[PATCH][U] UBUNTU: SAUCE: selftests: seccomp: bump up timeout to 5min

[Andrea Righi]

On Wed, Jun 16, 2021 at 07:31:19AM -0300, Thadeu Lima de Souza Cascardo wrote:
> On Wed, Jun 16, 2021 at 11:11:33AM +0200, Andrea Righi wrote:
> >  DEBUG| [stdout] # selftests: seccomp: seccomp_benchmark
> >  DEBUG| [stdout] # net.core.bpf_jit_enable = 1
> >  DEBUG| [stdout] # net.core.bpf_jit_harden = 0
> >  DEBUG| [stdout] #
> >  DEBUG| [stdout] not ok 2 selftests: seccomp: seccomp_benchmark # TIMEOUT 120 seconds
> > 
> > This test can easily fail if the testing environment is a bit
> > overloaded, so bump up the timeout to 5min to prevent false positive
> > failures.
> > 
> > Signed-off-by: Andrea Righi <andrea.righi at canonical.com>
> 
> There are two tests running here, seccomp_bpf and seccomp_benchmark.
> seccomp_benchmark tries to time its own run. Have you managed to identify if it
> is seccomp_benchmark that is running late, and why (and by how much), or if it
> is seccomp_bpf that is taking too long on these instances/workloads?

seccomp_benchmark is the one that requires more time, because it's
testing a lot of syscalls:

 # time ./seccomp_benchmark
 Current BPF sysctl settings:
 net.core.bpf_jit_enable = 1
 net.core.bpf_jit_harden = 0
 Calibrating sample size for 15 seconds worth of syscalls ...
 Benchmarking 29470215 syscalls...
 ...
 real	1m15.558s
 user	0m33.147s
 sys	0m42.405s

seccomp_bpf instead is quite fast, usually it takes between 2-3 sec in
my test VM, for example:

 # time ./seccomp_bpf
 ...
 real	0m2.819s
 user	0m0.039s
 sys	0m0.045s

> 
> And do we really run it concurrently with any other workload to justify it
> being late? Or is it the fact that it's running on a virtualized environment
> that is overcommitted?

I think the reason is that tests are running in a virtualized
environment and it's not so rare to have overcommitted/overloaded
conditions.

In my local VM, where the environment is not overcommitted, the elapsed
time to run both tests is very consistent, always around 75 sec.

-Andrea

> 
> Just curious.
> 
> Cascardo.
> 
> > ---
> >  tools/testing/selftests/seccomp/settings | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/tools/testing/selftests/seccomp/settings b/tools/testing/selftests/seccomp/settings
> > index 6091b45d226b..694d70710ff0 100644
> > --- a/tools/testing/selftests/seccomp/settings
> > +++ b/tools/testing/selftests/seccomp/settings
> > @@ -1 +1 @@
> > -timeout=120
> > +timeout=300
> > -- 
> > 2.31.1
> > 
> > 
> > -- 
> > kernel-team mailing list
> > kernel-team at lists.ubuntu.com
> > https://lists.ubuntu.com/mailman/listinfo/kernel-team