linux-xfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: bugzilla-daemon@kernel.org
To: linux-xfs@vger.kernel.org
Subject: [Bug 216007] XFS hangs in iowait when extracting large number of files
Date: Sat, 21 May 2022 22:31:59 +0000	[thread overview]
Message-ID: <bug-216007-201763-4bIsuoCV4o@https.bugzilla.kernel.org/> (raw)
In-Reply-To: <bug-216007-201763@https.bugzilla.kernel.org/>

https://bugzilla.kernel.org/show_bug.cgi?id=216007

--- Comment #5 from Dave Chinner (david@fromorbit.com) ---
On Sat, May 21, 2022 at 05:14:36AM +0000, bugzilla-daemon@kernel.org wrote:
> https://bugzilla.kernel.org/show_bug.cgi?id=216007
> 
> --- Comment #4 from Peter Pavlisko (bugzkernelorg8392@araxon.sk) ---
> > What sort of storage subsystem does this machine have? If it's a spinning
> > disk then you've probably just filled memory
> 
> Yes, all the disks are classic spinning CMR disks. But, out of all file
> systems
> tried, only XFS is doing this on the test machine. I can trigger this
> behavior
> every time. And kernels from 5.10 and bellow still work, even with my
> non-standard .config.
> 
> Here is the memory situation when it is stuck:
> 
> ftp-back ~ # free
>                total        used        free      shared  buff/cache  
>                available
> Mem:         3995528      175872       69240         416     3750416    
> 3763584

Doesn't tell us a whole lot except for "no free memory to allocate
without reclaim". /proc/meminfo, /proc/vmstat and /proc/slabinfo
would tell us a lot more.

Also, knowing if you've tweaked things like dirty ratios, etc would
also be helpful...

> This may not be a XFS bug, but so far only XFS seems to suffer from it.

Not that uncommon, really. XFS puts a different load on the memory
allocation/reclaim and cache subsystems compared to other
filesystems, so XFS tends to trip over bugs that others don't.

Cheers,

Dave.

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.

  parent reply	other threads:[~2022-05-21 22:32 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-20 11:56 [Bug 216007] New: XFS hangs in iowait when extracting large number of files bugzilla-daemon
2022-05-20 11:56 ` [Bug 216007] " bugzilla-daemon
2022-05-20 20:46 ` bugzilla-daemon
2022-05-20 23:05 ` [Bug 216007] New: " Dave Chinner
2022-05-20 23:05 ` [Bug 216007] " bugzilla-daemon
2022-05-21  5:14 ` bugzilla-daemon
2022-05-21 22:31   ` Dave Chinner
2022-05-21 22:31 ` bugzilla-daemon [this message]
2022-05-23  8:29 ` bugzilla-daemon
2022-05-23  8:31 ` bugzilla-daemon
2022-05-23 10:02 ` bugzilla-daemon
2022-05-23 10:28 ` bugzilla-daemon
2022-05-24  7:54 ` bugzilla-daemon
2022-05-24 10:00 ` bugzilla-daemon
2022-05-24 10:49 ` bugzilla-daemon
2022-05-24 10:52 ` bugzilla-daemon
2022-05-24 10:53 ` bugzilla-daemon
2022-05-24 11:21 ` bugzilla-daemon
2022-05-24 11:48 ` bugzilla-daemon
2022-05-24 11:49 ` bugzilla-daemon
2022-05-25 17:13 ` bugzilla-daemon
2022-05-26  4:04 ` bugzilla-daemon
2022-05-26  8:51 ` bugzilla-daemon
2022-05-26  9:16 ` bugzilla-daemon
2022-05-26 10:26 ` bugzilla-daemon
2022-05-26 10:37 ` bugzilla-daemon
2022-06-04 16:25 ` bugzilla-daemon
2022-06-05  7:51 ` bugzilla-daemon
2022-06-06  7:49 ` bugzilla-daemon

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=bug-216007-201763-4bIsuoCV4o@https.bugzilla.kernel.org/ \
    --to=bugzilla-daemon@kernel.org \
    --cc=linux-xfs@vger.kernel.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).