Bug 126729
Summary: | Kernel Hang on Dual Xeon System | ||
---|---|---|---|
Product: | Red Hat Enterprise Linux 3 | Reporter: | ravi <ravi> |
Component: | kernel | Assignee: | Arjan van de Ven <arjanv> |
Status: | CLOSED NOTABUG | QA Contact: | Brian Brock <bbrock> |
Severity: | high | Docs Contact: | |
Priority: | medium | ||
Version: | 3.0 | CC: | john, petrides, riel |
Target Milestone: | --- | ||
Target Release: | --- | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | Bug Fix | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2004-06-26 03:13:04 UTC | Type: | --- |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: |
Description
ravi
2004-06-25 12:51:24 UTC
Ummm, kernel 2.6.4 ???? I don't think Red Hat has ever shipped that kernel as part of any distribution. Fedora Core 2 is 2.6.5 or 2.6.6 based and RHEL3 is kernel 2.4.21 based. Exactly what distribution and kernel are you using ? Hello, Please refr the following information : ======================================================== root@perceptions [/etc]# uname -r 2.6.4 root@perceptions [/etc]# uname -a Linux perceptions.jhservers209.com 2.6.4 #1 SMP Tue Mar 16 01:46:57 EST 2004 i686 i686 i386 GNU/Linux root@perceptions [/etc]# ======================================================== Thank you, Best regards, ravi This is not a kernel from Red Hat. The 2.4.21-4.ELsmp and the newer 2.4.21-15.0.3.ELsmp kernels both hang with dual Xeon processors every few hours. It seems random but is probably not. A few minutes ago it hung when looking up a tag definition with nedit. Several times it hung while compiling and linking. If you don't run anything except normal system activities, once in a while it will run overnight. Usually not. When I say hung, I mean you can't ping it, there's no cursor and it never comes back (if you consider two days sufficient to say "never"). You have to manually reset the system. It does not seem to happen with the 2.4.21-4.EL uniprocessor kernel. I think you have a kernel race condition in your threads code. As you may know, the Xeon appears as two CPUs to the kernel and can execute two threads simultaneously (4 for a dual Xeon system). Perhaps this creates a problem. I have tried 2.4.21-4.ELsmp on two dual Xeon systems with the same results. |