Skip to content

Conversation

vszakats
Copy link
Member

@vszakats vszakats commented Sep 27, 2024

Jobs consistently fail to connect to the test server (run in GHA) since
2024-Aug-29:
https://ci.appveyor.com/project/libssh2org/libssh2/builds/50498393

There was an earlier phase of failures one month before that, that got
fixed by increasing the wait for the server in
bf3af90.

Thus, skip running tests in AppVeyor CI jobs, except: After some
experiments, it seems that running tests with the last OpenSSL job and
the last WinCrypt job work, which still leaves some coverage.
It remains to be seen how stable this is.

This is meant as a temporary fix till there is a solution to make all
jobs run tests reliable like up until a few months ago.


Bumping up the timeout to 10 minutes doesn't help. But it turns out
by accident, that running the tests only in the last AppVeyor job does
pick up the server and run correctly. So it's not serving as a "canary"
or a signal after all.

@vszakats vszakats changed the title CI/appveyor: stop running tests ci/appveyor: stop running tests Sep 28, 2024
@mback2k
Copy link
Member

mback2k commented Sep 28, 2024

Are the docker bridge containers CI jobs actually launched on GitHub in time?

@vszakats
Copy link
Member Author

It seemed to me they do launch, then remain idle. But my idea of "in time" is fuzzy and may be misinterpreting the visible logs.

@vszakats vszakats changed the title ci/appveyor: stop running tests ci/appveyor: stop running tests (workaround for infrastructure permafail) Oct 2, 2024
Jobs consistently fail to connect to the test server (run in GHA) since
2024-Aug-29:
https://ci.appveyor.com/project/libssh2org/libssh2/builds/50498393

There was an earlier phase of failures one month before that, that got
fixed by increasing the wait for the server in
bf3af90.

Let's make that wait 10 minues (was 5), in case it helps, though I tried
it earlier to no avail. Such long wait doesn't seem acceptable if it
works, and a sign of issues lurking somewhere.

Thus, skip running tests in AppVeyor CI. This leaves the Windows
platform with no tests at all.

Except in the last job to serve as a canary in case the tests would
"fix themselves", and to leave a signal that we have problem.
@vszakats vszakats force-pushed the appveyor-skip-tests branch from 2f6f5cd to f97a7da Compare October 2, 2024 15:05
@vszakats vszakats changed the title ci/appveyor: stop running tests (workaround for infrastructure permafail) ci/appveyor: reduce running tests (workaround for infrastructure permafails) Oct 2, 2024
@vszakats vszakats added the tests label Oct 2, 2024
@vszakats vszakats changed the title ci/appveyor: reduce running tests (workaround for infrastructure permafails) ci/appveyor: reduce test runs (workaround for infrastructure permafails) Oct 2, 2024
@vszakats
Copy link
Member Author

vszakats commented Oct 2, 2024

After some trial and error it seems that if the last OpenSSL and last WinCrypt job are enabled, they both finish fine. No idea what the logic is behind this, if any, or if this will stay this way or just a fluke.

@vszakats vszakats closed this in b5e68bd Oct 3, 2024
@vszakats
Copy link
Member Author

vszakats commented Oct 3, 2024

Merged. This PR is not a solution to this problem, just an attempt to keep the CI usable till a real fix is found and we can re-enable tests for these Windows CI jobs.

@vszakats vszakats deleted the appveyor-skip-tests branch October 3, 2024 17:16
vszakats added a commit to vszakats/libssh2 that referenced this pull request Oct 4, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants