Message ID | 20230407145252.32955-2-cconte@redhat.com |
---|---|
State | New |
Headers | show |
Series | [1/2] Add CI configuration for Kubernetes | expand |
On 07/04/2023 16.52, Camilla Conte wrote: > When running on the Kubernetes runner, this CI job is timing out. > Raise the limit to give the job enough time to run. > > Signed-off-by: Camilla Conte <cconte@redhat.com> > --- > .gitlab-ci.d/buildtest.yml | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/.gitlab-ci.d/buildtest.yml b/.gitlab-ci.d/buildtest.yml > index ba6f551752..893f8c69b4 100644 > --- a/.gitlab-ci.d/buildtest.yml > +++ b/.gitlab-ci.d/buildtest.yml > @@ -98,6 +98,7 @@ crash-test-debian: > needs: > - job: build-system-debian > artifacts: true > + timeout: 90m > variables: > IMAGE: debian-amd64 > script: Eldon recently discovered that the crash-test-debian jobs takes longer on his runner since it has KVM enabled and the job tests the devices twice - once with TCG and once with KVM. I assume this is the case for the Kubernetes runner, too? ... we should likely add a switch to the device-crash-test script to provide a way to enforce running with one accelerator only, so we can keep the run time in a sane limit. Anyway, for the time being, this patch is of course a fair work-around, thus: Reviewed-by: Thomas Huth <thuth@redhat.com>
On Tue, Apr 11, 2023 at 10:22:51AM +0200, Thomas Huth wrote: > On 07/04/2023 16.52, Camilla Conte wrote: > > When running on the Kubernetes runner, this CI job is timing out. > > Raise the limit to give the job enough time to run. > > > > Signed-off-by: Camilla Conte <cconte@redhat.com> > > --- > > .gitlab-ci.d/buildtest.yml | 1 + > > 1 file changed, 1 insertion(+) > > > > diff --git a/.gitlab-ci.d/buildtest.yml b/.gitlab-ci.d/buildtest.yml > > index ba6f551752..893f8c69b4 100644 > > --- a/.gitlab-ci.d/buildtest.yml > > +++ b/.gitlab-ci.d/buildtest.yml > > @@ -98,6 +98,7 @@ crash-test-debian: > > needs: > > - job: build-system-debian > > artifacts: true > > + timeout: 90m > > variables: > > IMAGE: debian-amd64 > > script: > > Eldon recently discovered that the crash-test-debian jobs takes longer on > his runner since it has KVM enabled and the job tests the devices twice - > once with TCG and once with KVM. I assume this is the case for the > Kubernetes runner, too? ... we should likely add a switch to the > device-crash-test script to provide a way to enforce running with one > accelerator only, so we can keep the run time in a sane limit. > > Anyway, for the time being, this patch is of course a fair work-around, thus: > > Reviewed-by: Thomas Huth <thuth@redhat.com> You added --tcg-only to the test a little while after your review here, so hopefully this patch can be dropped. With regards, Daniel
diff --git a/.gitlab-ci.d/buildtest.yml b/.gitlab-ci.d/buildtest.yml index ba6f551752..893f8c69b4 100644 --- a/.gitlab-ci.d/buildtest.yml +++ b/.gitlab-ci.d/buildtest.yml @@ -98,6 +98,7 @@ crash-test-debian: needs: - job: build-system-debian artifacts: true + timeout: 90m variables: IMAGE: debian-amd64 script:
When running on the Kubernetes runner, this CI job is timing out. Raise the limit to give the job enough time to run. Signed-off-by: Camilla Conte <cconte@redhat.com> --- .gitlab-ci.d/buildtest.yml | 1 + 1 file changed, 1 insertion(+)