Source: linux-2.6 Version: 2.6.32-39 Severity: normal Hello, Since the upgrade from lenny to squeeze I am getting in several servers random kernel lockups related to the XFS filesystem. The errors looks the same on all servers. All servers are running Debian kernel 2.6.32-39 Server 1 running Debian/Squeeze with linux-image-2.6.32-5-686/2.6.32-39 ----------------------------------------------------------------------------------------- [539160.520045] INFO: task xfssyncd:985 blocked for more than 120 seconds. [539160.520077] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [539160.520123] xfssyncd D f9197020 0 985 2 0x00000000 [539160.520129] c340fb40 00000046 f6c32200 f9197020 00000000 c141d100 c141d100 c14186ac [539160.520137] c340fcfc c2108100 00000001 0000a0bf 00000000 00000000 00000010 f6a7f280 [539160.520145] c21036ac c340fcfc 0806c251 f6a38b90 00000082 00000000 00000000 00000000 [539160.520154] Call Trace: [539160.520179] [<c126d9b9>] ? schedule_timeout+0x20/0xb0 [539160.520184] [<c1138433>] ? __lookup_tag+0x8e/0xee [539160.520189] [<c11385b1>] ? radix_tree_gang_lookup_tag+0x8a/0xd5 [539160.520193] [<c126d8c2>] ? wait_for_common+0xa4/0x100 [539160.520199] [<c102d500>] ? default_wake_function+0x0/0x8 [539160.520225] [<f8324433>] ? xfs_reclaim_inode+0x85/0xc8 [xfs] [539160.520259] [<f8324c26>] ? xfs_inode_ag_walk+0x73/0xbd [xfs] [539160.520280] [<f83243ae>] ? xfs_reclaim_inode+0x0/0xc8 [xfs] [539160.520302] [<f8324cc8>] ? xfs_inode_ag_iterator+0x58/0x8f [xfs] [539160.520324] [<f83243ae>] ? xfs_reclaim_inode+0x0/0xc8 [xfs] [539160.520345] [<f8324de4>] ? xfs_reclaim_inodes+0x12/0x16 [xfs] [539160.520367] [<f8324e0b>] ? xfs_sync_worker+0x23/0x5c [xfs] [539160.520388] [<f83246d7>] ? xfssyncd+0x134/0x17d [xfs] [539160.520410] [<f83245a3>] ? xfssyncd+0x0/0x17d [xfs] [539160.520415] [<c1043dfc>] ? kthread+0x61/0x66 [539160.520418] [<c1043d9b>] ? kthread+0x0/0x66 [539160.520423] [<c1003d47>] ? kernel_thread_helper+0x7/0x10 Server 2 running Debian/Squeeze with linux-image-2.6.32-5-vserver-amd64/2.6.32-39 ----------------------------------------------------------------------------------------- [864600.444518] INFO: task xfssyncd:933 blocked for more than 120 seconds. [864600.444563] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [864600.444608] xfssyncd D 0000000000000000 0 933 2 0x00000000 [864600.444612] ffff88022f06b980 0000000000000046 0000000000000000 ffff880050925e40 [864600.444617] 0000000000002000 ffffffff8118af2b 000000000000f9e0 ffff88022c607fd8 [864600.444620] 0000000000015780 0000000000015780 ffff88022f195640 ffff88022f195938 [864600.444625] Call Trace: [864600.444632] [<ffffffff8118af2b>] ? generic_make_request+0x299/0x2f9 [864600.444637] [<ffffffff8130a2fd>] ? schedule_timeout+0x2e/0xdd [864600.444659] [<ffffffffa01cea3d>] ? _xfs_buf_ioapply+0x204/0x22f [xfs] [864600.444663] [<ffffffff8119e9c8>] ? __lookup_tag+0xad/0x11b [864600.444666] [<ffffffff8130a1b6>] ? wait_for_common+0xde/0x15b [864600.444671] [<ffffffff8104a705>] ? default_wake_function+0x0/0x9 [864600.444685] [<ffffffffa01d6058>] ? xfs_reclaim_inode+0x95/0xe0 [xfs] [864600.444698] [<ffffffffa01d6975>] ? xfs_inode_ag_walk+0x92/0xef [xfs] [864600.444711] [<ffffffffa01d5fc3>] ? xfs_reclaim_inode+0x0/0xe0 [xfs] [864600.444724] [<ffffffffa01d6a43>] ? xfs_inode_ag_iterator+0x71/0xb2 [xfs] [864600.444737] [<ffffffffa01d5fc3>] ? xfs_reclaim_inode+0x0/0xe0 [xfs] [864600.444751] [<ffffffffa01d6bd0>] ? xfs_sync_worker+0x26/0x5f [xfs] [864600.444764] [<ffffffffa01d6338>] ? xfssyncd+0x150/0x1bb [xfs] [864600.444777] [<ffffffffa01d61e8>] ? xfssyncd+0x0/0x1bb [xfs] [864600.444780] [<ffffffff81065bf5>] ? kthread+0x79/0x81 [864600.444784] [<ffffffff81011baa>] ? child_rip+0xa/0x20 [864600.444788] [<ffffffff8118bd3d>] ? generic_unplug_device+0x0/0x34 [864600.444791] [<ffffffff81065b7c>] ? kthread+0x0/0x81 [864600.444794] [<ffffffff81011ba0>] ? child_rip+0x0/0x20 Server 3 running Debian/Squeeze with linux-image-2.6.32-5-vserver-amd64/2.6.32-39 ----------------------------------------------------------------------------------------- [ 5520.764046] INFO: task xfssyncd:1376 blocked for more than 120 seconds. [ 5520.764080] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [ 5520.764126] xfssyncd D 0000000000000000 0 1376 2 0x00000000 [ 5520.764132] ffff88007fb5d640 0000000000000046 0000000000000000 ffff88001f342540 [ 5520.764138] 00000000000000ff ffffffff8118af2b 000000000000f9e0 ffff88007e291fd8 [ 5520.764143] 0000000000015780 0000000000015780 ffff880037baab20 ffff880037baae18 [ 5520.764148] Call Trace: [ 5520.764159] [<ffffffff8118af2b>] ? generic_make_request+0x299/0x2f9 [ 5520.764166] [<ffffffff8130a2fd>] ? schedule_timeout+0x2e/0xdd [ 5520.764200] [<ffffffffa0268a3d>] ? _xfs_buf_ioapply+0x204/0x22f [xfs] [ 5520.764205] [<ffffffff8119e9c8>] ? __lookup_tag+0xad/0x11b [ 5520.764210] [<ffffffff8130a1b6>] ? wait_for_common+0xde/0x15b [ 5520.764216] [<ffffffff8104a705>] ? default_wake_function+0x0/0x9 [ 5520.764234] [<ffffffffa0270058>] ? xfs_reclaim_inode+0x95/0xe0 [xfs] [ 5520.764251] [<ffffffffa0270975>] ? xfs_inode_ag_walk+0x92/0xef [xfs] [ 5520.764268] [<ffffffffa026ffc3>] ? xfs_reclaim_inode+0x0/0xe0 [xfs] [ 5520.764285] [<ffffffffa0270a43>] ? xfs_inode_ag_iterator+0x71/0xb2 [xfs] [ 5520.764301] [<ffffffffa026ffc3>] ? xfs_reclaim_inode+0x0/0xe0 [xfs] [ 5520.764318] [<ffffffffa0270bd0>] ? xfs_sync_worker+0x26/0x5f [xfs] [ 5520.764335] [<ffffffffa0270338>] ? xfssyncd+0x150/0x1bb [xfs] [ 5520.764351] [<ffffffffa02701e8>] ? xfssyncd+0x0/0x1bb [xfs] [ 5520.764356] [<ffffffff81065bf5>] ? kthread+0x79/0x81 [ 5520.764362] [<ffffffff81011baa>] ? child_rip+0xa/0x20 [ 5520.764366] [<ffffffff81065b7c>] ? kthread+0x0/0x81 [ 5520.764369] [<ffffffff81011ba0>] ? child_rip+0x0/0x20 Server 4 running Debian/Squeeze with linux-image-2.6.32-5-vserver-amd64/2.6.32-39 ----------------------------------------------------------------------------------------- [279117.747342] INFO: task xfssyncd:1985 blocked for more than 120 seconds. [279117.747401] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [279117.747463] xfssyncd D 0000000000000000 0 1985 2 0x00000000 [279117.747466] ffff88023f06c0b0 0000000000000046 0000000000800012 ffffffff8118a673 [279117.747468] ffff88023f0a6c00 000000000000f9e0 ffff88023cd93fd8 0000000000015780 [279117.747470] 0000000000015780 ffff88023c1547e0 ffff88023c154ad8 0000000298bc2318 [279117.747473] Call Trace: [279117.747479] [<ffffffff8118a673>] ? generic_make_request+0x299/0x2f9 [279117.747483] [<ffffffff810114ce>] ? common_interrupt+0xe/0x13 [279117.747501] [<ffffffffa01b45f4>] ? xfs_ioend_wait+0x84/0x9c [xfs] [279117.747503] [<ffffffff8106599e>] ? autoremove_wake_function+0x0/0x2e [279117.747513] [<ffffffffa0199460>] ? xfs_ilock_nowait+0x32/0x92 [xfs] [279117.747520] [<ffffffffa01bce36>] ? xfs_sync_inode_data+0x91/0xa8 [xfs] [279117.747528] [<ffffffffa01bcf61>] ? xfs_inode_ag_walk+0x92/0xef [xfs] [279117.747535] [<ffffffffa01bcda5>] ? xfs_sync_inode_data+0x0/0xa8 [xfs] [279117.747542] [<ffffffffa01bd02f>] ? xfs_inode_ag_iterator+0x71/0xb2 [xfs] [279117.747549] [<ffffffffa01bcda5>] ? xfs_sync_inode_data+0x0/0xa8 [xfs] [279117.747557] [<ffffffffa01bd2fe>] ? xfs_sync_data+0x20/0x42 [xfs] [279117.747564] [<ffffffffa01bd344>] ? xfs_flush_inodes_work+0x24/0x31 [xfs] [279117.747571] [<ffffffffa01bc924>] ? xfssyncd+0x150/0x1bb [xfs] [279117.747578] [<ffffffffa01bc7d4>] ? xfssyncd+0x0/0x1bb [xfs] [279117.747580] [<ffffffff810656d1>] ? kthread+0x79/0x81 [279117.747582] [<ffffffff81011baa>] ? child_rip+0xa/0x20 [279117.747584] [<ffffffff81065658>] ? kthread+0x0/0x81 [279117.747585] [<ffffffff81011ba0>] ? child_rip+0x0/0x20 A bit of googling [1] suggests that perhaps the commit 17b3847 [2] could have fixed this Also I can say that I have switched some weeks ago one of this servers to a vainilla/vserver 3.1 Kernel and since then I didn't saw this lockup anymore. ----------------------------------------------------------------------------------------- [1] http://comments.gmane.org/gmane.comp.file-systems.xfs.general/41907 [2] https://git.kernel.org/linus/17b3847 -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Carlos Alberto Lopez Perez http://neutrino.es Igalia - Free Software Engineering http://www.igalia.com ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Attachment:
signature.asc
Description: OpenPGP digital signature