linux-kernel - Re: [REGRESSION] 6.7.1: md: raid5 hang and unresponsive system; successfully bisected

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <c9b3466c-7972-47c0-b590-9c1012b53a82@oracle.com>
Date: Wed, 21 Feb 2024 11:15:35 -0800
From: junxiao.bi@...cle.com
To: Mateusz Kusiak <mateusz.kusiak@...ux.intel.com>,
        Dan Moulding <dan@...m.net>
Cc: gregkh@...uxfoundation.org, linux-kernel@...r.kernel.org,
        linux-raid@...r.kernel.org, regressions@...ts.linux.dev,
        song@...nel.org, stable@...r.kernel.org
Subject: Re: [REGRESSION] 6.7.1: md: raid5 hang and unresponsive system;
 successfully bisected

On 2/21/24 6:50 AM, Mateusz Kusiak wrote:
> On 21.02.2024 00:15, junxiao.bi@...cle.com wrote:
>>
>> The thing is we can't reproduce this issue at all. If you can 
>> generate a vmcore when the hung happened, then we can review which 
>> processes are stuck.
>>
> Hi,
> don't know if that be any of help, but I run below scenario with SATA 
> and NVMe drives. For me, the issue is reproducible on NVMe drives only.
>
> Scenario:
> 1. Create R5D3 with native metadata
>     # mdadm -CR /dev/md/vol -l5 -n3 /dev/nvme[0-2]n1 --assume-clean
> 2. Create FS on the array
>     # mkfs.ext4 /dev/md/vol -F
> 3. Remove single member drive via "--incremental --fail"
>     # mdadm -If nvme0n1
>
> The result is almost instant.

This is not the same issue that Dan reported, it looks like another 
regression that Yu Kuai fixed , can you please try this patch?

https://lore.kernel.org/lkml/95f2e08e-2daf-e298-e696-42ebfa7b9bbf@huaweicloud.com/

Thanks,

Junxiao.

>
> Thanks,
> Mateusz