On Wed, Sep 24, 2014 at 12:20 PM, Micha Krause <micha at krausam.de> wrote: > Hi, > >> So does it actually crash or it's just the blocked I/Os? If it doesn't >> >> crash, you should be able to get everything off dmesg. > > > it's blocked I/Os, I just wrote another mail to the list, with more dmesg > Output > from a Centos machine. > > >>> dmesg: >>> >>> [18102.981064] INFO: task nfsd:2769 blocked for more than 120 seconds. >>> [18102.981112] Not tainted 3.14-0.bpo.1-amd64 #1 >>> [18102.981150] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" >>> disables >>> this message. >>> [18102.981216] nfsd D ffff88003fc14340 0 2769 2 >>> 0x00000000 >>> [18102.981218] ffff88003bac6e20 0000000000000046 0000000000000000 >>> ffff88003d47ada0 >>> [18102.981219] 0000000000014340 ffff88003ce31fd8 0000000000014340 >>> ffff88003bac6e20 >>> [18102.981221] ffff88003ce31728 ffff8800029539f0 7fffffffffffffff >>> 7fffffffffffffff >>> [18102.981223] Call Trace: >>> [18102.981225] [<ffffffff814eedbd>] ? schedule_timeout+0x1ed/0x250 >>> [18102.981231] [<ffffffffa04b0f92>] ? _xfs_buf_find+0xd2/0x280 [xfs] >>> [18102.981234] [<ffffffff8117fc2c>] ? kmem_cache_alloc+0x1bc/0x1f0 >>> [18102.981236] [<ffffffff814f193c>] ? __down_common+0x97/0xea >>> [18102.981241] [<ffffffffa04b0faa>] ? _xfs_buf_find+0xea/0x280 [xfs] >>> [18102.981243] [<ffffffff810aa697>] ? down+0x37/0x40 >>> [18102.981247] [<ffffffffa04b0e02>] ? xfs_buf_lock+0x32/0xf0 [xfs] >>> [18102.981252] [<ffffffffa04b0faa>] ? _xfs_buf_find+0xea/0x280 [xfs] >>> [18102.981257] [<ffffffffa04b1215>] ? xfs_buf_get_map+0x35/0x1a0 [xfs] >>> [18102.981263] [<ffffffffa04b2153>] ? xfs_buf_read_map+0x33/0x130 [xfs] >>> [18102.981269] [<ffffffffa05161da>] ? xfs_trans_read_buf_map+0x34a/0x4f0 >>> [xfs] >>> [18102.981275] [<ffffffffa05036f9>] ? xfs_imap_to_bp+0x69/0xf0 [xfs] >>> [18102.981281] [<ffffffffa0503bcd>] ? xfs_iread+0x7d/0x3f0 [xfs] >>> [18102.981284] [<ffffffff810e8939>] ? make_kgid+0x9/0x10 >>> [18102.981286] [<ffffffff811b148e>] ? inode_init_always+0x10e/0x1d0 >>> [18102.981292] [<ffffffffa04ba11a>] ? xfs_iget+0x2ba/0x810 [xfs] >>> [18102.981298] [<ffffffffa04fd9a6>] ? xfs_ialloc+0xe6/0x740 [xfs] >>> [18102.981305] [<ffffffffa04ca1ee>] ? kmem_zone_alloc+0x6e/0xf0 [xfs] >>> [18102.981311] [<ffffffffa04fe083>] ? xfs_dir_ialloc+0x83/0x300 [xfs] >>> [18102.981317] [<ffffffffa04c8e43>] ? xfs_trans_reserve+0x213/0x220 >>> [xfs] >>> [18102.981323] [<ffffffffa04fe87e>] ? xfs_create+0x4fe/0x720 [xfs] >>> [18102.981329] [<ffffffffa04bfd02>] ? xfs_vn_mknod+0xd2/0x200 [xfs] >>> [18102.981331] [<ffffffff811a6b54>] ? vfs_create+0xe4/0x160 >>> [18102.981335] [<ffffffffa0400d9e>] ? do_nfsd_create+0x53e/0x610 [nfsd] >>> [18102.981339] [<ffffffffa0407f4d>] ? nfsd3_proc_create+0x16d/0x250 >>> [nfsd] >>> [18102.981342] [<ffffffffa03f9d74>] ? nfsd_dispatch+0xe4/0x230 [nfsd] >>> [18102.981347] [<ffffffffa035dd64>] ? svc_process_common+0x354/0x690 >>> [sunrpc] >>> [18102.981349] [<ffffffff81096ab0>] ? try_to_wake_up+0x280/0x280 >>> [18102.981353] [<ffffffffa035e3fb>] ? svc_process+0x10b/0x160 [sunrpc] >>> [18102.981359] [<ffffffffa03f96d7>] ? nfsd+0xb7/0x130 [nfsd] >>> [18102.981363] [<ffffffffa03f9620>] ? nfsd_destroy+0x70/0x70 [nfsd] >>> [18102.981365] [<ffffffff81086d6c>] ? kthread+0xbc/0xe0 >>> [18102.981367] [<ffffffff81086cb0>] ? flush_kthread_worker+0xa0/0xa0 >>> [18102.981369] [<ffffffff814faecc>] ? ret_from_fork+0x7c/0xb0 >>> [18102.981371] [<ffffffff81086cb0>] ? flush_kthread_worker+0xa0/0xa0 >> >> >> Is that the only hung task in dmesg? > > > I think it was, could be that this Message was repeated a few times. Like I mentioned in my other reply, I'd be very interested in any similar messages on kernel other than 3.15.*, 3.16.1 and 3.16.2. One hung task stack trace is usually not enough to diagnose this sort of problems.
 |