Hide Forgot
Description of problem: The machine feels slow at install time already as it takes very long for the standard installation to finish. Also general system performances is lacking. A kernel build with make -j32 often stalls completely! Also while doing the kernel build I observed: INFO: task jbd2/dm-0-8:1899 blocked for more than 120 seconds. "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. jbd2/dm-0-8 D ffff88027fe75000 0 1899 2 0x00000000 ffff880273653d10 0000000000000046 ffff880273653cd8 ffff880273653cd4 ffff8801e243ac90 ffff88027fe75000 ffff880287416980 0000000100204367 ffff8802736a5ad8 ffff880273653fd8 0000000000010518 ffff8802736a5ad8 Call Trace: [<ffffffffa00618f0>] jbd2_journal_commit_transaction+0x1c0/0x14e0 [jbd2] [<ffffffff8107d5ac>] ? lock_timer_base+0x3c/0x70 [<ffffffff81091ca0>] ? autoremove_wake_function+0x0/0x40 [<ffffffffa00680b8>] kjournald2+0xb8/0x220 [jbd2] [<ffffffff81091ca0>] ? autoremove_wake_function+0x0/0x40 [<ffffffffa0068000>] ? kjournald2+0x0/0x220 [jbd2] [<ffffffff81091936>] kthread+0x96/0xa0 [<ffffffff810141ca>] child_rip+0xa/0x20 [<ffffffff810918a0>] ? kthread+0x0/0xa0 [<ffffffff810141c0>] ? child_rip+0x0/0x20 INFO: task flush-253:0:4600 blocked for more than 120 seconds. "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. flush-253:0 D ffff88027fe74c00 0 4600 2 0x00000080 ffff880270dc7420 0000000000000046 0000000000000000 ffffffffa000471c ffff880270dc73a0 ffffffff814c9cf6 ffffffff8175fe80 0000000100204088 ffff880270da0678 ffff880270dc7fd8 0000000000010518 ffff880270da0678 Call Trace: [<ffffffffa000471c>] ? dm_table_unplug_all+0x5c/0xd0 [dm_mod] [<ffffffff814c9cf6>] ? down_write+0x16/0x40 [<ffffffff8119dcf0>] ? sync_buffer+0x0/0x50 [<ffffffff814c8a23>] io_schedule+0x73/0xc0 [<ffffffff8119dd30>] sync_buffer+0x40/0x50 [<ffffffff814c929f>] __wait_on_bit+0x5f/0x90 [<ffffffff8119dcf0>] ? sync_buffer+0x0/0x50 [<ffffffff814c9348>] out_of_line_wait_on_bit+0x78/0x90 [<ffffffff81091ce0>] ? wake_bit_function+0x0/0x50 [<ffffffff8119dce6>] __wait_on_buffer+0x26/0x30 [<ffffffffa00b927d>] ext4_mb_init_cache+0x34d/0x970 [ext4] [<ffffffff8110c3f9>] ? add_to_page_cache_locked+0xc9/0x140 [<ffffffffa00b9974>] ext4_mb_init_group+0xd4/0x260 [ext4] [<ffffffffa00b9e8d>] ext4_mb_load_buddy+0x38d/0x3d0 [ext4] [<ffffffffa00bcc86>] ext4_mb_regular_allocator+0x1e6/0x550 [ext4] [<ffffffffa00037c7>] ? dm_request+0x1e7/0x230 [dm_mod] [<ffffffffa00bd37d>] ext4_mb_new_blocks+0x38d/0x560 [ext4] [<ffffffffa00b12be>] ? ext4_ext_find_extent+0x2be/0x320 [ext4] [<ffffffffa00b4230>] ext4_ext_get_blocks+0xfa0/0x1800 [ext4] [<ffffffff8124117f>] ? submit_bio+0x8f/0x120 [<ffffffffa0092610>] ext4_get_blocks+0x1f0/0x2a0 [ext4] [<ffffffffa00934ec>] mpage_da_map_blocks+0xac/0x430 [ext4] [<ffffffffa009402e>] ext4_da_writepages+0x2be/0x600 [ext4] [<ffffffff81120951>] do_writepages+0x21/0x40 [<ffffffff81195cdd>] writeback_single_inode+0xdd/0x2c0 [<ffffffff811960de>] writeback_sb_inodes+0xce/0x180 [<ffffffff81196233>] writeback_inodes_wb+0xa3/0x1a0 [<ffffffff811965cb>] wb_writeback+0x29b/0x3f0 [<ffffffff814c8286>] ? thread_return+0x4e/0x778 [<ffffffff811968b9>] wb_do_writeback+0x199/0x240 [<ffffffff811969c3>] bdi_writeback_task+0x63/0x1b0 [<ffffffff81091b67>] ? bit_waitqueue+0x17/0xd0 [<ffffffff8112ecd0>] ? bdi_start_fn+0x0/0x100 [<ffffffff8112ed56>] bdi_start_fn+0x86/0x100 [<ffffffff8112ecd0>] ? bdi_start_fn+0x0/0x100 [<ffffffff81091936>] kthread+0x96/0xa0 [<ffffffff810141ca>] child_rip+0xa/0x20 [<ffffffff810918a0>] ? kthread+0x0/0xa0 [<ffffffff810141c0>] ? child_rip+0x0/0x20 System was unreachable after that. Version-Release number of selected component (if applicable): 2.6.32-71.el6.x86_64 How reproducible: always Steps to Reproduce: 1. build kernel with make -j32 2. 3.
I installed RHEL6.0 (GA) on the system, checked out a kernel and watched a -j32 compile. During the compile there are several "long hangs". I cleaned the tree, and compiled again and did a echo 30 > /proc/sys/kernel/hung_task_timeout_secs After doing that I see one or two messages similar to the ones above which seem to imply an issue with ext4 and the journaling code (AFAICT). esandeen -- this *could* be a weird storage issue, but for now I'm assigning it to you. Please note that this system is supposed to be converted into a Westmere system in the next few weeks, and we have NOT seen this issue on other Nehalem based boxes. P.
I think that this is more of a storage level issue and ext4 is just the thing waiting on something else. It would be good to retry this on the current crop of 6.1 kernels....
Since RHEL 6.2 External Beta has begun, and this bug remains unresolved, it has been rejected as it is not proposed as exception or blocker. Red Hat invites you to ask your support representative to propose this request, if appropriate and relevant, in the next release of Red Hat Enterprise Linux.