To accurately reflect the state of a build request we also need a
cancelling state. This is set when we've started a build and then for
whatever reason cancel it, cancelling is not instantaneous so we have
this state to indicate that a cancel is in progress.
Also add a state transition guard. As the state of a BuildRequest can
currently be modified by three processes; Toastergui,
Runbuilds/bldcontrol and the buildinofhelper we cannot say for sure
which process will be running at the time of cancellation so in order to
avoid one of these processes making an incorrect transition only allow
transitions of state to increase.
e.g. CREATED -> QUEUED -> INPROGRESS
And to ignore such requested changes such as
INPROGRESS -> CREATED
(Bitbake rev: 449598c8e6be75bd0c9d59e7bdf859d1d6f83858)
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
A new state CANCELLED is introduced to, distinguish
the state of build.
[YOCTO #6787]
(Bitbake rev: 404f406fecae879703bcfe96f3b65086b115fa8a)
Signed-off-by: Sujith H <sujith.h@gmail.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Remove getBBController function from BuildEnvironmentController.
The constructor of BitbakeController is updated appropriately so that
call can be made to connect to running server.
The call to startBBServer is removed from bbcontroller and handledin
localhostbecontroller.
[YOCTO #6787]
(Bitbake rev: e15151106aae21d3b164ca868be42bd63905f0a1)
Signed-off-by: Sujith H <sujith.h@gmail.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Update the implementation of startABuild and cancelAbuild to reflect
changes to the backend api. We now have a dedicated endpoint to make
calls into so add this url to libtoaster.ctx and allow passing null in
as a url value to indicate that we want to use the current project
Also:
- Fix some documentation comments
- Add the convenience of passing in an array of targets to startABuild
(Bitbake rev: 61a21d96abab113cbd13376cdb8b08a426b50538)
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Update the implementation of the backend api for cancelling builds with
the new cancelling BuildRequest state and cancelled Build state.
Also added some docstring about general usage.
Co-Author: Sujith H <sujith.h@gmail.com>
(Bitbake rev: 0d76084f5d896e4199e1446e2d6d43190a4fcc3a)
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Move the backend xhr implementation of the build request changes
into it's own file and out of the ToasterTable definition.
It used to live in the views.py but in a hope of starting to collate logical
groups of views move this to a new file called api.
(Bitbake rev: 29572f0e6bd3b5e8315f3b93d55bdb8967b86bc3)
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add forceShutDown function to BitbakeController class. This function
provides a mechanism to cancel the build from toaster. An API which
can be used safely to cancel build and hence shutdown running bitbake
server.
[YOCTO #6787]
(Bitbake rev: 89dc2ee8f5b7255538082ce4f6cb5277839875a8)
Signed-off-by: Sujith H <sujith.h@gmail.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
toasterui exits event loop on one of the following events:
CommandCompleted, CommandFailed or CommandExit.
Unfortunately none of them come from bitbake when build fails.
This is normai if toasterui runs in observer mode. However, if it's
in build mode this causes toasterui to stuck in the infinite loop
waiting for new events.
The only event we can rely on is BuildCompleted as it always
comes from bitbake unlike 3 above mentioned events.
Modified the code to always shutdown toasterui in build mode
on BuildCompleted event.
(Bitbake rev: 9cd60f98b13cf7b1c518851a51e1cbaa596d8f81)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Toaster can't run builds on Ubuntu as default /bin/sh points
to dash there. The reason is that oe-init-build-env can't
be sourced under dash for various reasons. It can be fixed
or work arounded, but it wouldn't fix builds for older releases.
Explicitly using bash to start builds should fix the issue.
(Bitbake rev: e4352603b21d43a8b5d9ee417c036fff0a263d40)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The keys 'started', 'ended', 'cpu_time_user', 'disk_io_read' and
'disk_io_write' were added to the event recently, so they don't
exist in the events generated by bitbake server from older releases.
Checking if task_to_update structure has these keys before using
them should fix build of older releases.
(Bitbake rev: 79611d0ea742263074fbb0bf5f1e39df75fd9f55)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Added 'INHERIT+="toaster buildhistory"' line to the conf/local conf
when Toaster starts. It should make commandline builds to provide
all required information to Toaster backend.
(Bitbake rev: 1271cf430087c66f87c46689b37b8a3538c35739)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bitbake server is used only during the build. There is no need
in keeping server running after the build.
Running bitbake -m in the subshell after the build should stop
the server.
(Bitbake rev: 062c68e3e7d00834118fe07ceb0899874be714ae)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Added 'nowait' parameter to _shellcmd method to support
running chain of commands in a subshell.
This is going to be used to stop bitbake server after
the build.
(Bitbake rev: 89fae3c8712bfaac48686c58b480e40c1abdcfdc)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Reimplemented triggerBuild method to support one build directory
per project:
- start bitbake server from the cloned repository
- don't run observer
- run bitbake build directly instead of triggering it
through xmlrpc
[YOCTO #7880]
[YOCTO #9058]
[YOCTO #8958]
(Bitbake rev: aba8e19bf609196f749e185d43571f706707e408)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Removed updating of bblayers.conf. It will be done in runBuild method.
Changed return value: return list of layers.
Removed _updateBBLayers method.
(Bitbake rev: 198bf7e6b8d7f847f2619b71c6bd86a9a76156c9)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Called triggerBuild with brbe parameter instead of adding TOASTER_BRBE
variable to the database and fetching it in triggerBuild.
(Bitbake rev: 669bb3f6865d8bb89cd94ef9d0eba3346c47d7b2)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This API is quite dangerous as it removes build directory.
It's not used anywhere and most probably will not be used in future
as toaster is going to have one build directory per project.
(Bitbake rev: ef9e126227d3bdee86c1592e2baa43d21e6a32df)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We still will have to run bitbake server, but it will be done
different way and the code will be in triggerBuild function.
Removed startBBServer API from BuildEnvironmentController and
LocalhostBEController classes.
(Bitbake rev: a8f986d73a84f28fa8b116b91fb30c5d2cd4edbe)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
buildinfohelper.brbe is lost when buildinfohelper is closed.
This causes incorrect report of brbe when build is done.
Saved brbe attribute before closing buildinfohelper and used
it to report correct brbe.
Got rid of useless and confusing 'ToasterUI build done 1'
log message.
(Bitbake rev: 5d7cce0d0ed70f6b3ebd6cbad300d86964a13398)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
DepTreeGenerated event doesn't contain 'providermap' data in jethro.
Modified buildinfohelper to handle events without this data. This
should make it possible to handle jethro events coming from jethro
bitbake server by the latest buildinfohelper.
(Bitbake rev: f6dcb1c9967f042beae024146781cb8235a9e1f2)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Return value of self.BBServer.registerEventHandler differs between
jethro and master. To be able to build jethro toaster should be
able to communicate with jethro bitbake server i.e. it must work
with both old and new registerEventHandler call.
(Bitbake rev: f356c154016c428a3b53af61a075de6f14d9d1d9)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
In current toaster code BRBE(build request:build environment) value
is passed from toaster to buildinfohelper through the 'SetBRBE' event.
Passing it through environment variable is easier as it doesn't
involve rpc communication between toaster and bitbake server.
It also eliminates the need in running bitbake observer process.
Added parameter 'brbe' to BuildInfoHelper.__init__
Used environment variable TOASTER_BRBE to set brbe for
buildinfohelper object.
(Bitbake rev: a0c8e2b309055e5927a8ff729d292ccaa69d0575)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Set BITBAKE_UI variable to 'toastergui' for command line builds
to use toasterui as a default ui module for bitbake.
(Bitbake rev: 9ad6393d30cb6196cf7c9a5adcf33febd724d294)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
noui command line option doesn't makes sense anymore as toaster doesn't
run bitbake. It should be safe to to remove it.
The purpose of this option was to skip running bitbake observer process.
This was never used before as it's not possible to run toaster build
without running observer.
(Bitbake rev: 7506719090e8bb39231cf389c4a5b47f1b37a01f)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
>From now on toaster script will not run bitbake server.
It will be started by runbuilds and stopped after the build.
(Bitbake rev: 3fbd8534149e87c5a5d1bc1691711cfca05cafd1)
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
It was used for workaround git 1.7.9.2 which was released in 2012 which
should not be existed on nowadays host, so remove it to avoid
confusions.
(Bitbake rev: 6140d0cc9aecf1029ca16fed47071dfcc92f4269)
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Python 2's sqlite3 module defaults to returning Unicode strings for SQL
text queries, which could trickle down to other parts of bitbake code and
cause unexpected Unicode conversions. Using byte strings avoids this issue.
For example, the git fetcher's AUTOREV support caches HEAD SHA1's using
bb.persist_data, so sometimes the git command strings passed to fetch2's
runfetchcmd() were unicode, potentially causing UnicodeDecodeErrors when
it appended the values of environment variables containing non-ASCII chars.
[YOCTO #9382]
(Bitbake rev: 09623a0811c613a47a01ae465b822d8156faca30)
Signed-off-by: Daniel Klauer <daniel.klauer@gin.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
sstate.bbclass for example writes siginfo files to a separate location
but we need to read taint data from the standard path.
(Bitbake rev: da444c9761ee15a59ea8880e3f812a5d3f1a1aaa)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The taint values need to be passed from the server to the workers to
ensure they see the same stamp values. Also ensure that the "nostamp:"
prefix isn't included in the checksum value to match the server
calculation. This ensures the checksums are all consistent.
(Bitbake rev: f80ba20e90f3746f7faee3e0ff7f249025fec8ee)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
In theory all the information to recalcuate the task signatures was written
into the siginfo/sigdata files. In reality, some of the information was
written into the filename.
Firstly this patch duplicates that info into the file itself just for easy
of use since its small.
Secondly, we abstract out the existing "calculate the checksum" code for
the taskhash, and add a function to calculate the bashhash based on the
informaiton within the file.
Finally, we call these functions when we're writing out the data to check
that the data we're writing is consistent. I've found a couple of places
it wasn't and its good to know about these in advance, rather than having
a siginfo/sigdata file which a given hash in its filename but a contents
which give a different result.
This should all combine to avoid a certain class of checksum bugs making
it into world, and identifying problems in advance.
(Bitbake rev: 0f50a18d7a0ea0d68edd8e5217e29111f4b1ea0b)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When I enabled debugging of the checksum code, I found the value calculated
from siginfo/sigdata files for do_fetch tasks never matched. This was due
to an error in the way the data was being stored for these, it wasn't ordered
correctly. This patch fixes things so the checksums calculated from
siginfo/sigdata files is correct when file checksums are present.
(Bitbake rev: 046c1be7594fae2eec3d1f242ba3e9a85f1a1880)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The real method is a few lines later, this one is incorrect and
just causing confusion. Remove it.
(Bitbake rev: a896f263300f069400eae533be0daf5dedf41c95)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The tests were skipped when running without network even though they
didn't require network. This commit also adds a test case for URLs with
ports in them (the ports should not be considered when doing trusted
network checks).
(Bitbake rev: 77747de6b20538063eef3b188489a35bef225359)
Signed-off-by: Olof Johansson <olof.johansson@axis.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bitbake would fail to classify the following URL as belonging to a
allowed network, because of the port number in the url.
BB_ALLOWED_NETWORKS = "*.example.com"
SRC_URI = "http://git.example.com:8080/foo.tar.gz"
Since protocols aren't specified in the BB_ALLOWED_NETWORKS variable,
it's reasonable to believe that this should work regardless of protocol
being used.
(Bitbake rev: ff603df23037e10fb2cfdf150429cba3f65072cd)
Signed-off-by: Olof Johansson <olof.johansson@axis.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We were doing a string search for the literal string "require" and
assuming that a match meant the recipe uses the require keyword to
include the contents of another recipe. This test isn't very smart (it
should at least have matched on "require ") and triggers on the word
require anywhere in the recipe, i.e. the word "required" in the
DESCRIPTION of core-image-lsb.bb, and then breaks as the much smarter
regex fails to match anything.
Instead always run the regex search and only try to update the
require entry when the regex matches.
(Bitbake rev: a6add0e95d3d1e9a6a9fcabd73543bc5c278915f)
Signed-off-by: Joshua Lock <joshua.g.lock@intel.com>
Signed-off-by: Elliot Smith <elliot.smith@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The help text said to append a semicolon and a task name to a target to
run a task other than do_build. What you need to append is a colon.
[YOCTO #9326]
(Bitbake rev: 691419e23a94a94129b177e71d2c728b12689139)
Signed-off-by: Belen Barros Pena <belen.barros.pena@linux.intel.com>
Signed-off-by: Elliot Smith <elliot.smith@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add additional metadata to the layer created for build history to be
able to identify the layer and recipe later on. Specifically this is the
branch and release to which the recipe and layer are associated with
enabling differentiation of two recipes which are local release and
master and 'master' release.
[YOCTO #8528]
[YOCTO #8545]
(Bitbake rev: 3deebd887bddbbd02fd9829a180aab494b1af7c4)
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Elliot Smith <elliot.smith@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(Bitbake rev: 20ce1e1b39a9b602eb51ca0ba3954ea3e999c874)
Signed-off-by: Elliot Smith <elliot.smith@intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(Bitbake rev: 0e5f45d68e423f8462937879eed3253db31b2bb5)
Signed-off-by: Elliot Smith <elliot.smith@intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(Bitbake rev: 31204937f71a7e0aa08361c3e20d02d063788a86)
Signed-off-by: Elliot Smith <elliot.smith@intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(Bitbake rev: 4fda6be831d10e6d266b11975a0e9a35a7f35a77)
Signed-off-by: Elliot Smith <elliot.smith@intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Create a new folder for Selenium tests.
Add a new base Selenium testcase class and a helper which
instantiates a webdriver for a given browser.
Add a sample Selenium test case which can be used as a template
for creating new tests.
(Bitbake rev: b7a377aa2ab36390d619e2a0436ccb4b8d186c23)
Signed-off-by: Elliot Smith <elliot.smith@intel.com>
Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
There appears to have been a lot of copy and pasting of the code
which prints tracebacks upon failure and limits the stack trace to
5 entries. This obscures the real error and is very confusing to the user
it look me an age to work out why some tracebacks weren't useful.
This patch removes the limit, making tracebacks much more useful for
debugging.
[YOCTO #9230]
(Bitbake rev: 5549748a200b5df259fc7352477ec59471b87b2f)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
It's very possible that we added layer as:
BBLAYERS += "/path/to/meta/"
then there would be warning:
WARNING: No bb files matched BBFILE_PATTERN_core '^/path/to/meta//'
This patch can fix the problem.
(Bitbake rev: 2b1cb21d18fb18399e682021b866babeced9a4aa)
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
With the code as it stands today it not possible to execute a python function
and get "normal" python exception handling behaviour. If a python function
raises an exception, it forces a traceback to be printed and the exception
becomes a FuncFailed exception.
This adds in a parameter 'pythonexception' which allows standard python
exceptions to be passed unchanged with no traceback. Ultimately we may want
to change to this convention in various places but at least it means we can
start to add sane functions now.
(Bitbake rev: 85cf22fd0ed26bb7dc7738ef2a10441891f38ae2)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
There are some codepaths where the file checksum is verified and can
be found to mismatch but the 'rename' logic doesn't kick in. If code
relies on the presence of a file for the checksum having been checked
(e.g. uninative.bbclass) then it can be used when the checksum hasn't
matched.
Therefore rename the file whenever an invalid checksum is encountered.
(Bitbake rev: 69ef6c8a9db02bfa0e3fac72481ec26586a29a01)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Prevent a hang when shutdown() is called during parsing (e.g. after
SIGINT). We must not append 'None' to the jobs queue. Otherwise the
worker loop inside Parser.realrun() may break out at the wrong point,
causing the results queue thread blocking bitbake indefinitely.
[YOCTO #9319]
(Bitbake rev: 7ebea3e9a60232222efa8a546a0ff28a53029949)
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Currently bbappend files in a layer are applied in the order they're
found on disk (as reported by glob) which means things are not
deterministic.
By sorting the glob results, the order becomes deterministic, the parsing
order for .bb files also should be deterministic as a result of this change.
[YOCTO #9138]
(Bitbake rev: 3f8febc4212fbd3485ac9bdd4ac71b8fb0a05693)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Before this patch, directory symlinks mathcing filename pattern (either
a file name or a glob pattern) were followed. However, directory
symlinks deeper in the search chain were omitted by os.walk(). Now
directory traversal behaves consistently, ignoring syminks on all
levels.
One reason for choosing not to "walk into" directory symlinks is that
dir symlinks in externalsrc.bbclass in oe-core are causing problems in
source tree checksumming.
[YOCTO #8853]
(Bitbake rev: 66dff37ebcd1dd14ebd6933d727df9cf0a641866)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Michael Wood
Sujith H
Ed Bartosh
Robert Yang
Klauer, Daniel
Richard Purdie
Olof Johansson
Joshua Lock
Belen Barros Pena
Elliot Smith
Markus Lehtonen