The initial idea to link an error to another one was a quick solution
to group them if they where related, but this became challenging
to copute metada regarding errors.
- The displayed error message was not always consistent with the real
root cause/the error that lead here.
- The aggregates (lets says, linked buils ids) could be the one of the
error, or from all error messages. Same for the versions, first seen, ..
This is confusing to knwo what is the leist we are managing and what is
the expecte result to display
Main motivation:
on a standard error page (will be changed to "assignment"), we want to
have the list of error message that is related to this one. We want to
know for each message (a real build error) what is the version,
first seen, ...
This will give more flexibility on the display,
The assigned person/team/test-tags, ... are moved to this model
The appearance data remains on the build error but are aggregate on the
assignation.
When a type error occurs when trying to format a message in a build log,
the suspicious are are joined with the message. But as the args may be a
tuple, an errors occurs when concatenating the message with the args
during the join.
With this commit, we ensure that the args are casted into a list.
When an image is build for the first time and the build fails, an
ImageNotfound Exception is raised when trying to push the image.
Whith this commit, a warning is emmited instead of crashing in such a
case.
A recent task introduced a record loop using self inside.
It doesn't respect an api multi classic loop but it isn't an issue
since onchange are always called on a single record.
Removing the loop on all onchange to clean it up.
Idea is to help runbot/QA team to monitor other peoples tasks
easier. This way we will have a responsible person for each task
that someone else has assigned.
The goal of pr is two-fold:
1) We need to be able to better monitor the progress of tasks that
are not our own.
2) It introduces another metric with which we can measure individual
progress, and this can become main measuring metric for future non-technical
employees whose main goal will be making sure that none of the tasks become
rotten(not have any visible progress for months).
When a lot of Docker images are updated at the same time, all runbot
hosts will try to pull them at the same moment.
With this commit, only the images marked as `always_pull` will be pulled
and if one of them takes too much time to be pulled, we let the host
make another loop turn before continuing the images pull.
Finally, if a host uses the Docker registry, it will pull the remote
image when running a step, that way the image will be pulled
automatically if needed.
The docker operation are called often and cannot be logged each time.
If a docker operation is slow, we log it at the end but waiting for this
output we have no idea what the process is doing.
This pr proposed to go a lower level and get the stream from the docker
operation to be able to log earlier if we started a potentially slow
operation.
When disabling tests on runbot by using the test_tags on a build error,
the tests are disabled on every tested version. This can be annoying
when a test only fails from one version up to another.
With this commit, a min and max version can be specified on a
build_error when the test_tags field is used.
If the min and max are not used, the test will be disabled cross
versions.
A common error on runbot is to generate link containing a __init__.py
file
[/some/path/to/__init__.py](/some/path/to/__init__.py)
This would be rendered as
<a href="/some/path/to/<ins>init<ins>.py">/some/path/to/<ins>init<ins>.py</a>
Breaking the link, and the display of the name
By default markdown will not render links avoiding this issue, but it
will remain for the content of the a, needing to manage some kind of
escaping.
The way to escape markdown is to add a \ before any special character
This must be done upront before formating, adding the method
markdown_escape
Our implementation of markdown is not meant to meet the exact
specification of markdown but better suit our needs. One of the
requirements is to be able to use it to format message easily but adding
dynamic countent comming from the outside. One of the error than can
occur is also
'Some code `%s`' % code can also cause problem if code contains `
This issue could be solved using indented code block, but this would
need to complexify the generated string, have a dedicated method to
escape the code blocs, ...
Since we have the controll on the input, we can easily sanitize all
ynamic content to avoid such issues. For code block we introduce a way
to escape backtick (\`). It is non standard but will be easier to use.
Combine with that, the build._log method now allows to add args with the
values to format the string (similar to logging) but will escape
params by default. (cr.execute spirit)
name = '__init__.py'
url = 'path/to/__init__.py'
code = '# comment `for` something'
build._log('f', 'Some message [%s](%s) \n `%s`', name, url, code)
name, url and code will be escaped
The stats are only kept when the build finishes, this makes sence to
avoid collecting stats of a manually killed build (or because of newer
build found), but the stats mays be interresting when the build timeouts.
Adding a manual collection of stats in this case.
The underscore was previously used in database name, but this is not a
valid character for certificates (still need some investigation about
the limitations)
If it worked to access it it was causing issues for IOT andother.
_ is forbidden on saas actually.
The design_theme database was changed to design-theme, but this is not
accessible through the nginx condig. This should solve the issue.
The current runbot infrastructure has 100+ machine that will each build
all docker images.
This is unpractical for multiple reasons:
- impotant load on all machine when the build could be done once
- possible differences for the same image depending on the moment the
base was pulled on the host. An extreme example was the version of
python (3.11 or 3.12) when building the noble docker image before it was
stabilized.
- increase the chance to have a docker build failure in case of network
problem. (random)
A centralized registry will help with that, and allow future devlopment
to generate more docker images. All docker images will be exactly the
same, the pull time is faster than build time, only one build per docker
image, ...
When building a docker image the error is part of the stream, and
at the end.
The current behaviour will append the error message at the begining of
the "result log" breaking the temporality of the output. Adding it at
the end should be more intuitive to read. This will also help to get a
more usefull error sumary in some cases.
The migration to the new dockerfile also includes a change to the noble
docker image
This is done in a separate commit to ease the comparaison.
Notes:
The postgresql-client was initially pinned to version 16 and removed
from debian control, but this version proposes add the repositories and
let apt chose the latest version. This is experimental but was tested
on ubuntu:jammy, with 16 version installed as expected.
The requirements are installed as user instead of root to avoid the need
to use the ignore_installed flag (and solve a bunch of issues related
to this flag)
When a docker fails to build, the output is logged in the chatter
leading to a lot of noise and a not so readable output. Moreover, the
output tries to format markdown and don't render line break correctly.
This commit proposes to introduce a model to store this output, as well
as some other info like the image identifier, build time, ...
This will help to compare images versions between hosts and should be
useful later to have multiple version of the same image with variant
once the docker registry is introduced.
The batch prepare fill missing won't work if the repo is single version,
in a foreign bundle.
This commit simply adds the branches depending on their version instead
of the bundle version.
When cleaning build errors before fingerprinting, it's only possible to
replace the matching regex with something else but not an empty string.
Since the python 3.11 that may adds lines in error message in order to
visually improve them, the fingerprint of those errors does not match
anymore between different versions.
With this commit, when the replacement string is two consecutive simple
quotes, the matching element is replaced by an empty sting, allowing to
remove unwanted characters.
The order of requirements may have an impact on final outcome.
In documentation builds, we have two requirements with conflicting needs
Lets make the doc requirements install after the odoo ones
(sequence dependant)
The runbot colors can lack contrast in some case, especially for
colorblinded people.
This commit introduces a theme that should help a little.
Note that this was done without much analysis and should be tweaked in
the future.
There is an issue in unidiff 0.7.3 fixed in 0.7.4
a3faffc54e
This version is not able to parse a diff with removed files.
Since the unidiff packaged version in noble is 0.7.3 patching it looks
like the easiest solution
The MAKE_CELL opcode appeared in python 3.11 and is needed in some
python steps when using closures and generators.
Like:
`(all(s > e for e in [1,2]) for s in [0,1])`
