curtin-dev team mailing list archive

[Chad Smith <chad.smith@xxxxxxxxxxxxx>]

Diff comments:

> diff --git a/tests/vmtests/test_basic.py b/tests/vmtests/test_basic.py
> index bb6d3a4..b442d8b 100644
> --- a/tests/vmtests/test_basic.py
> +++ b/tests/vmtests/test_basic.py
> @@ -132,7 +132,11 @@ class TestBasicAbs(VMBaseClass):
>  
>      def test_partition_numbers(self):
>          # pnum_disk should have partitions 1 2, and 10
> -        disk = self._dname_to_kname('pnum_disk')
> +        if self.target_release != 'centos66':
> +            disk = self._dname_to_kname('pnum_disk')
> +        else:
> +            disk = self._serial_to_kname('disk-d')

Is it worth noting here for future us that we may sometimes experience intermittent CI errors that we were trying to avoid with your dname_to_kname changes?

> +
>          expected = [disk + s for s in ["", "1", "2", "10"]]
>          self._test_partition_numbers(disk, expected)
>  
> @@ -291,7 +295,10 @@ class TestBasicScsiAbs(TestBasicAbs):
>  
>      def test_partition_numbers(self):
>          # pnum_disk should have partitions 1, 2, and 10
> -        disk = self._dname_to_kname('pnum_disk')
> +        if self.target_release != 'centos66':
> +            disk = self._dname_to_kname('pnum_disk')
> +        else:
> +            disk = 'sdd'

Same question here too. Can we comment that this could lead to intermittent CI errors if disks are swapped?

>          expected = [disk + s for s in ["", "1", "2", "10"]]
>          self._test_partition_numbers(disk, expected)
>  


-- 
https://code.launchpad.net/~raharper/curtin/+git/curtin/+merge/382704
Your team curtin developers is requested to review the proposed merge of ~raharper/curtin:fix/vmtest-centos-basic into curtin:master.