asterisk/main/sdp_srtp.c

442 lines
11 KiB
C
Raw Normal View History

/*
* Asterisk -- An open source telephony toolkit.
*
* Copyright (C) 2006 - 2007, Mikael Magnusson
*
* Mikael Magnusson <mikma@users.sourceforge.net>
*
* 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.
*/
/*! \file
*
* \brief SRTP and SDP Security descriptions
*
* Specified in RFC 3711
* Specified in RFC 4568
*
* \author Mikael Magnusson <mikma@users.sourceforge.net>
*/
/*** MODULEINFO
<support_level>core</support_level>
***/
#include "asterisk.h"
git migration: Refactor the ASTERISK_FILE_VERSION macro Git does not support the ability to replace a token with a version string during check-in. While it does have support for replacing a token on clone, this is somewhat sub-optimal: the token is replaced with the object hash, which is not particularly easy for human consumption. What's more, in practice, the source file version was often not terribly useful. Generally, when triaging bugs, the overall version of Asterisk is far more useful than an individual SVN version of a file. As a result, this patch removes Asterisk's support for showing source file versions. Specifically, it does the following: * Rename ASTERISK_FILE_VERSION macro to ASTERISK_REGISTER_FILE, and remove passing the version in with the macro. Other facilities than 'core show file version' make use of the file names, such as setting a debug level only on a specific file. As such, the act of registering source files with the Asterisk core still has use. The macro rename now reflects the new macro purpose. * main/asterisk: - Refactor the file_version structure to reflect that it no longer tracks a version field. - Remove the "core show file version" CLI command. Without the file version, it is no longer useful. - Remove the ast_file_version_find function. The file version is no longer tracked. - Rename ast_register_file_version/ast_unregister_file_version to ast_register_file/ast_unregister_file, respectively. * main/manager: Remove value from the Version key of the ModuleCheck Action. The actual key itself has not been removed, as doing so would absolutely constitute a backwards incompatible change. However, since the file version is no longer tracked, there is no need to attempt to include it in the Version key. * UPGRADE: Add notes for: - Modification to the ModuleCheck AMI Action - Removal of the "core show file version" CLI command Change-Id: I6cf0ff280e1668bf4957dc21f32a5ff43444a40e
2015-04-12 02:38:22 +00:00
ASTERISK_REGISTER_FILE()
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
#include <math.h>
#include "asterisk/options.h"
#include "asterisk/utils.h"
#include "asterisk/sdp_srtp.h"
#define SRTP_MASTER_LEN 30
#define SRTP_MASTERKEY_LEN 16
#define SRTP_MASTERSALT_LEN ((SRTP_MASTER_LEN) - (SRTP_MASTERKEY_LEN))
#define SRTP_MASTER_LEN64 (((SRTP_MASTER_LEN) * 8 + 5) / 6 + 1)
extern struct ast_srtp_res *res_srtp;
extern struct ast_srtp_policy_res *res_srtp_policy;
struct ast_sdp_srtp *ast_sdp_srtp_alloc(void)
{
if (!ast_rtp_engine_srtp_is_registered()) {
ast_debug(1, "No SRTP module loaded, can't setup SRTP session.\n");
return NULL;
}
return ast_calloc(1, sizeof(struct ast_sdp_srtp));
}
void ast_sdp_srtp_destroy(struct ast_sdp_srtp *srtp)
{
if (srtp->crypto) {
ast_sdp_crypto_destroy(srtp->crypto);
}
srtp->crypto = NULL;
ast_free(srtp);
}
struct ast_sdp_crypto {
char *a_crypto;
unsigned char local_key[SRTP_MASTER_LEN];
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
int tag;
char local_key64[SRTP_MASTER_LEN64];
unsigned char remote_key[SRTP_MASTER_LEN];
};
static int set_crypto_policy(struct ast_srtp_policy *policy, int suite_val, const unsigned char *master_key, unsigned long ssrc, int inbound);
void ast_sdp_crypto_destroy(struct ast_sdp_crypto *crypto)
{
ast_free(crypto->a_crypto);
crypto->a_crypto = NULL;
ast_free(crypto);
}
struct ast_sdp_crypto *ast_sdp_crypto_alloc(void)
{
struct ast_sdp_crypto *p;
int key_len;
unsigned char remote_key[SRTP_MASTER_LEN];
if (!ast_rtp_engine_srtp_is_registered()) {
return NULL;
}
if (!(p = ast_calloc(1, sizeof(*p)))) {
return NULL;
}
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
p->tag = 1;
if (res_srtp->get_random(p->local_key, sizeof(p->local_key)) < 0) {
ast_sdp_crypto_destroy(p);
return NULL;
}
ast_base64encode(p->local_key64, p->local_key, SRTP_MASTER_LEN, sizeof(p->local_key64));
key_len = ast_base64decode(remote_key, p->local_key64, sizeof(remote_key));
if (key_len != SRTP_MASTER_LEN) {
ast_log(LOG_ERROR, "base64 encode/decode bad len %d != %d\n", key_len, SRTP_MASTER_LEN);
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
ast_sdp_crypto_destroy(p);
return NULL;
}
if (memcmp(remote_key, p->local_key, SRTP_MASTER_LEN)) {
ast_log(LOG_ERROR, "base64 encode/decode bad key\n");
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
ast_sdp_crypto_destroy(p);
return NULL;
}
ast_debug(1 , "local_key64 %s len %zu\n", p->local_key64, strlen(p->local_key64));
return p;
}
static int set_crypto_policy(struct ast_srtp_policy *policy, int suite_val, const unsigned char *master_key, unsigned long ssrc, int inbound)
{
const unsigned char *master_salt = NULL;
if (!ast_rtp_engine_srtp_is_registered()) {
return -1;
}
master_salt = master_key + SRTP_MASTERKEY_LEN;
if (res_srtp_policy->set_master_key(policy, master_key, SRTP_MASTERKEY_LEN, master_salt, SRTP_MASTERSALT_LEN) < 0) {
return -1;
}
if (res_srtp_policy->set_suite(policy, suite_val)) {
ast_log(LOG_WARNING, "Could not set remote SRTP suite\n");
return -1;
}
res_srtp_policy->set_ssrc(policy, ssrc, inbound);
return 0;
}
static int crypto_activate(struct ast_sdp_crypto *p, int suite_val, unsigned char *remote_key, struct ast_rtp_instance *rtp)
{
struct ast_srtp_policy *local_policy = NULL;
struct ast_srtp_policy *remote_policy = NULL;
struct ast_rtp_instance_stats stats = {0,};
int res = -1;
if (!ast_rtp_engine_srtp_is_registered()) {
return -1;
}
if (!p) {
return -1;
}
if (!(local_policy = res_srtp_policy->alloc())) {
return -1;
}
if (!(remote_policy = res_srtp_policy->alloc())) {
goto err;
}
if (ast_rtp_instance_get_stats(rtp, &stats, AST_RTP_INSTANCE_STAT_LOCAL_SSRC)) {
goto err;
}
if (set_crypto_policy(local_policy, suite_val, p->local_key, stats.local_ssrc, 0) < 0) {
goto err;
}
if (set_crypto_policy(remote_policy, suite_val, remote_key, 0, 1) < 0) {
goto err;
}
/* Add the SRTP policies */
if (ast_rtp_instance_add_srtp_policy(rtp, remote_policy, local_policy)) {
ast_log(LOG_WARNING, "Could not set SRTP policies\n");
goto err;
}
ast_debug(1 , "SRTP policy activated\n");
res = 0;
err:
if (local_policy) {
res_srtp_policy->destroy(local_policy);
}
if (remote_policy) {
res_srtp_policy->destroy(remote_policy);
}
return res;
}
int ast_sdp_crypto_process(struct ast_rtp_instance *rtp, struct ast_sdp_srtp *srtp, const char *attr)
{
char *str = NULL;
char *tag = NULL;
char *suite = NULL;
char *key_params = NULL;
char *key_param = NULL;
char *session_params = NULL;
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
char *key_salt = NULL; /* The actual master key and key salt */
char *lifetime = NULL; /* Key lifetime (# of RTP packets) */
char *mki = NULL; /* Master Key Index */
int found = 0;
int key_len = 0;
int suite_val = 0;
unsigned char remote_key[SRTP_MASTER_LEN];
int taglen = 0;
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
double sdes_lifetime;
struct ast_sdp_crypto *crypto = srtp->crypto;
if (!ast_rtp_engine_srtp_is_registered()) {
return -1;
}
str = ast_strdupa(attr);
tag = strsep(&str, " ");
suite = strsep(&str, " ");
key_params = strsep(&str, " ");
session_params = strsep(&str, " ");
if (!tag || !suite) {
ast_log(LOG_WARNING, "Unrecognized crypto attribute a=%s\n", attr);
return -1;
}
/* RFC4568 9.1 - tag is 1-9 digits, greater than zero */
if (sscanf(tag, "%30d", &crypto->tag) != 1 || crypto->tag <= 0 || crypto->tag > 999999999) {
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
ast_log(LOG_WARNING, "Unacceptable a=crypto tag: %s\n", tag);
return -1;
}
if (!ast_strlen_zero(session_params)) {
ast_log(LOG_WARNING, "Unsupported crypto parameters: %s\n", session_params);
return -1;
}
if (!strcmp(suite, "AES_CM_128_HMAC_SHA1_80")) {
suite_val = AST_AES_CM_128_HMAC_SHA1_80;
ast_set_flag(srtp, AST_SRTP_CRYPTO_TAG_80);
taglen = 80;
} else if (!strcmp(suite, "AES_CM_128_HMAC_SHA1_32")) {
suite_val = AST_AES_CM_128_HMAC_SHA1_32;
ast_set_flag(srtp, AST_SRTP_CRYPTO_TAG_32);
taglen = 32;
} else {
ast_log(LOG_WARNING, "Unsupported crypto suite: %s\n", suite);
return -1;
}
while ((key_param = strsep(&key_params, ";"))) {
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
unsigned int n_lifetime;
char *method = NULL;
char *info = NULL;
method = strsep(&key_param, ":");
info = strsep(&key_param, ";");
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
sdes_lifetime = 0;
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
if (strcmp(method, "inline")) {
continue;
}
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
key_salt = strsep(&info, "|");
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
/* The next parameter can be either lifetime or MKI */
lifetime = strsep(&info, "|");
if (!lifetime) {
found = 1;
break;
}
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
mki = strchr(lifetime, ':');
if (mki) {
mki = lifetime;
lifetime = NULL;
} else {
mki = strsep(&info, "|");
}
if (mki && *mki != '1') {
ast_log(LOG_NOTICE, "Crypto MKI handling is not supported: ignoring attribute %s\n", attr);
continue;
}
if (lifetime) {
if (!strncmp(lifetime, "2^", 2)) {
char *lifetime_val = lifetime + 2;
/* Exponential lifetime */
if (sscanf(lifetime_val, "%30u", &n_lifetime) != 1) {
ast_log(LOG_NOTICE, "Failed to parse lifetime value in crypto attribute: %s\n", attr);
continue;
}
if (n_lifetime > 48) {
/* Yeah... that's a bit big. */
ast_log(LOG_NOTICE, "Crypto lifetime exponent of '%u' is a bit large; using 48\n", n_lifetime);
n_lifetime = 48;
}
sdes_lifetime = pow(2, n_lifetime);
} else {
/* Decimal lifetime */
if (sscanf(lifetime, "%30u", &n_lifetime) != 1) {
ast_log(LOG_NOTICE, "Failed to parse lifetime value in crypto attribute: %s\n", attr);
continue;
}
sdes_lifetime = n_lifetime;
}
/* Accept anything above 10 hours. Less than 10; reject. */
if (sdes_lifetime < 1800000) {
ast_log(LOG_NOTICE, "Rejecting crypto attribute '%s': lifetime '%f' too short\n", attr, sdes_lifetime);
continue;
}
}
ast_debug(2, "Crypto attribute '%s' accepted with lifetime '%f', MKI '%s'\n",
attr, sdes_lifetime, mki ? mki : "-");
found = 1;
break;
}
if (!found) {
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
ast_log(LOG_NOTICE, "SRTP crypto offer not acceptable: '%s'\n", attr);
return -1;
}
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
key_len = ast_base64decode(remote_key, key_salt, sizeof(remote_key));
if (key_len != SRTP_MASTER_LEN) {
ast_log(LOG_WARNING, "SRTP descriptions key length '%d' != master length '%d'\n",
key_len, SRTP_MASTER_LEN);
return -1;
}
if (!memcmp(crypto->remote_key, remote_key, sizeof(crypto->remote_key))) {
ast_debug(1, "SRTP remote key unchanged; maintaining current policy\n");
ast_set_flag(srtp, AST_SRTP_CRYPTO_OFFER_OK);
return 0;
}
memcpy(crypto->remote_key, remote_key, sizeof(crypto->remote_key));
if (crypto_activate(crypto, suite_val, remote_key, rtp) < 0) {
return -1;
}
/* Finally, rebuild the crypto line */
if (ast_sdp_crypto_build_offer(crypto, taglen)) {
return -1;
}
ast_set_flag(srtp, AST_SRTP_CRYPTO_OFFER_OK);
return 0;
}
int ast_sdp_crypto_build_offer(struct ast_sdp_crypto *p, int taglen)
{
/* Rebuild the crypto line */
if (p->a_crypto) {
ast_free(p->a_crypto);
}
channels/sip/sdp_crypto: Handle SRTP keys negotiated with key lifetime/MKI Prior to this patch, SDP offers negotiating SDES-SRTP crypto attributes would be rejected if those crypto attributes contained either a key lifetime or a MKI parameter. While from a theoretical point of view this was defensible - Asterisk does not support key lifetimes or multiple crypto keys - from a practical point of view, this is quite a problem. A large number of endpoints offer lifetimes/MKI, which Asterisk can tolerate so long as it doesn't actually have to support anything more than a single key or refresh the key. In reality, this is (so far as we've seen) always the case. This patch is a forward port of Olle's work in the lingon-srtp-key-lifetime-1.8 branch. To quote Olle from ASTERISK-17721, it handles lifetime/MKI parameters in the following fashion: > The Lingon branch now handle lifetime and MKI parameters. > > We only accept lifetimes up to max for the crypto and higher than 10 hours > for packetization of 20 ms (50 pps). > > We only handle MKI with index 1. > > We do not really bother with counting packets and reinviting at end of > lifetime, so the min of 10 hours kind of takes care of most calls. If there > are longer ones, we rely on the other side for re-invites. > > It's still not perfect, but I personally think this is an improvement. A > configuration option for minimum lifetime accepted could be added. When the patch was ported forward, I decided against adding a configuration option as Olle's handling was more than sufficient for every case I've seen come through the issue tracker or through interoperability testing. We can revisit that decision if it proves to be false. A few small other tweaks were made to the surrounding code to reduce indentation and provide better type safety for the 'tag' parameter. Review: https://reviewboard.asterisk.org/r/4419/ Review: https://reviewboard.asterisk.org/r/4418/ ASTERISK-17721 #close Reported by: Terry Wilson ASTERISK-17899 #close Reported by: Dwayne Hubbard patches: lingon-srtp-key-lifetime-1.8.diff uploaded by oej (License 5267) ASTERISK-20233 Reported by: tootai ASTERISK-22748 Reported by: Alejandro Mejia ........ Merged revisions 432239 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 432258 from http://svn.asterisk.org/svn/asterisk/branches/13 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@432259 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2015-02-25 21:42:39 +00:00
if (ast_asprintf(&p->a_crypto, "%d AES_CM_128_HMAC_SHA1_%i inline:%s",
p->tag, taglen, p->local_key64) == -1) {
ast_log(LOG_ERROR, "Could not allocate memory for crypto line\n");
return -1;
}
ast_debug(1, "Crypto line: a=crypto:%s\n", p->a_crypto);
return 0;
}
const char *ast_sdp_srtp_get_attrib(struct ast_sdp_srtp *srtp, int dtls_enabled, int default_taglen_32)
{
int taglen = default_taglen_32 ? 32 : 80;
if (!srtp) {
return NULL;
}
/* Set encryption properties */
if (!srtp->crypto) {
srtp->crypto = ast_sdp_crypto_alloc();
}
if (dtls_enabled) {
/* If DTLS-SRTP is enabled the key details will be pulled from TLS */
return NULL;
}
/* set the key length based on INVITE or settings */
if (ast_test_flag(srtp, AST_SRTP_CRYPTO_TAG_80)) {
taglen = 80;
} else if (ast_test_flag(srtp, AST_SRTP_CRYPTO_TAG_32)) {
taglen = 32;
}
if (srtp->crypto && (ast_sdp_crypto_build_offer(srtp->crypto, taglen) >= 0)) {
return srtp->crypto->a_crypto;
}
ast_log(LOG_WARNING, "No SRTP key management enabled\n");
return NULL;
}
char *ast_sdp_get_rtp_profile(unsigned int sdes_active, struct ast_rtp_instance *instance, unsigned int using_avpf,
unsigned int force_avp)
{
struct ast_rtp_engine_dtls *dtls;
if ((dtls = ast_rtp_instance_get_dtls(instance)) && dtls->active(instance)) {
if (force_avp) {
return using_avpf ? "RTP/SAVPF" : "RTP/SAVP";
} else {
return using_avpf ? "UDP/TLS/RTP/SAVPF" : "UDP/TLS/RTP/SAVP";
}
} else {
if (using_avpf) {
return sdes_active ? "RTP/SAVPF" : "RTP/AVPF";
} else {
return sdes_active ? "RTP/SAVP" : "RTP/AVP";
}
}
}