From Bugzilla Helper: User-Agent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.7.8) Gecko/20050427 Camino/0.8.4 Description of problem: About once every month, we get a kernel oops in kswapd0. All stacks show the problem being somewhere in shrinking the dcache. Two console logs from 2 different events included as attachments. Version-Release number of selected component (if applicable): kernel-smp-2.6.9-22.0.1.EL How reproducible: Sometimes Steps to Reproduce: 1. Let RHEL4 fileserver run for over a month 2. 3. Additional info:
Created attachment 122978 [details] console log from january crash
Created attachment 122979 [details] console log from december crash
There has been some discussion of this issue on lkml, please see: http://marc.theaimsgroup.com/?l=linux-kernel&m=113745093326567&w=2
I seem to also just have had a recent rash of these. Has there been any progress at all? Is there anything in the updated kernel in U3 that is expected to address this issue?
hi Bill, Yes we are aware that this is a hot issue and we are actively persuing a fix. No, U3 does not fix this issue. I'm marking this as a duplicate of bug #173843. thanks, -Jason
*** This bug has been marked as a duplicate of 173843 ***
The bug this has been marked a duplicate of, is another and separate issue; there has been no unmounting in the case I've been seeing (matching other reports here), so shrink_dcache_parent() is not being called from kill_super() on unmount. Since this is with post-U4 2.6.9-42.0.3, it is after the "[fix] committed in stream U4 build 35.1", above, and I am not seeing the 'busy-inodes on unmount' message either, and the fix for the bug this is marked as a duplicate of is present, yet I'm still seeing the crash. I'm hitting this kswapd->prune_dcache bug quite frequently - around once a week, on a few machines under constant load and high memory pressure. Configuration is stock RHEL4 U4 + latest errata 2.6.9-42.0.3, dual-SMP, x86-64, 4GB memory Crash signature is: __down_read_trylock+18 prune_dcache+568 shrink_dcache_memory+20 shrink_slab+188 balance_pgdat+538 kswapd+252 autoremove_wake_function+0 autoremove_wake_function+0 child_rip+8 kswapd+0 child_rip+0 RIP: _spin_lock_irqsave+40 Can this be reopened please, to preserve the reports from other users? Let me know if getting a crash-dump and making it available to someone would help.
Bug 224134 is a duplicate of this: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=224134
*** Bug 224134 has been marked as a duplicate of this bug. ***
I too am seeing this on machines running the 42.0.2 kernel. Any progress on this in terms of patches to test?
It would be a good data-point to establish if the newer kernels suffer the same issue: http://people.redhat.com/~jbaron/rhel4/RPMS.kernel/
This request was evaluated by Red Hat Product Management for inclusion in a Red Hat Enterprise Linux maintenance release. Product Management has requested further review of this request by Red Hat Engineering, for potential inclusion in a Red Hat Enterprise Linux Update release for currently deployed products. This request is not yet committed for inclusion in an Update release.
*** Bug 177122 has been marked as a duplicate of this bug. ***
This request was previously evaluated by Red Hat Product Management for inclusion in the current Red Hat Enterprise Linux release, but Red Hat was unable to resolve it in time. This request will be reviewed for a future Red Hat Enterprise Linux release.
Would it be possible to get a test patch posted here before release?
Can someone who can reproduce this problem see if it can still happens with RHEL4-U6? If it does crash, please get a dump, I hav never been able to reproduce this problem internally. Larry Woodman