Comment 64 for bug 245779

Following up on my previous post. I have downgraded the kernel back to 2.6.18-2 on 20 of our guests. These machines have been running for over a week with no incident. We have been able to reproduce this issue on 2.6.26-2 very consistently. We also attempted to upgrade to 2.6.30 but we experienced a very similar symptom, however not identical. In the 2.6.30 case both cpus locked up, not just one.

Our application runs the following: JBoss, MySQL. The application is very disk I/O intensive and generates a fair share of Network IO as well.

Came across a changelog entry for 2.6.31 that may be interesting:

http://www.kernel.org/pub/linux/kernel/v2.6/testing/ChangeLog-2.6.31-rc8