Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.debian.kernel > #77918

Bug#1025537: nfsd: Kernel Oops while serving NFS

From Olivier Mehani <shtrom+debian@ssji.net>
Newsgroups linux.debian.bugs.dist, linux.debian.kernel
Subject Bug#1025537: nfsd: Kernel Oops while serving NFS
Date 2023-01-14 03:20 +0100
Message-ID <FNEjf-fov-1@gated-at.bofh.it> (permalink)
References <FzF5v-8LUu-1@gated-at.bofh.it>
Organization linux.* mail to news gateway

Cross-posted to 2 groups.

Show all headers | View raw


Package: linux-image-5.19.0-10315-g310d9d5a5009
Followup-For: Bug #1025537

Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
From: Olivier Mehani <shtrom+debian@ssji.net>
To: Debian Bug Tracking System <1025537@bugs.debian.org>
Subject: Re: nfsd: Kernel Oops while serving NFS
Message-ID: <167361292314.1643173.3991370234085262557.reportbug@supahwinch.narf.ssji.net>
X-Mailer: reportbug 11.6.0
Date: Fri, 13 Jan 2023 23:28:43 +1100

Package: linux-image-5.19.0-10316-gf0f6b614f83d
Followup-For: Bug #1025537

OK, I think I have a strong suspect [0]: 

   f0f6b614f83dbae99d283b7b12ab5dd2e04df979 (tags/v6.0-rc1~55^2~2)

This is a pretty generic commit, that doesn't seem to touch the nfs 
stack, but changes something that looks deep enough in the iovec code 
that it could have an impact.

I have solidly re-tested the preceding commit,
310d9d5a5009a93377200b98daa2d84aa2bd8160 (tags/v6.0-rc1~55^2~3), and 
the Oops doesn't happen there.

There is a series of commits that were suggested as part of the `git 
bisect`, that I wasn't able to build and had to skip. The error was

   FAILED: load BTF from vmlinux: Invalid argument

but no amount of cajoling `pahole` has helped.

The merge parent for the series is at 
8447d0e75099eb54eea9306c2d43ecfc956d09ed (tags/v6.0-rc1~56^2), which is 
the preceding merge to the one containing the first bad commit above. I 
think this is noteworthy as a lot of the commits on this branch are 
labeled as `remoteproc` which I, as a lay person, think might be related 
to RPCs.

Is there any other tests that I can perform to help diagnose / fix this 
further?

[0] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=f0f6b614f83dbae99d283b7b12ab5dd2e04df97 (but dockerd Oops, 10623384), nah, BAD


-- System Information:
Debian Release: bookworm/sid
  APT prefers testing
  APT policy: (990, 'testing')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 5.19.0-10315-g310d9d5a5009 (SMP w/2 CPU threads; PREEMPT)
Locale: LANG=en_AU, LC_CTYPE=en_AU (charmap=UTF-8) (ignored: LC_ALL set to en_AU.UTF8), LANGUAGE=en_AU:en_GB
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

-- no debconf information

Back to linux.debian.kernel | Previous | Next | Find similar | Unroll thread


Thread

Bug#1025537: nfsd: Kernel Oops while serving NFS Olivier Mehani <shtrom+debian@ssji.net> - 2023-01-14 03:20 +0100

csiph-web