[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#1020831: marked as done (linux-signed-arm64: iouring tmpfs pwrite returns 125 error sporadically)

Your message dated Thu, 20 Feb 2025 15:07:28 +0100 (CET)
with message-id <20250220140728.E4A64BE2EE7@eldamar.lan>
and subject line Closing this bug (BTS maintenance for src:linux bugs)
has caused the Debian Bug report #1020831,
regarding linux-signed-arm64: iouring tmpfs pwrite returns 125 error sporadically
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
1020831: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1020831
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message --- 
Package: linux-signed-arm64


Dear Maintainer,

This is derived from https://jira.mariadb.org/browse/MDEV-29610

MariaDB-10.6+ uses liburing as its default async IO layer. In our CI
infrastructure there are a considerable amount of encryption.innodb*
test errors where uring returns the error IO Error: 125 during async
write of 49152 bytes, for file 11, returned 0. In all cases this is
a pwrite uring call.

We are using various distro containers on the Debian kernel host and
this is a common error.

An example error is:
https://buildbot.mariadb.org/#/builders/482/builds/1493
The encryption.innodb-checksum-algorithm error where the complete
mariadb log is https://ci.mariadb.org/28662/logs/aarch64-rhel-9/mysqld.1.err.6

In this case the kernel versions is printed just before the next
test as:
* 5.10.0-17-cloud-arm64  (debian-kernel@lists.debian.org) (gcc-10 (Debi

A KVM build with a bullseye is:
https://buildbot.askmonty.org/buildbot/builders/kvm-deb-bullseye-aarch64/builds/111

Its kernel version is: 5.10.0-17-arm64 (debian-kernel@lists.debian.org)

Though largely aarch failures have been observed there are ppc64le
examples like:
* https://buildbot.mariadb.org/#/builders/478/builds/1676 error log
https://ci.mariadb.org/28662/logs/ppc64le-rhel-9/mysqld.1.err.11
Linux version 5.10.0-18-powerpc64le (debian-kernel@lists.debian.org)

The one amd64 case is in:
https://ci.mariadb.org/28153/logs/amd64-debian-sid/mysqld.1.err.14
The kernel (though not in the logs) was 5.10.0-13-amd64
(debian-kernel@lists.debian.org)

Our CI works across quite a number of host kernels and the Debian 5.10
is the only one showing these errors. Even newer kernels like
5.18.0-0.deb11 / Debian 5.18.16-1~bpo11+1 (2022-08-12)
isn't showing errors.

So its looking like a missing upstream iouring patch missing.


To reproduce:
(mariadb-10.6+ test)

curl -sS https://downloads.mariadb.com/MariaDB/mariadb_repo_setup | bash
apt-get install -y mariadb-test patch
cd /usr/share/mysql/mysql-test/
mkdir var
chown mysql: var

as mysql user:
./mtr --mem --repeat=1000 --parallel=8
encryption.innodb-compressed-blob{,,,,,,,,,,,,,}


-- System Information:
Debian Release: 11.5
  APT prefers stable-updates
  APT policy: (500, 'stable-updates'), (500, 'stable-security'), (500, 'stable')
Architecture: amd64 (x86_64)

Kernel: Linux version 5.10.0-17-arm64
Kernel taint flags: TAINT_WARN
Locale: LANG=C, LC_CTYPE=C.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /bin/dash
Init: unable to detect

--- End Message ---
--- Begin Message --- 
Hi

This bug was filed for a very old kernel or the bug is old itself
without resolution.

If you can reproduce it with

- the current version in unstable/testing
- the latest kernel from backports

please reopen the bug, see https://www.debian.org/Bugs/server-control
for details.

Regards,
Salvatore

--- End Message ---
Reply to: