Fix flake on failed podman-remote build

[rhatdan]

We have a race condition where podman build can fail
but still return an exit code of 0. This PR ensures
that as soon as the build fails, the failed flag is set
eliminating the race.

Fixes: #10029

[NO TESTS NEEDED] Tests of failed builds are already in place, and
the elimination of the race should be enough.

Signed-off-by: Daniel J Walsh [email protected]

[openshift-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: rhatdan

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [rhatdan]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[rhatdan]

@edsantiago I ran your tests for a few minutes and could not get this PR to fail.

[edsantiago]

@rhatdan I'm sorry... no-go. I'm still seeing it fail, anywhere from 7 seconds to 208 seconds. New reproducer shows how long it takes:

$ t0=$SECONDS;while :;do echo $(($SECONDS - $t0));../bin/podman-remote build -t build_test --pull-never . && break;done

[edsantiago]

@Luap99 we got a compose flake! Rootless, and curl shows one ECONNRESET followed by consistent ECONNREFUSED. To me that implies that the container has died. The timestamps are consistent with that. I don't think any amount of retrying is going to help here; we need a different way to debug.

(Sorry to hijack the thread, Dan)

[mheon]

This LGTM, even if it's not a complete fix for the problem - the change on its own does seem to make things better.

[edsantiago]

/lgtm
/hold
Le mieux est l'ennemi du bien

[jwhonce]

/hold cancel