From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; .NET CLR 1.1.4322) Description of problem: We are running a cluster of 16 Dual Xeons (Supermicro X5DPR-iG2). The m'board is based on the 7501 chipset with onboard dual Intel PRO1000 NIC's. There is heavy cpu usage 24/7 and the network traffic can be heavy but intermittent. Over the last 6 months we have experienced random lock ups on several different nodes. We have not been able to reproduce the problem at will but it recurs every few weeks. After the lockup the computer still responds to a ping but not to a remote or console command. A reboot always clears the problem (until the next time). There are never any messages in the system log files. We observed this problem with the stock RH8 kernel and then upgraded to 2.4.20, which exhibits the same problem. Version-Release number of selected component (if applicable): How reproducible: Couldn't Reproduce Steps to Reproduce: 1. Run 16 nodes with heavy cpu and heavy but intermittent network traffic. 2. Wait 1 month 3. Actual Results: Periodic freezing of one of the nodes Expected Results: Kept running Additional info: There are similar bugs listed, but I am not sure they are the same. They seem to occur more frequently for one thing. 101264 (laptop) 79997 (supposed to be tg3 driver issue, but some reports of problems with PRO1000)
Can you try the latest errata kernel ? It's possible you're being hit by some of the obscure bugs fixed in 2.4.20-20
I updated to 2.4.20-20smp last October as you suggested, and more recently to 2.4.20-28.8smp. There has been no recurrence so I consider the issue closed. Thanks
Thanks for the bug report. However, Red Hat no longer maintains this version of the product. Please upgrade to the latest version and open a new bug if the problem persists. The Fedora Legacy project (http://fedoralegacy.org/) maintains some older releases, and if you believe this bug is interesting to them, please report the problem in the bug tracker at: http://bugzilla.fedora.us/