ACK/Cmnt: [ACT][PATCH] UBUNTU: SAUCE: ubuntu_lxc: destroy leftover container before tests

[Kleber Souza]

On 12.07.21 17:37, Krzysztof Kozlowski wrote:
> On 12/07/2021 12:59, Kleber Sacilotto de Souza wrote:
>> BugLink: https://bugs.launchpad.net/bugs/1788574
>>
>> Commits 1a50e77ca803 ("UBUNTU: SAUCE: ubuntu_lxc: remove the reboot
>> container after test") and 298cd01e ("UBUNTU: SAUCE: ubuntu_lxc: destroy
>> the leftover container before exit") fixed the issue with the leftover
>> "reboot" contained on the 'exercise' script. However, this script is run
>> only for series older than artful. For newer series, the testcase is run
>> from the source package via 'autopkgtest'. Fix it by destroying the
>> container before running any test.
>>
>> Signed-off-by: Kleber Sacilotto de Souza <kleber.souza at canonical.com>
>> ---
>>   ubuntu_lxc/ubuntu_lxc.py | 6 ++++++
>>   1 file changed, 6 insertions(+)
>>
>> diff --git a/ubuntu_lxc/ubuntu_lxc.py b/ubuntu_lxc/ubuntu_lxc.py
>> index e201b0d4..cc8466ba 100644
>> --- a/ubuntu_lxc/ubuntu_lxc.py
>> +++ b/ubuntu_lxc/ubuntu_lxc.py
>> @@ -44,6 +44,12 @@ class ubuntu_lxc(test.test):
>>       def run_once(self, test_name):
>>           if test_name == 'setup':
>>               return
>> +
>> +        # Destroy the "reboot" container which might have been left
>> +        # behind (LP#1788574)
>> +        cmd = 'lxc-destroy reboot'
>> +        utils.system(cmd, ignore_status=True)
>> +
>>           if self.series in ['precise', 'trusty', 'xenial', 'artful']:
>>               cmd = '/bin/sh %s/exercise' % self.bindir
>>           else:
>>
> 
> Maybe it should be rather at the end of test in try-finally block to
> leave the machine/container in a clean state? But I don't have any
> strong preference or arguments, so:

I thought of doing it before to make sure that if any previous run that
might have left the container behind would not impact the current run.
But if we can guarantee that doing it after the test is run would
*always* clean it up then it sounds like a better approach. The only
issue would be that the first time we run this script with the fix it
will still fail, but that wouldn't be a big deal.

> 
> Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski at canonical.com>
> 
> 
> Best regards,
> Krzysztof
>