Hello, I was copying a 60 GB file onto btrfs, and at the same time tried to stop/restart Squid, which has its cache on the same partition. Squid locked up until the copying was finished. This filesystem was resized by about +100% just before, if this could matter. The kernel is 3.6.0-rc5. From dmesg: [27175.299657] INFO: task squid3:15475 blocked for more than 120 seconds. [27175.299665] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [27175.299670] squid3 D ffff88041fc13dc0 0 15475 1 0x00000004 [27175.299673] ffff880189735b38 0000000000000082 ffff88029588c470 ffff880189735fd8 [27175.299677] ffff880189735fd8 ffff880189735fd8 ffff8803c6ef5b40 ffff88029588c470 [27175.299679] ffff880189735b48 ffff880407a07b00 ffff88037ca3b9e8 0000000000000000 [27175.299681] Call Trace: [27175.299688] [<ffffffff8166afc9>] schedule+0x29/0x70 [27175.299715] [<ffffffffa00e875d>] wait_current_trans.isra.28+0x9d/0x100 [btrfs] [27175.299718] [<ffffffff810794b0>] ? finish_wait+0x80/0x80 [27175.299729] [<ffffffffa00e9e48>] start_transaction+0x148/0x3d0 [btrfs] [27175.299737] [<ffffffffa00c3efa>] ? btrfs_free_path+0x2a/0x40 [btrfs] [27175.299747] [<ffffffffa00ea3a3>] btrfs_start_transaction+0x13/0x20 [btrfs] [27175.299757] [<ffffffffa00f6a97>] btrfs_create+0x47/0x210 [btrfs] [27175.299760] [<ffffffff812a9a5c>] ? security_inode_permission+0x1c/0x30 [27175.299763] [<ffffffff8118683c>] ? __inode_permission+0x5c/0xd0 [27175.299765] [<ffffffff811894f5>] vfs_create+0xb5/0x110 [27175.299767] [<ffffffff8118a285>] do_last+0xd35/0xe30 [27175.299769] [<ffffffff8116d7fb>] ? kmem_cache_alloc_trace+0x11b/0x120 [27175.299772] [<ffffffff8118a433>] path_openat+0xb3/0x4b0 [27175.299774] [<ffffffff81185353>] ? putname+0x33/0x50 [27175.299775] [<ffffffff8118a9eb>] ? user_path_at_empty+0x6b/0xa0 [27175.299777] [<ffffffff8166bd9e>] ? _raw_spin_lock+0xe/0x20 [27175.299779] [<ffffffff8117f7d7>] ? inode_get_bytes+0x47/0x60 [27175.299781] [<ffffffff8118aa81>] do_filp_open+0x41/0xa0 [27175.299783] [<ffffffff81197306>] ? alloc_fd+0xc6/0x110 [27175.299786] [<ffffffff81179e46>] do_sys_open+0xf6/0x1e0 [27175.299788] [<ffffffff81179f51>] sys_open+0x21/0x30 [27175.299791] [<ffffffff81673e6d>] system_call_fastpath+0x1a/0x1f [27295.008758] INFO: task squid3:15475 blocked for more than 120 seconds. [27295.008766] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [27295.008770] squid3 D ffff88041fc13dc0 0 15475 1 0x00000004 [27295.008774] ffff880189735b38 0000000000000082 ffff88029588c470 ffff880189735fd8 [27295.008777] ffff880189735fd8 ffff880189735fd8 ffff8803c6ef5b40 ffff88029588c470 [27295.008779] ffff880189735b48 ffff880407a07b00 ffff88037ca3b9e8 0000000000000000 [27295.008781] Call Trace: [27295.008787] [<ffffffff8166afc9>] schedule+0x29/0x70 [27295.008809] [<ffffffffa00e875d>] wait_current_trans.isra.28+0x9d/0x100 [btrfs] [27295.008813] [<ffffffff810794b0>] ? finish_wait+0x80/0x80 [27295.008823] [<ffffffffa00e9e48>] start_transaction+0x148/0x3d0 [btrfs] [27295.008831] [<ffffffffa00c3efa>] ? btrfs_free_path+0x2a/0x40 [btrfs] [27295.008841] [<ffffffffa00ea3a3>] btrfs_start_transaction+0x13/0x20 [btrfs] [27295.008851] [<ffffffffa00f6a97>] btrfs_create+0x47/0x210 [btrfs] [27295.008853] [<ffffffff812a9a5c>] ? security_inode_permission+0x1c/0x30 [27295.008856] [<ffffffff8118683c>] ? __inode_permission+0x5c/0xd0 [27295.008858] [<ffffffff811894f5>] vfs_create+0xb5/0x110 [27295.008860] [<ffffffff8118a285>] do_last+0xd35/0xe30 [27295.008863] [<ffffffff8116d7fb>] ? kmem_cache_alloc_trace+0x11b/0x120 [27295.008865] [<ffffffff8118a433>] path_openat+0xb3/0x4b0 [27295.008867] [<ffffffff81185353>] ? putname+0x33/0x50 [27295.008868] [<ffffffff8118a9eb>] ? user_path_at_empty+0x6b/0xa0 [27295.008870] [<ffffffff8166bd9e>] ? _raw_spin_lock+0xe/0x20 [27295.008872] [<ffffffff8117f7d7>] ? inode_get_bytes+0x47/0x60 [27295.008874] [<ffffffff8118aa81>] do_filp_open+0x41/0xa0 [27295.008876] [<ffffffff81197306>] ? alloc_fd+0xc6/0x110 [27295.008878] [<ffffffff81179e46>] do_sys_open+0xf6/0x1e0 [27295.008880] [<ffffffff81179f51>] sys_open+0x21/0x30 [27295.008883] [<ffffffff81673e6d>] system_call_fastpath+0x1a/0x1f -- With respect, Roman ~~~~~~~~~~~~~~~~~~~~~~~~~~~ "Stallman had a printer, with code he could not see. So he began to tinker, and set the software free."
Josef Bacik
2012-Sep-13 12:56 UTC
Re: "Task blocked for more than 120 seconds" on file copy
On Wed, Sep 12, 2012 at 03:16:23PM -0600, Roman Mamedov wrote:> Hello, > > I was copying a 60 GB file onto btrfs, and at the same time tried to stop/restart Squid, > which has its cache on the same partition. Squid locked up until the copying was finished. > This filesystem was resized by about +100% just before, if this could matter. > The kernel is 3.6.0-rc5. From dmesg: >When you get these sort of lockups we need sysrq+w to get all of the blocked tasks, not just hte ones that the hung task timer catches. Thanks, Josef -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html