On Tue, 16 Jul 2024 at 18:45, Thomas Huth <th...@redhat.com> wrote: > > On 16/07/2024 15.06, Michael S. Tsirkin wrote: > > On Tue, Jul 16, 2024 at 02:59:30PM +0200, Igor Mammedov wrote: > >> CI often fails 'cross-i686-tci' job due to runner slowness > >> Log shows that test almost complete, with a few remaining > >> when bios-tables-test timeout hits: > >> > >> 19/270 qemu:qtest+qtest-aarch64 / qtest-aarch64/bios-tables-test > >> TIMEOUT 610.02s killed by signal 15 SIGTERM > >> ... > >> stderr: > >> TAP parsing error: Too few tests run (expected 8, got 7) > >> > >> At the same time overall job running time is only ~30 out of 1hr allowed. > >> > >> Increase bios-tables-test instance timeout on 5min as a fix > >> for slow CI runners. > >> > >> Signed-off-by: Igor Mammedov <imamm...@redhat.com> > > > > We can't just keep increasing the timeout. > > The issue is checking wall time on a busy host, > > isn't it? Let's check CPU time instead. > > The timeout setting comes from meson, not sure whether you can switch that > easily to use CPU time instead of wall time? > > Anyway, if the bios-tables-test is getting more and more complex, it's maybe > not such a good idea to run it in a job that is using TCI ... Maybe it's > best to remove aarch64-softmmu from the cross-i686-tci job?
It's one of the few tests that actually runs code in the guest: we definitely shouldn't reduce the coverage of the actual TCI part of the TCI job, I think. I continue to think we need to find out why this CI job is perpetually flaky and fix the underlying cause, not simply increase timeouts or drop test cases or configs from it. -- PMM
