
ast_free(dest_tn);
ast_party_id_free(&connected_id);
SCOPE_EXIT_RTN("Unable to get caller id\n");
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
as_rc = ast_stir_shaken_as_ctx_create(connected_id.number.str,
dest_tn, session->channel,
session->endpoint->stir_shaken_profile,
session_name, &ctx);
ast_free(dest_tn);
ast_party_id_free(&connected_id);
if (as_rc == AST_STIR_SHAKEN_AS_DISABLED) {
SCOPE_EXIT_RTN("%s: AS Disabled\n", session_name);
} else if (as_rc != AST_STIR_SHAKEN_AS_SUCCESS) {
SCOPE_EXIT_RTN("%s: Unable to create context\n",
session_name);
}
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
add_date_header(session, tdata);
Stir/Shaken Refactor Why do we need a refactor? The original stir/shaken implementation was started over 3 years ago when little was understood about practical implementation. The result was an implementation that wouldn't actually interoperate with any other stir-shaken implementations. There were also a number of stir-shaken features and RFC requirements that were never implemented such as TNAuthList certificate validation, sending Reason headers in SIP responses when verification failed but we wished to continue the call, and the ability to send Media Key(mky) grants in the Identity header when the call involved DTLS. Finally, there were some performance concerns around outgoing calls and selection of the correct certificate and private key. The configuration was keyed by an arbitrary name which meant that for every outgoing call, we had to scan the entire list of configured TNs to find the correct cert to use. With only a few TNs configured, this wasn't an issue but if you have a thousand, it could be. What's changed? * Configuration objects have been refactored to be clearer about their uses and to fix issues. * The "general" object was renamed to "verification" since it contains parameters specific to the incoming verification process. It also never handled ca_path and crl_path correctly. * A new "attestation" object was added that controls the outgoing attestation process. It sets default certificates, keys, etc. * The "certificate" object was renamed to "tn" and had it's key change to telephone number since outgoing call attestation needs to look up certificates by telephone number. * The "profile" object had more parameters added to it that can override default parameters specified in the "attestation" and "verification" objects. * The "store" object was removed altogther as it was never implemented. * We now use libjwt to create outgoing Identity headers and to parse and validate signatures on incoming Identiy headers. Our previous custom implementation was much of the source of the interoperability issues. * General code cleanup and refactor. * Moved things to better places. * Separated some of the complex functions to smaller ones. * Using context objects rather than passing tons of parameters in function calls. * Removed some complexity and unneeded encapsuation from the config objects. Resolves: #351 Resolves: #46 UserNote: Asterisk's stir-shaken feature has been refactored to correct interoperability, RFC compliance, and performance issues. See https://docs.asterisk.org/Deployment/STIR-SHAKEN for more information. UpgradeNote: The stir-shaken refactor is a breaking change but since it's not working now we don't think it matters. The stir_shaken.conf file has changed significantly which means that existing ones WILL need to be changed. The stir_shaken.conf.sample file in configs/samples/ has quite a bit more information. This is also an ABI breaking change since some of the existing objects needed to be changed or removed, and new ones added. Additionally, if res_stir_shaken is enabled in menuselect, you'll need to either have the development package for libjwt v1.15.3 installed or use the --with-libjwt-bundled option with ./configure.
2023-10-26 16:27:35 +00:00
add_fingerprints_if_present(session, ctx);
as_rc = ast_stir_shaken_attest(ctx, &identity_str);
if (as_rc != AST_STIR_SHAKEN_AS_SUCCESS) {
ao2_cleanup(ctx);
SCOPE_EXIT_LOG(LOG_ERROR,
"%s: Failed to create attestation\n", session_name);
}
ast_trace(1, "%s: Identity header: %s\n", session_name, identity_str);
identity_val = pj_str(identity_str);
identity_hdr = pjsip_generic_string_hdr_create(tdata->pool, &identity_hdr_str, &identity_val);
ast_free(identity_str);
if (!identity_hdr) {
ao2_cleanup(ctx);
SCOPE_EXIT_LOG_RTN(LOG_ERROR,
"%s: Unable to create Identity header\n", session_name);
}
pjsip_msg_add_hdr(tdata->msg, (pjsip_hdr *)identity_hdr);
ao2_cleanup(ctx);
SCOPE_EXIT_RTN("Done\n");
}
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
static struct ast_sip_session_supplement stir_shaken_supplement = {
.method = "INVITE",
.priority = AST_SIP_SUPPLEMENT_PRIORITY_CHANNEL + 1, /* Run AFTER channel creation */
.incoming_request = stir_shaken_incoming_request,
.outgoing_request = stir_shaken_outgoing_request,
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
};
static int unload_module(void)
{
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
ast_sip_session_unregister_supplement(&stir_shaken_supplement);
return 0;
}
static int load_module(void)
{
res_stir_shaken: Add inbound INVITE support. Integrated STIR/SHAKEN support with incoming INVITES. Upon receiving an INVITE, the Identity header is retrieved, parsing the message to verify the signature. If any of the parsing fails, AST_STIR_SHAKEN_VERIFY_NOT_PRESENT will be added to the channel for this caller ID. If verification itself fails, AST_STIR_SHAKEN_VERIFY_SIGNATURE_FAILED will be added. If anything in the payload does not line up with the SIP signaling, AST_STIR_SHAKEN_VERIFY_MISMATCH will be added. If all of the above steps pass, then AST_STIR_SHAKEN_VERIFY_PASSED will be added, completing the verification process. A new config option has been added to the general section for stir_shaken.conf. "signature_timeout" is the amount of time a signature will be considered valid. If an INVITE is received and the amount of time between when it was received and when it was signed is greater than signature_timeout, verification will fail. Some changes were also made to signing and verification. There was an error where the whole JSON string was being signed rather than the header combined with the payload. This has been changed to sign the correct thing. Verification has been changed to do this as well, and the unit tests have been updated to reflect these changes. A couple of utility functions have also been added. One decodes a BASE64 string and returns the decoded string, doing all the length calculations for you. The other retrieves a string value from a header in a rdata object. Change-Id: I855f857be3d1c63b64812ac35d9ce0534085b913
2020-05-19 19:46:45 +00:00
ast_sip_session_register_supplement(&stir_shaken_supplement);
return AST_MODULE_LOAD_SUCCESS;
}
#undef AST_BUILDOPT_SUM
#define AST_BUILDOPT_SUM ""
AST_MODULE_INFO(ASTERISK_GPL_KEY, AST_MODFLAG_GLOBAL_SYMBOLS | AST_MODFLAG_LOAD_ORDER, "PJSIP STIR/SHAKEN Module for Asterisk",
.support_level = AST_MODULE_SUPPORT_CORE,
.load = load_module,
.unload = unload_module,
.load_pri = AST_MODPRI_DEFAULT,
.requires = "res_pjsip,res_pjsip_session,res_stir_shaken",
);