Bug 154596
Summary: | kernel panics when running memory test with 32gb ram | ||||||
---|---|---|---|---|---|---|---|
Product: | Red Hat Enterprise Linux 3 | Reporter: | erik nguyen <erik.nguyen> | ||||
Component: | kernel | Assignee: | Jim Paradis <jparadis> | ||||
Status: | CLOSED NOTABUG | QA Contact: | |||||
Severity: | high | Docs Contact: | |||||
Priority: | medium | ||||||
Version: | 3.0 | CC: | linux-bugs, peterm, petrides, tao, wtogami | ||||
Target Milestone: | --- | ||||||
Target Release: | --- | ||||||
Hardware: | x86_64 | ||||||
OS: | Linux | ||||||
Whiteboard: | |||||||
Fixed In Version: | Doc Type: | Bug Fix | |||||
Doc Text: | Story Points: | --- | |||||
Clone Of: | Environment: | ||||||
Last Closed: | 2005-10-04 00:35:33 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: | |||||||
Attachments: |
|
Description
erik nguyen
2005-04-12 21:47:26 UTC
Why did you report this against memtest86+? memtest86+ has nothing to do with the kernel. Also you cannot run an i686 hugemem kernels with a x86_64 64-bit userspace. i686 kernels can only run 32bit programs. Oops, sorry, thanks for moving it to kernel, where it belongs. Please ignore the comments about the hugemem i686 kernel - this is happening with x86_64 (the tester recalled the multiple kernels provided on i686, and thought he should also test with the hugemem kernel provided as part of x86_64 bits). The problem is simply 32gb memory + rhel 3 update 5 beta + v40z = kernel panic with memtest. Created attachment 113259 [details]
lspci -vvvx output
http://www.memtest.org/ Please test it with memtest86+ for at least 24 hours to be 100% sure all 32GB of RAM is good and there exists no other hardware trouble. Yes this is unlikely but it is better to be sure than waste time of engineers, both yours and ours. The failure is proving elusive/difficult to reproduce - we've had 3 systems rerunning tests for multiple days without recreating. Recently a processor errata was announced about stale tlb entries - this may be a possible root cause. Closing due to expectation that this was not a kernel problem. |