From: Christopher Zimmermann <christopher@gmerlin.de>
To: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Cc: linux-nilfs@vger.kernel.org
Subject: Re: nilfs_readdir: bad page in #
Date: Tue, 4 Nov 2025 17:04:18 +0100 [thread overview]
Message-ID: <aQokBe0xUtcitVnj@merari.gmerlin.de> (raw)
In-Reply-To: <CAKFNMonYLKSikthtoGP9z6Loetu0LxrUsGC6vMSAwaCPE6muqQ@mail.gmail.com>
[-- Attachment #1: Type: text/plain, Size: 5014 bytes --]
On Wed, Nov 05, 2025 at 12:28:54AM +0900, Ryusuke Konishi wrote:
>Based on what's happening with the .xsession-errors file, I guess that
>the problem is indeed occurring with files that are renamed or have a
>short lifespan.
>
>I'll try to create an environment that reproduces these conditions.
Not sure if this helps. Here are some more observations:
The nilfs_readdir: bad page in #170719 error was triggered by a
find ~/ -iname …
The following
nilfs_bmap_lookup_contig: broken bmap (inode number=259)
errors were then triggered by running a backup (restic) or simply
reading ~/.xsession-errors. Some content of the broken
~/.xsession-errors could then still be read.
What I noticed now while killing processes preventing the unmounting of
the errored filesystem is that it is passed to many x-clients as
stdout/stderr fd:
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
herbstluf 1506 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
herbstluf 1506 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
dunst 1642 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
dunst 1642 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
conky 1643 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
conky 1643 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
nm-applet 1644 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
nm-applet 1644 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
blueman-a 1645 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
blueman-a 1645 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
firefox-e 1655 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
firefox-e 1655 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
crashhelp 1727 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
crashhelp 1727 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
blueman-t 1821 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
blueman-t 1821 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
polybar 2069 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
polybar 2069 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
Socket 2087 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
Socket 2087 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
Privilege 2122 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
Privilege 2122 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
RDD\x20Pr 2129 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
RDD\x20Pr 2129 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
Isolated 2167 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
Isolated 2167 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
WebExtens 2230 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
WebExtens 2230 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
Utility 2317 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
Utility 2317 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
Web\x20Co 2449 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
Web\x20Co 2449 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
Web\x20Co 2525 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
Web\x20Co 2525 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
Web\x20Co 2559 madroach 1w REG 259,5 3637 11 /home/madroach/.xsession-errors
Web\x20Co 2559 madroach 2w REG 259,5 3637 11 /home/madroach/.xsession-errors
Most content comes from herbstluftwm logging unbound keypresses, but
that's not a huge amount of logs. A typical .xsession-errors on that
machine has less than or a few hundred lines of content.
Logging the inode numbers reveals back-and-forth renameing of
.xsession-errors and .xsession-errors.old:
11 -rw------- 1 madroach users 5,0K 4. Nov 16:50 .xsession-errors
126604 -rw------- 1 madroach users 11K 4. Nov 15:03 .xsession-errors.old
120042 -rw------- 1 madroach users 3,2K 4. Nov 16:51 .xsession-errors
11 -rw------- 1 madroach users 5,5K 4. Nov 16:51 .xsession-errors.old
11 -rw------- 1 madroach users 3,2K 4. Nov 16:54 .xsession-errors
120042 -rw------- 1 madroach users 4,4K 4. Nov 16:52 .xsession-errors.old
118708 -rw------- 1 madroach users 3,0K 4. Nov 16:55 .xsession-errors
11 -rw------- 1 madroach users 4,2K 4. Nov 16:55 .xsession-errors.old
Thanks for you efforts,
Christopher
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]
next prev parent reply other threads:[~2025-11-04 16:04 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-17 14:06 nilfs_readdir: bad page in # Christopher Zimmermann
2025-10-18 11:04 ` Ryusuke Konishi
2025-11-04 14:00 ` Christopher Zimmermann
2025-11-04 15:28 ` Ryusuke Konishi
2025-11-04 16:04 ` Christopher Zimmermann [this message]
2025-11-04 16:43 ` Christopher Zimmermann
2025-11-05 15:01 ` Hideki EIRAKU
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=aQokBe0xUtcitVnj@merari.gmerlin.de \
--to=christopher@gmerlin.de \
--cc=konishi.ryusuke@gmail.com \
--cc=linux-nilfs@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).