From: "Chris Murphy" <lists@colorremedies.com>
To: "Josef Bacik" <josef@toxicpanda.com>, paolo.valente@linaro.org
Cc: "Btrfs BTRFS" <linux-btrfs@vger.kernel.org>,
Linux-RAID <linux-raid@vger.kernel.org>,
linux-block@vger.kernel.org,
linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: stalling IO regression since linux 5.12, through 5.18
Date: Fri, 12 Aug 2022 12:05:12 -0400 [thread overview]
Message-ID: <e36fe80f-a33b-4750-b593-3108ba169611@www.fastmail.com> (raw)
In-Reply-To: <ad78a32c-7790-4e21-be9f-81c5848a4953@www.fastmail.com>
On Wed, Aug 10, 2022, at 3:34 PM, Chris Murphy wrote:
> Booted with cgroup_disable=io, and confirmed cat
> /sys/fs/cgroup/cgroup.controllers does not list io.
The problem still reproduces with the cgroup IO controller disabled.
On a whim, I decided to switch the IO scheduler from Fedora's default bfq for rotating drives to mq-deadline. The problem does not reproduce for 15+ hours, which is not 100% conclusive but probably 99% conclusive. I then switched live while running the workload to bfq on all eight drives, and within 10 minutes the system cratered, all new commands just hang. Load average goes to triple digits, i/o wait increasing, i/o pressure for the workload tasks to 100%, and IO completely stalls to zero. I was able to switch only two of the drive queues back to mq-deadline and then lost responsivness in that shell and had to issue sysrq+b...
Before that I was able to extra sysrq+w and sysrq+t.
https://drive.google.com/file/d/16hdQjyBnuzzQIhiQT6fQdE0nkRQJj7EI/view?usp=sharing
I can't tell if this is a bfq bug, or if there's some negative interaction between bfq and scsi or megaraid_sas. Obviously it's rare because otherwise people would have been falling over this much sooner. But at this point there's strong correlation that it's bfq related and is a kernel regression that's been around since 5.12.0 through 5.18.0, and I suspect also 5.19.0 but it's being partly masked by other improvements.
--
Chris Murphy
next prev parent reply other threads:[~2022-08-12 16:05 UTC|newest]
Thread overview: 58+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-08-10 16:35 stalling IO regression in linux 5.12 Chris Murphy
2022-08-10 17:48 ` Josef Bacik
2022-08-10 18:33 ` Chris Murphy
2022-08-10 18:42 ` Chris Murphy
2022-08-10 19:31 ` Josef Bacik
2022-08-10 19:34 ` Chris Murphy
2022-08-12 16:05 ` Chris Murphy [this message]
2022-08-12 17:59 ` stalling IO regression since linux 5.12, through 5.18 Josef Bacik
2022-08-12 18:02 ` Jens Axboe
2022-08-14 20:28 ` Chris Murphy
2022-08-16 14:22 ` Chris Murphy
2022-08-16 15:25 ` Nikolay Borisov
2022-08-16 15:34 ` Chris Murphy
2022-08-17 9:52 ` Holger Hoffstätte
2022-08-17 11:49 ` Jan Kara
2022-08-17 14:37 ` Chris Murphy
2022-08-17 15:09 ` Chris Murphy
2022-08-17 16:30 ` Jan Kara
2022-08-17 16:47 ` Chris Murphy
2022-08-17 17:57 ` Chris Murphy
2022-08-17 18:15 ` Jan Kara
2022-08-17 18:18 ` Chris Murphy
2022-08-17 18:33 ` Jan Kara
2022-08-17 18:54 ` Chris Murphy
2022-08-17 19:23 ` Chris Murphy
2022-08-18 2:31 ` Chris Murphy
2022-08-17 18:21 ` Holger Hoffstätte
2022-08-17 11:57 ` Chris Murphy
2022-08-17 12:31 ` Holger Hoffstätte
2022-08-17 18:16 ` Chris Murphy
2022-08-17 18:38 ` Holger Hoffstätte
2022-08-17 12:06 ` Ming Lei
2022-08-17 14:34 ` Chris Murphy
2022-08-17 14:53 ` Ming Lei
2022-08-17 15:02 ` Chris Murphy
2022-08-17 15:34 ` Ming Lei
2022-08-17 16:34 ` Chris Murphy
2022-08-18 1:03 ` Ming Lei
2022-08-18 2:30 ` Chris Murphy
2022-08-18 3:24 ` Ming Lei
2022-08-18 4:12 ` Chris Murphy
2022-08-18 4:18 ` Chris Murphy
2022-08-18 4:27 ` Chris Murphy
2022-08-18 4:32 ` Chris Murphy
2022-08-18 5:15 ` Ming Lei
2022-08-18 18:52 ` Chris Murphy
2022-08-18 5:24 ` Ming Lei
2022-08-18 13:50 ` Chris Murphy
2022-08-18 15:10 ` Ming Lei
2022-08-19 19:20 ` Chris Murphy
2022-08-20 7:00 ` Ming Lei
2022-09-01 7:02 ` Yu Kuai
2022-09-01 8:03 ` Jan Kara
2022-09-01 8:19 ` Yu Kuai
2022-09-06 9:49 ` Paolo Valente
2022-09-02 16:53 ` Chris Murphy
2022-09-06 9:45 ` Paolo Valente
2022-08-15 11:25 ` stalling IO regression in linux 5.12 Thorsten Leemhuis
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=e36fe80f-a33b-4750-b593-3108ba169611@www.fastmail.com \
--to=lists@colorremedies.com \
--cc=josef@toxicpanda.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-btrfs@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-raid@vger.kernel.org \
--cc=paolo.valente@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).