Re: ANNOUNCE: mdadm 4.0 - A tool for managing md Soft RAID under Linux

[Jes Sorensen]

On 01/11/17 23:24, Guoqing Jiang wrote:
> 
> 
> On 01/12/2017 12:59 AM, Jes Sorensen wrote:
>> On 01/11/17 11:52, Shaohua Li wrote:
>>> On Tue, Jan 10, 2017 at 11:49:04AM -0600, Bruce Dubbs wrote:
>>>> Jes Sorensen wrote:
>>>>> I am pleased to announce the availability of
>>>>>      mdadm version 4.0
>>>>>
>>>>> It is available at the usual places:
>>>>>      http://www.kernel.org/pub/linux/utils/raid/mdadm/
>>>>> and via git at
>>>>>      git://git.kernel.org/pub/scm/utils/mdadm/mdadm.git
>>>>>      http://git.kernel.org/cgit/utils/mdadm/
>>>>>
>>>>> The update in major version number primarily indicates this is a
>>>>> release by it's new maintainer. In addition it contains a large number
>>>>> of fixes in particular for IMSM RAID and clustered RAID support.  In
>>>>> addition this release includes support for IMSM 4k sector drives,
>>>>> failfast and better documentation for journaled RAID.
>>>> Thank you for the new release.  Unfortunately I get 9 failures
>>>> running the
>>>> test suite:
>>>>
>>>> tests/00raid1...          FAILED
>>>> tests/07autoassemble...   FAILED
>>>> tests/07changelevels...   FAILED
>>>> tests/07revert-grow...    FAILED
>>>> tests/07revert-inplace... FAILED
>>>> tests/07testreshape5...   FAILED
>>>> tests/10ddf-fail-twice... FAILED
>>>> tests/20raid5journal...   FAILED
>>>> tests/10ddf-incremental-wrong-order...  FAILED
>>> Yep, several tests usually fail. It appears some checks aren't always
>>> good.  At
>>> least the 'check' function for reshape/resync isn't reliable in my
>>> test, I saw
>>> 07changelevelintr fails frequently.
>> That is my experience as well - some of them are affected by the kernel
>> version too. We probably need to look into making them more reliable.
> 
> If possible, it could be a potential topic for lsf/mm raid discussion as
> Coly suggested
> in previous mail.
> 
> Is current test can run the test for different raid level, say, "./test
> --raidtype=raid1" could
> execute all the *r1* tests, does it make sense to do it if we don't
> support it now.

We could have a discussion about this at LSF/MM, if someone is willing
to sponsor getting it accepted and we can get the right people there.

Note that the test suite also allows you to run all the 01 tests by
specifying ./test 01. I do like to see the test suite improved and made
more resilient.

Cheers,
Jes