Running kernel 2.6.28-4 on a hp proliant with dual-core xeon running centos 5.2 64bits with custom kernel, getting this: Call Trace: [<ffffffff802f8d6f>] ext4_da_writepages+0xcc/0x333 [<ffffffff8026eb67>] do_writepages+0x20/0x2d [<ffffffff802aaf73>] __writeback_single_inode+0x147/0x285 [<ffffffff80231948>] dequeue_task_fair+0x96/0x10d [<ffffffff8022a150>] __dequeue_entity+0x1c/0x32 [<ffffffff802ab3f2>] generic_sync_sb_inodes+0x1bf/0x302 [<ffffffff802ab82d>] writeback_inodes+0x56/0x9d [<ffffffff8026f469>] wb_kupdate+0x9e/0x10d [<ffffffff8026f812>] pdflush+0x0/0x1f8 [<ffffffff8026f95e>] pdflush+0x14c/0x1f8 [<ffffffff8026f3cb>] wb_kupdate+0x0/0x10d [<ffffffff80247232>] kthread+0x47/0x75 [<ffffffff802309b4>] schedule_tail+0x28/0x60 [<ffffffff8020cbe9>] child_rip+0xa/0x11 [<ffffffff802471eb>] kthread+0x0/0x75 [<ffffffff8020cbdf>] child_rip+0x0/0x11 ext4_da_writepages: jbd2_start: 1024 pages, ino 21705; err -30 Pid: 272, comm: pdflush Not tainted 2.6.28.4-onda #1 Call Trace: [<ffffffff802f8d6f>] ext4_da_writepages+0xcc/0x333 [<ffffffff8026eb67>] do_writepages+0x20/0x2d [<ffffffff802aaf73>] __writeback_single_inode+0x147/0x285 [<ffffffff80231948>] dequeue_task_fair+0x96/0x10d [<ffffffff8022a150>] __dequeue_entity+0x1c/0x32 [<ffffffff802ab3f2>] generic_sync_sb_inodes+0x1bf/0x302 [<ffffffff802ab82d>] writeback_inodes+0x56/0x9d [<ffffffff8026f469>] wb_kupdate+0x9e/0x10d [<ffffffff8026f812>] pdflush+0x0/0x1f8 [<ffffffff8026f95e>] pdflush+0x14c/0x1f8 [<ffffffff8026f3cb>] wb_kupdate+0x0/0x10d [<ffffffff80247232>] kthread+0x47/0x75 [<ffffffff802309b4>] schedule_tail+0x28/0x60 [<ffffffff8020cbe9>] child_rip+0xa/0x11 [<ffffffff802471eb>] kthread+0x0/0x75 [<ffffffff8020cbdf>] child_rip+0x0/0x11 ext4_da_writepages: jbd2_start: 1024 pages, ino 21705; err -30 Pid: 272, comm: pdflush Not tainted 2.6.28.4-onda #1 Call Trace: [<ffffffff802f8d6f>] ext4_da_writepages+0xcc/0x333 [<ffffffff8026eb67>] do_writepages+0x20/0x2d [<ffffffff802aaf73>] __writeback_single_inode+0x147/0x285 [<ffffffff80231948>] dequeue_task_fair+0x96/0x10d [<ffffffff802ab3f2>] generic_sync_sb_inodes+0x1bf/0x302 [<ffffffff802ab82d>] writeback_inodes+0x56/0x9d [<ffffffff8026f469>] wb_kupdate+0x9e/0x10d [<ffffffff8026f812>] pdflush+0x0/0x1f8 [<ffffffff8026f95e>] pdflush+0x14c/0x1f8 [<ffffffff8026f3cb>] wb_kupdate+0x0/0x10d [<ffffffff80247232>] kthread+0x47/0x75 [<ffffffff802309b4>] schedule_tail+0x28/0x60 [<ffffffff8020cbe9>] child_rip+0xa/0x11 [<ffffffff802471eb>] kthread+0x0/0x75 [<ffffffff8020cbdf>] child_rip+0x0/0x11 ext4_da_writepages: jbd2_start: 1024 pages, ino 21705; err -30 Pid: 272, comm: pdflush Not tainted 2.6.28.4-onda #1 Call Trace: [<ffffffff802f8d6f>] ext4_da_writepages+0xcc/0x333 [<ffffffff8026eb67>] do_writepages+0x20/0x2d [<ffffffff802aaf73>] __writeback_single_inode+0x147/0x285 [<ffffffff80231948>] dequeue_task_fair+0x96/0x10d [<ffffffff802ab3f2>] generic_sync_sb_inodes+0x1bf/0x302 [<ffffffff802ab82d>] writeback_inodes+0x56/0x9d [<ffffffff8026f469>] wb_kupdate+0x9e/0x10d [<ffffffff8026f812>] pdflush+0x0/0x1f8 [<ffffffff8026f95e>] pdflush+0x14c/0x1f8 [<ffffffff8026f3cb>] wb_kupdate+0x0/0x10d [<ffffffff80247232>] kthread+0x47/0x75 [<ffffffff802309b4>] schedule_tail+0x28/0x60 [<ffffffff8020cbe9>] child_rip+0xa/0x11 [<ffffffff802471eb>] kthread+0x0/0x75 [<ffffffff8020cbdf>] child_rip+0x0/0x11 ext4_da_writepages: jbd2_start: 1024 pages, ino 21705; err -30 Pid: 272, comm: pdflush Not tainted 2.6.28.4-onda #1 Call Trace: [<ffffffff802f8d6f>] ext4_da_writepages+0xcc/0x333 [<ffffffff8026eb67>] do_writepages+0x20/0x2d [<ffffffff802aaf73>] __writeback_single_inode+0x147/0x285 [<ffffffff80231948>] dequeue_task_fair+0x96/0x10d [<ffffffff802ab3f2>] generic_sync_sb_inodes+0x1bf/0x302 [<ffffffff802ab82d>] writeback_inodes+0x56/0x9d [<ffffffff8026f469>] wb_kupdate+0x9e/0x10d [<ffffffff8026f812>] pdflush+0x0/0x1f8 [<ffffffff8026f95e>] pdflush+0x14c/0x1f8 [<ffffffff8026f3cb>] wb_kupdate+0x0/0x10d [<ffffffff80247232>] kthread+0x47/0x75 [<ffffffff802309b4>] schedule_tail+0x28/0x60 [<ffffffff8020cbe9>] child_rip+0xa/0x11 [<ffffffff802471eb>] kthread+0x0/0x75 [<ffffffff8020cbdf>] child_rip+0x0/0x11 ext4_da_writepages: jbd2_start: 1024 pages, ino 21705; err -30 Pid: 272, comm: pdflush Not tainted 2.6.28.4-onda #1 Call Trace: [<ffffffff802f8d6f>] ext4_da_writepages+0xcc/0x333 [<ffffffff8026eb67>] do_writepages+0x20/0x2d [<ffffffff802aaf73>] __writeback_single_inode+0x147/0x285 [<ffffffff80231948>] dequeue_task_fair+0x96/0x10d [<ffffffff802ab3f2>] generic_sync_sb_inodes+0x1bf/0x302 [<ffffffff802ab82d>] writeback_inodes+0x56/0x9d [<ffffffff8026f469>] wb_kupdate+0x9e/0x10d [<ffffffff8026f812>] pdflush+0x0/0x1f8 [<ffffffff8026f95e>] pdflush+0x14c/0x1f8 [<ffffffff8026f3cb>] wb_kupdate+0x0/0x10d [<ffffffff80247232>] kthread+0x47/0x75 [<ffffffff802309b4>] schedule_tail+0x28/0x60 [<ffffffff8020cbe9>] child_rip+0xa/0x11 [<ffffffff802471eb>] kthread+0x0/0x75 [<ffffffff8020cbdf>] child_rip+0x0/0x11
-30 is EROFS I assume you've had some error that shut down the filesystem previously, can you look? This seems to ring a bell for me but can't remember for sure if it's fixed, yet. /* start a new transaction*/ handle = ext4_journal_start(inode, needed_blocks); if (IS_ERR(handle)) { ret = PTR_ERR(handle); printk(KERN_CRIT "%s: jbd2_start: " "%ld pages, ino %lu; err %d\n", __func__, wbc->nr_to_write, inode->i_ino, ret); dump_stack(); goto out_writepages; } we should maybe check for a shutdown fs here before we generate all these messages.
Yes, it's fixed already in 2.6.29-rcX, and it's in the queue of patches that I'm preparing for 2.6.27 and 2.6.28.
Thanks Ted, thought so but couldn't remember for sure :)