On 24/08/2023 9:52 am, Anthony PERARD wrote:
> On Wed, Aug 23, 2023 at 07:05:56PM +0100, Andrew Cooper wrote:
>> On 23/08/2023 4:23 pm, Anthony PERARD wrote:
>>> On failure of "build"-each-commit script, the next command that move
>>> the log back into the build directory isn't executed. Fix that by
>>> using "after_script" which is always executed even if the main
>>> "script" fails. (We would still miss the log when the jobs times out.)
>>>
>>> Signed-off-by: Anthony PERARD <anthony.per...@citrix.com>
>>> ---
>>> automation/gitlab-ci/test.yaml | 1 +
>>> 1 file changed, 1 insertion(+)
>>>
>>> diff --git a/automation/gitlab-ci/test.yaml b/automation/gitlab-ci/test.yaml
>>> index 810631bc46..5099f2e6b6 100644
>>> --- a/automation/gitlab-ci/test.yaml
>>> +++ b/automation/gitlab-ci/test.yaml
>>> @@ -140,6 +140,7 @@ build-each-commit-gcc:
>>> CC: gcc
>>> script:
>>> - BASE=${BASE_SHA:-${CI_COMMIT_BEFORE_SHA}}
>>> TIP=${TIP_SHA:-${CI_COMMIT_SHA}}
>>> ./automation/gitlab-ci/build-each-commit.sh 2>&1 | tee
>>> ../build-each-commit-gcc.log
>>> + after_script:
>>> - mv ../build-each-commit-gcc.log .
>>> artifacts:
>>> paths:
>> Thanks for looking into this, and yeah that is dumb, but why play games
>> with the parent directory?
> `git clean -ffdx` has the tendency to remove everything that's not
> committed, that's why. But maybe we can teach ./build-each-commit.sh to
> ignore that logfile.
Oh, right. Yeah, lets not lose the log file like that.
I'd say that teaching `git clean` to leave the file interacted and not
copying it is going to be a more robust option.
~Andrew
