Hi Brian-
A fix for this one has been pushed to 'unstable'. Thanks!
sage
On Tue, 3 Mar 2009, Brian Koebbe wrote:
> So, I untarred the kernel on to a fresh ceph fs, then did a make
> menuconfig, then unmounted it so I could remount with all the -o options
> below, but I didn't get that far. When I attempted a umount, I got the
> following:
>
> Mar 3 12:45:40 node100 kernel: [ 349.582955]
> /srv/cgs/local/ceph-dev/src/ke: kill_sb ffff8101a91a1c00
> Mar 3 12:45:40 node100 kernel: [ 349.696501] BUG: Dentry
> ffff81016c41e820{i=0,n=.7770.tmp} still in use (1) [unmount of ceph ceph]
> Mar 3 12:45:40 node100 kernel: [ 349.696619] ------------[ cut here
> ]------------
> Mar 3 12:45:40 node100 kernel: [ 349.696678] kernel BUG at
> /build/buildd/linux-2.6.24/fs/dcache.c:648!
> Mar 3 12:45:40 node100 kernel: [ 349.696737] invalid opcode: 0000 [1] SMP
> Mar 3 12:45:40 node100 kernel: [ 349.696864] CPU 0
> Mar 3 12:45:40 node100 kernel: [ 349.696951] Modules linked in: ceph btrfs
> libcrc32c nfs nfsd lockd nfs_acl auth_rpcgss sunrpc exportfs autofs4
> iptable_filter ip_tables x_tables ipv6 psmouse serio_raw evdev dcdbas pcspkr
> iTCO_wdt iTCO_vendor_support e752x_edac button edac_core shpchp pci_hotplug
> ext3 jbd ext2 mbcache sr_mod sg cdrom af_packet ehci_hcd e1000 uhci_hcd
> usbcore dm_mirror dm_snapshot dm_mod thermal processor fan fbcon tileblit
> font bitblit softcursor fuse aoe sd_mod mptspi mptscsih mptbase
> scsi_transport_spi scsi_mod
> Mar 3 12:45:40 node100 kernel: [ 349.699288] Pid: 8282, comm: umount Not
> tainted 2.6.24-23-server #1
> Mar 3 12:45:40 node100 kernel: [ 349.699345] RIP: 0010:[<ffffffff802c8a84>]
> [<ffffffff802c8a84>] shrink_dcache_for_umount_subtree+0x274/0x280
> Mar 3 12:45:40 node100 kernel: [ 349.699462] RSP: 0018:ffff81018d509df8
> EFLAGS: 00010296
> Mar 3 12:45:40 node100 kernel: [ 349.699518] RAX: 0000000000000068 RBX:
> ffff81016c41e820 RCX: 0000000000000001
> Mar 3 12:45:40 node100 kernel: [ 349.700916] RDX: ffffffff8058ff68 RSI:
> 0000000000000086 RDI: ffffffff8058ff60
> Mar 3 12:45:40 node100 kernel: [ 349.700974] RBP: ffff81018e0a4750 R08:
> 0000000a49539867 R09: 0000000000000000
> Mar 3 12:45:40 node100 kernel: [ 349.701032] R10: ffff810001044c60 R11:
> 0000000000000001 R12: ffff81016c41e880
> Mar 3 12:45:40 node100 kernel: [ 349.701090] R13: 000000000000593f R14:
> ffff8101a91a1c00 R15: ffff81018d509f38
> Mar 3 12:45:40 node100 kernel: [ 349.701149] FS: 00007fcaa0e676e0(0000)
> GS:ffffffff805c5000(0000) knlGS:0000000000000000
> Mar 3 12:45:40 node100 kernel: [ 349.701220] CS: 0010 DS: 0000 ES: 0000
> CR0: 000000008005003b
> Mar 3 12:45:40 node100 kernel: [ 349.701277] CR2: 00007fcaa09410b6 CR3:
> 0000000179983000 CR4: 00000000000006e0
> Mar 3 12:45:40 node100 kernel: [ 349.701335] DR0: 0000000000000000 DR1:
> 0000000000000000 DR2: 0000000000000000
> Mar 3 12:45:40 node100 kernel: [ 349.701394] DR3: 0000000000000000 DR6:
> 00000000ffff0ff0 DR7: 0000000000000400
> Mar 3 12:45:40 node100 kernel: [ 349.701452] Process umount (pid: 8282,
> threadinfo ffff81018d508000, task ffff81017a936fe0)
> Mar 3 12:45:40 node100 kernel: [ 349.701523] Stack: ffff8101a91a1e38
> ffff8101a91a1c00 ffffffff88447a60 00007fffa8e77478
> Mar 3 12:45:40 node100 kernel: [ 349.701777] 0000000000000000
> ffffffff802c9349 ffff8101a91a1c00 ffffffff802b7369
> Mar 3 12:45:40 node100 kernel: [ 349.701992] ffff8101a91a1c00
> 0000000000000022 ffff8101a800e000 ffffffff802b7489
> Mar 3 12:45:40 node100 kernel: [ 349.702161] Call Trace:
> Mar 3 12:45:40 node100 kernel: [ 349.702266] [<ffffffff802c9349>]
> shrink_dcache_for_umount+0x29/0x50
> Mar 3 12:45:40 node100 kernel: [ 349.702326] [<ffffffff802b7369>]
> generic_shutdown_super+0x19/0x110
> Mar 3 12:45:40 node100 kernel: [ 349.702388] [<ffffffff802b7489>]
> kill_anon_super+0x9/0x40
> Mar 3 12:45:40 node100 kernel: [ 349.702460] [<ffffffff88415414>]
> :ceph:ceph_kill_sb+0x64/0xc0
> Mar 3 12:45:40 node100 kernel: [ 349.702526] [<ffffffff802b754f>]
> deactivate_super+0x6f/0x90
> Mar 3 12:45:40 node100 kernel: [ 349.702588] [<ffffffff802cdfab>]
> sys_umount+0x6b/0x2e0
> Mar 3 12:45:40 node100 kernel: [ 349.702652] [<ffffffff802b91d7>]
> sys_newstat+0x27/0x50
> Mar 3 12:45:40 node100 kernel: [ 349.702714] [<ffffffff803523c2>]
> __up_write+0x22/0x130
> Mar 3 12:45:40 node100 kernel: [ 349.702780] [<ffffffff8020c39e>]
> system_call+0x7e/0x83
> Mar 3 12:45:40 node100 kernel: [ 349.702845]
> Mar 3 12:45:40 node100 kernel: [ 349.702894]
> Mar 3 12:45:40 node100 kernel: [ 349.702895] Code: 0f 0b eb fe 0f 0b eb fe
> 0f 1f 40 00 55 48 89 fd 53 48 89 f3
> Mar 3 12:45:40 node100 kernel: [ 349.703777] RIP [<ffffffff802c8a84>]
> shrink_dcache_for_umount_subtree+0x274/0x280
> Mar 3 12:45:40 node100 kernel: [ 349.703885] RSP <ffff81018d509df8>
> Mar 3 12:45:40 node100 kernel: [ 349.703942] ---[ end trace
> fe1c2e0b1491c96f ]---
>
>
> On Tue, 2009-03-03 at 12:21 -0600, Brian Koebbe wrote:
> > On Tue, 2009-03-03 at 09:41 -0800, Sage Weil wrote:
> > >
> > > (I'm still looking at the bug you sent yesterday, by the
> > > way. The latest unstable _may_ fix it, though I'm not quite done
> > > yet.)
> > >
> >
> > yeah... still getting that bug.
> >
> > > This is with teh latest unstable? Can you do it again, but mount with
> > > '-o
> > > debug_msgr=1,debug=100,debug_console', and send along the kern.log
> > > output?
> >
> > Will do...
> >
> >
> > Brian Koebbe
> > Scientific Computing Systems Manager
> > Laboratory for Computational Genomics
> > Center for Genome Sciences
> > Washington University
> > Campus Box 8510
> > 4444 Forest Park
> > St. Louis, MO 63108
>
> Brian Koebbe
> Scientific Computing Systems Manager
> Laboratory for Computational Genomics
> Center for Genome Sciences
> Washington University
> Campus Box 8510
> 4444 Forest Park
> St. Louis, MO 63108
>
>
------------------------------------------------------------------------------
Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA
-OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise
-Strategies to boost innovation and cut costs with open source participation
-Receive a $600 discount off the registration fee with the source code: SFAD
http://p.sf.net/sfu/XcvMzF8H
_______________________________________________
Ceph-devel mailing list
Ceph-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ceph-devel
