Public bug reported:
Several different systems, both vm and bare metal, running Ubuntu 20.04
LTS and different Ubuntu kernel versions with the generic kernel,
5.4.0-91.81, a mainline 5.15 build from kernel.ubuntu.com, 5.11-hwe,
show random panics.
The crash dumps all show panics within the SLUB memory management stuff.
Analyzing the kdump show invalid free list pointers:
CACHE OBJSIZE ALLOCATED TOTAL SLABS SSIZE NAME
kmem: kmalloc-256: slab: ffffefc73e319900 invalid freepointer: d28996eab0548a12
ffff8a863dc06f40 256 15454 76096 1189 16k kmalloc-256
The kernel log shows warnings about wrong slab cache:
[148148.037307] cache_from_obj: Wrong slab cache. jbd2_journal_handle but
object is from kmalloc-256
[148148.037348] WARNING: CPU: 20 PID: 4141624 at mm/slab.h:521
kmem_cache_free+0x260/0x2b0
This shows up if there are several hundred MB per second backup traffic
via ceph (I think using rbd). Sometimes it only takes minutes for the
system to panic.
Hardware:
HP DL380 Gen10
AMD EPYC 7502 32-Core Processor
256GB RAM
Firmware and microcode are uptodate.
I attached one backtraces, the warning message, the panic.
** Affects: linux (Ubuntu)
Importance: Undecided
Status: New
** Attachment added: "Panic 1"
https://bugs.launchpad.net/bugs/1952425/+attachment/5543591/+files/out
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1952425
Title:
SLUB freelist corruption with ceph
Status in linux package in Ubuntu:
New
Bug description:
Several different systems, both vm and bare metal, running Ubuntu
20.04 LTS and different Ubuntu kernel versions with the generic
kernel, 5.4.0-91.81, a mainline 5.15 build from kernel.ubuntu.com,
5.11-hwe, show random panics.
The crash dumps all show panics within the SLUB memory management
stuff. Analyzing the kdump show invalid free list pointers:
CACHE OBJSIZE ALLOCATED TOTAL SLABS SSIZE NAME
kmem: kmalloc-256: slab: ffffefc73e319900 invalid freepointer:
d28996eab0548a12
ffff8a863dc06f40 256 15454 76096 1189 16k kmalloc-256
The kernel log shows warnings about wrong slab cache:
[148148.037307] cache_from_obj: Wrong slab cache. jbd2_journal_handle but
object is from kmalloc-256
[148148.037348] WARNING: CPU: 20 PID: 4141624 at mm/slab.h:521
kmem_cache_free+0x260/0x2b0
This shows up if there are several hundred MB per second backup
traffic via ceph (I think using rbd). Sometimes it only takes minutes
for the system to panic.
Hardware:
HP DL380 Gen10
AMD EPYC 7502 32-Core Processor
256GB RAM
Firmware and microcode are uptodate.
I attached one backtraces, the warning message, the panic.
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1952425/+subscriptions
--
Mailing list: https://launchpad.net/~kernel-packages
Post to : kernel-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kernel-packages
More help : https://help.launchpad.net/ListHelp
