diff mbox series

[03/13] tests/avocado/intel_iommu.py: increase timeout

Message ID 20240726134438.14720-4-crosa@redhat.com
State New
Headers show
Series Bump Avocado to 103.0 LTS and update tests for compatibility and new features | expand

Commit Message

Cleber Rosa July 26, 2024, 1:44 p.m. UTC
Based on many runs, the average run time for these 4 tests is around
250 seconds, with 320 seconds being the ceiling.  In any way, the
default 120 seconds timeout is inappropriate in my experience.

Let's increase the timeout so these tests get a chance to completion.

Signed-off-by: Cleber Rosa <crosa@redhat.com>
---
 tests/avocado/intel_iommu.py | 2 ++
 1 file changed, 2 insertions(+)

Comments

Daniel P. Berrangé July 29, 2024, 10:30 a.m. UTC | #1
On Fri, Jul 26, 2024 at 09:44:28AM -0400, Cleber Rosa wrote:
> Based on many runs, the average run time for these 4 tests is around
> 250 seconds, with 320 seconds being the ceiling.  In any way, the
> default 120 seconds timeout is inappropriate in my experience.
> Let's increase the timeout so these tests get a chance to completion.

A high watermark of over 5 minutes is pretty long for a test.

Looking at the test I see it runs

   self.ssh_command('dnf -y install numactl-devel')

but then never actually uses the installed package.

I expect that most of the wallclock time here is coming from having
dnf download all the repodata, 4 times over.

If the intention was to test networking, then replace this with
something that doesn't have to download 100's of MB of data, then
see what kind of running time we get before increasing any timeout. 


> Signed-off-by: Cleber Rosa <crosa@redhat.com>
> ---
>  tests/avocado/intel_iommu.py | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/tests/avocado/intel_iommu.py b/tests/avocado/intel_iommu.py
> index 008f214397..9e7965c5df 100644
> --- a/tests/avocado/intel_iommu.py
> +++ b/tests/avocado/intel_iommu.py
> @@ -25,6 +25,8 @@ class IntelIOMMU(LinuxTest):
>      :avocado: tags=flaky
>      """
>  
> +    timeout = 360
> +
>      IOMMU_ADDON = ',iommu_platform=on,disable-modern=off,disable-legacy=on'
>      kernel_path = None
>      initrd_path = None
> -- 
> 2.45.2
> 
> 

With regards,
Daniel
Cleber Rosa Aug. 1, 2024, 1:02 a.m. UTC | #2
On Mon, Jul 29, 2024 at 6:30 AM Daniel P. Berrangé <berrange@redhat.com> wrote:
>
> On Fri, Jul 26, 2024 at 09:44:28AM -0400, Cleber Rosa wrote:
> > Based on many runs, the average run time for these 4 tests is around
> > 250 seconds, with 320 seconds being the ceiling.  In any way, the
> > default 120 seconds timeout is inappropriate in my experience.
> > Let's increase the timeout so these tests get a chance to completion.
>
> A high watermark of over 5 minutes is pretty long for a test.
>

I agree.

> Looking at the test I see it runs
>
>    self.ssh_command('dnf -y install numactl-devel')
>
> but then never actually uses the installed package.
>
> I expect that most of the wallclock time here is coming from having
> dnf download all the repodata, 4 times over.
>

Exactly.

> If the intention was to test networking, then replace this with
> something that doesn't have to download 100's of MB of data, then
> see what kind of running time we get before increasing any timeout.
>
>

I was trying not to get in the way of the original test writer.

Eric,

Are you OK with replacing this command for a simpler file transfer?
Any suggestions?

Regards,
- Cleber.
Eric Auger Aug. 12, 2024, 7:33 a.m. UTC | #3
Hi Cleber,

On 8/1/24 03:02, Cleber Rosa wrote:
> On Mon, Jul 29, 2024 at 6:30 AM Daniel P. Berrangé <berrange@redhat.com> wrote:
>> On Fri, Jul 26, 2024 at 09:44:28AM -0400, Cleber Rosa wrote:
>>> Based on many runs, the average run time for these 4 tests is around
>>> 250 seconds, with 320 seconds being the ceiling.  In any way, the
>>> default 120 seconds timeout is inappropriate in my experience.
>>> Let's increase the timeout so these tests get a chance to completion.
>> A high watermark of over 5 minutes is pretty long for a test.
>>
> I agree.
>
>> Looking at the test I see it runs
>>
>>    self.ssh_command('dnf -y install numactl-devel')
>>
>> but then never actually uses the installed package.
>>
>> I expect that most of the wallclock time here is coming from having
>> dnf download all the repodata, 4 times over.
>>
> Exactly.
>
>> If the intention was to test networking, then replace this with
>> something that doesn't have to download 100's of MB of data, then
>> see what kind of running time we get before increasing any timeout.
>>
>>
> I was trying not to get in the way of the original test writer.
>
> Eric,
>
> Are you OK with replacing this command for a simpler file transfer?
> Any suggestions?
Sorry I am just returning from PTO. Yes I am OK to replace it with a
file transfer if it takes too much time. When the test was written we
just found out a bug with that exact dnf install command so at that time
this was a reproducer. Either replace it with a dummy file transfer or
withdraw it from the automated tests.

Thanks

Eric
>
> Regards,
> - Cleber.
>
diff mbox series

Patch

diff --git a/tests/avocado/intel_iommu.py b/tests/avocado/intel_iommu.py
index 008f214397..9e7965c5df 100644
--- a/tests/avocado/intel_iommu.py
+++ b/tests/avocado/intel_iommu.py
@@ -25,6 +25,8 @@  class IntelIOMMU(LinuxTest):
     :avocado: tags=flaky
     """
 
+    timeout = 360
+
     IOMMU_ADDON = ',iommu_platform=on,disable-modern=off,disable-legacy=on'
     kernel_path = None
     initrd_path = None