Project

General

Profile

Actions
Copy link

Bug #8859

closed

krbd crash while serving linux-lio iscsi: rbd_assert(img_request != NULL);

Added by Walter Huf almost 10 years ago. Updated over 9 years ago.

Status:
Closed
Priority:
Normal
Assignee:
-
Target version:
-
% Done:

0%

Source:
other
Tags:
Backport:
Regression:
Severity:
2 - major
Reviewed:
Affected Versions:
ceph-qa-suite:
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

We have Linux-HA configuring a pair of nodes to make highly-available iSCSI targets with Linux-LIO, and so it maps the RBD on all the nodes and has a roving iSCSI resource that wanders around the cluster.
Sometime every night, the node that is hosting the iSCSI resource kernel panics and shows a stack trace, with varying degrees of hard death. This morning, the node was able to log some messages to kern.log, and so I've attached them to this report.

The Ceph cluster is running Emperor (0.72.2-1precise). The iSCSI target hosts are Ubuntu 12.04, and I've tried a few different kernels to see if it alleviated issues. This report is from the server running the Ubuntu 12.04 package linux-image-generic-lts-trusty 3.13.0.30.26, which has installed kernel version 3.13.0-30.

Files

kern.20140717.log (1.76 MB) kern.20140717.log Kernel log of the crash Walter Huf, 07/17/2014 07:10 AM
Actions
Copy link

Also available in: Atom PDF