Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 434400 - sys-kernel/gentoo-sources-3.3.8: BUG: bad page map
Summary: sys-kernel/gentoo-sources-3.3.8: BUG: bad page map
Status: RESOLVED OBSOLETE
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: [OLD] Core system (show other bugs)
Hardware: All Linux
: Normal major (vote)
Assignee: Gentoo Kernel Bug Wranglers and Kernel Maintainers
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2012-09-09 07:51 UTC by Raymond Jennings
Modified: 2012-11-01 17:15 UTC (History)
0 users

See Also:
Package list:
Runtime testing required: ---


Attachments
huge mess of debug from corrupted swap (swapbug.txt.gz,79.11 KB, text/plain)
2012-09-09 07:57 UTC, Raymond Jennings
Details
emerge --info (einfo.txt,4.92 KB, text/plain)
2012-09-09 07:59 UTC, Raymond Jennings
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Raymond Jennings 2012-09-09 07:51:17 UTC
Where do I begin.

See attached logs.

Summary version:

having 8 different swap devices during an emerge --emptytree caused a large number of bad page errors, and later caused the system to hang when attempting to do an orderly shutdown and it choked on swap partition shutdown.  Also numerous processes segfaulted on "bad page" kernel bugchecks.

My advice is just to mask version 3.3.8 as buggy and raise the urgency of stabilizing a newer kernel.

3.3.8 has been EOL by upstream and now I need to venture into ~arch land to get a working kernel.

emerge --info is forthcoming.
Comment 1 Raymond Jennings 2012-09-09 07:57:57 UTC
Created attachment 323256 [details]
huge mess of debug from corrupted swap
Comment 2 Raymond Jennings 2012-09-09 07:58:56 UTC
Also, swapoff -a managed to get SIGKILL due to BUG in kernel
Comment 3 Raymond Jennings 2012-09-09 07:59:21 UTC
Created attachment 323258 [details]
emerge --info
Comment 4 Raymond Jennings 2012-09-09 07:59:51 UTC
Also quick note

The debug log for corrupted swap was so large it has been gzipped.
Comment 5 Mike Pagano gentoo-dev 2012-09-09 17:49:39 UTC
Can you please venture in ~arch land and let me know if it's fixed? I have no plans on releasing any more kernels in this version.
Comment 6 Raymond Jennings 2012-09-10 17:23:18 UTC
I'm not sure if it's fixed but venturing into 3.5 land is already improving things in other areas.  I took the initiative of doing this precisely because of the point I reemphasise in the next paragraph.

As I mentioned in my first post however I was advocating that 3.3.8 be masked for removal for a) being buggy and b) being EOL'ed upstream.

I would rather as mentioned appreciate it if stabilization was stepped up on the newer versions.

I consider 3.5.3 stable enough to use, personally, and I'm using it right now.
Comment 7 Tomasz Chilinski 2012-09-20 15:00:09 UTC
That's quite strange that 3.2.x has been removed completely, but in the same moment 3.3.x is still the last stable kernel for x86 arch :/
Comment 8 Raymond Jennings 2012-09-21 03:48:30 UTC
I agree, this is ridiculous.

Would one of the kernel maintainers for gentoo please unmask something higher than 3.3.8 and give 3.3.8 the heave ho?

This version is both broken and EOL upstream.
Comment 9 Mike Pagano gentoo-dev 2012-11-01 17:15:34 UTC
3.5.7 is stable for x86. If this is still an issue with >=3.5.7.  Please comment and I will reopen.


  22 Oct 2012; Agostino Sarubbo <ago@gentoo.org> vanilla-sources-3.5.7.ebuild:
  Stable for X86, wrt bug #432426