0017623: xfs filesystem hung up - CentOS Bug Tracker

ID	Project	Category	View Status	Date Submitted	Last Update

0017623	CentOS-7	kernel	public	2020-07-27 13:46	2020-07-27 13:50

Reporter	petrilak.m	Assigned To
Priority	urgent	Severity	block	Reproducibility	always
Status	new	Resolution	open
Platform	x86-64	OS	CentOS	OS Version	7.8.2003
Product Version	7.8-2003

Summary	0017623: xfs filesystem hung up
Description	I run script for process some files (about 23 thousands files). Program convert files to other format via temporary file (located in data directory). After each conversion temp file is deleted. Temp file is deleted by rm command. After some time, in some iteration, process rm hung up in state D+. If temporary file is created in /tmp (ramdisk), there are no problem. Normal disk is xfs filesystem on disk array connected via fibrechannel. There is also problem in other operation other than deletion. Some program for process data hungup, also. State D+ stay for very long time (>24 hours). iotop indicate all data transfers are zero. A have physical computer with only one virtual machine (oVirt virtualization). Data disk is connected to virtual machine as "direct LUN". I run xfs_repair with some arguments for repair filesystem, but it does not help. I think there is some problem in xfs filesystem driver. If you need some logs or dumps, feel free to inform me. iostat output: Device: rrqm/s wrqm/s r/s w/s rMB/s wMB/s avgrq-sz avgqu-sz await r_await w_await svctm %util sdd 0.00 0.00 0.00 0.00 0.00 0.00 0.00 22.00 0.00 0.00 0.00 0.00 100.00 ps aufx: fischer+ 4541 0.0 0.0 125976 2304 pts/1 Ss Jul24 0:00 \| \_ bash fischer+ 4617 0.0 0.0 129824 2136 pts/1 S+ Jul24 0:00 \| \| \_ /usr/bin/perl ./pl/run_domains.pl fischer+ 4618 0.0 0.0 113292 1204 pts/1 S+ Jul24 0:00 \| \| \_ sh -c /mnt/data/DisALEXI/CZECH/code/Programs/preprocess_inputs.pl >& /mnt/data/DisALEXI/CZECH/domains/DryPan_2001-2020/messages/preprocess_inputs.ou fischer+ 4620 0.0 0.0 129960 2360 pts/1 S+ Jul24 0:00 \| \| \_ /usr/bin/perl /mnt/data/DisALEXI/CZECH/code/Programs/preprocess_inputs.pl fischer+ 77528 0.0 0.0 113292 1204 pts/1 S+ Jul26 0:00 \| \| \_ sh -c /mnt/data/DisALEXI/CZECH/code/SysPrograms/bin/modlai_daily_smooth.csh MCD15A3H 2000210 2020180 >& /mnt/data/DisALEXI/CZECH/domains/Dry fischer+ 77529 0.0 0.0 137404 3064 pts/1 S+ Jul26 0:16 \| \| \_ /bin/csh -f /mnt/data/DisALEXI/CZECH/code/SysPrograms/bin/modlai_daily_smooth.csh MCD15A3H 2000210 2020180 fischer+ 114454 6.0 0.0 64660 58072 pts/1 D+ Jul26 138:44 \| \| \_ /mnt/data/DisALEXI/CZECH/code/SysPrograms/bin/modlai_daily_smooth.exe lai_input.2000210-2020180.txt Below is dmesg log after program stuck in D+ state: [133922.264189] INFO: task kworker/u480:2:67137 blocked for more than 120 seconds. [133922.265249] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [133922.266034] kworker/u480:2 D ffff8b1201583150 0 67137 2 0x00000080 [133922.266770] Workqueue: writeback bdi_writeback_workfn (flush-253:2) [133922.267526] Call Trace: [133922.268275] [<ffffffffb1ce0628>] ? __enqueue_entity+0x78/0x80 [133922.269039] [<ffffffffb2385da9>] schedule+0x29/0x70 [133922.269780] [<ffffffffb23838b1>] schedule_timeout+0x221/0x2d0 [133922.270534] [<ffffffffb1ccc5dd>] ? down_trylock+0x2d/0x40 [133922.271308] [<ffffffffc086821f>] ? xfs_buf_trylock+0x1f/0xc0 [xfs] [133922.272054] [<ffffffffb23851b7>] __down_common+0xaa/0x104 [133922.272791] [<ffffffffc0868500>] ? _xfs_buf_find.isra.9+0x170/0x330 [xfs] [133922.273551] [<ffffffffb238522e>] __down+0x1d/0x1f [133922.274297] [<ffffffffb1ccc631>] down+0x41/0x50 [133922.275053] [<ffffffffc08682fc>] xfs_buf_lock+0x3c/0xd0 [xfs] [133922.275784] [<ffffffffc0868500>] _xfs_buf_find.isra.9+0x170/0x330 [xfs] [133922.276551] [<ffffffffc0868755>] xfs_buf_get_map+0x35/0x250 [xfs] [133922.277323] [<ffffffffc0868f20>] xfs_buf_read_map+0x30/0x160 [xfs] [133922.278089] [<ffffffffc0899d49>] xfs_trans_read_buf_map+0xe9/0x2c0 [xfs] [133922.278821] [<ffffffffc083e934>] xfs_btree_read_buf_block.constprop.33+0xa4/0xe0 [xfs] [133922.279589] [<ffffffffc0842a45>] xfs_btree_lookup_get_block+0x95/0x1a0 [xfs] [133922.280364] [<ffffffffc0842cff>] xfs_btree_lookup+0xdf/0x420 [xfs] [133922.281130] [<ffffffffc083418f>] xfs_bmbt_lookup_eq+0x2f/0x40 [xfs] [133922.281863] [<ffffffffc08371e4>] xfs_bmap_add_extent_delay_real+0x864/0x11e0 [xfs] [133922.282628] [<ffffffffc0839624>] ? xfs_bmap_btalloc+0x2d4/0x7e0 [xfs] [133922.283385] [<ffffffffb1e28852>] ? kmem_cache_alloc+0x1c2/0x1f0 [133922.284137] [<ffffffffc083b7c8>] xfs_bmapi_write+0x7f8/0xb70 [xfs] [133922.284854] [<ffffffffc0888d67>] ? kmem_zone_alloc+0x97/0x130 [xfs] [133922.285593] [<ffffffffc0877cf2>] xfs_iomap_write_allocate+0x182/0x380 [xfs] [133922.286333] [<ffffffffc0861446>] xfs_map_blocks+0x1a6/0x220 [xfs] [133922.287071] [<ffffffffc0862484>] xfs_do_writepage+0x174/0x550 [xfs] [133922.287766] [<ffffffffb1dca1bc>] write_cache_pages+0x21c/0x470 [133922.288501] [<ffffffffc0862310>] ? xfs_vm_writepages+0xa0/0xa0 [xfs] [133922.289232] [<ffffffffb1f546a0>] ? submit_bio+0x70/0x150 [133922.290707] [<ffffffffc08622db>] xfs_vm_writepages+0x6b/0xa0 [xfs] [133922.291879] [<ffffffffb1dcb211>] do_writepages+0x21/0x50 [133922.292870] [<ffffffffb1e7cc30>] __writeback_single_inode+0x40/0x260 [133922.293816] [<ffffffffb1cc7745>] ? wake_up_bit+0x25/0x30 [133922.294726] [<ffffffffb1e7d7c4>] writeback_sb_inodes+0x1c4/0x430 [133922.295624] [<ffffffffb1e7dacf>] __writeback_inodes_wb+0x9f/0xd0 [133922.296642] [<ffffffffb1e7dfb3>] wb_writeback+0x263/0x2f0 [133922.297562] [<ffffffffb1e7eaac>] bdi_writeback_workfn+0x1cc/0x460 [133922.298441] [<ffffffffb1cbe6bf>] process_one_work+0x17f/0x440 [133922.299294] [<ffffffffb1cbf7d6>] worker_thread+0x126/0x3c0 [133922.300148] [<ffffffffb1cbf6b0>] ? manage_workers.isra.26+0x2a0/0x2a0 [133922.300967] [<ffffffffb1cc6691>] kthread+0xd1/0xe0 [133922.301757] [<ffffffffb1cc65c0>] ? insert_kthread_work+0x40/0x40 [133922.302615] [<ffffffffb2392d24>] ret_from_fork_nospec_begin+0xe/0x21 [133922.303442] [<ffffffffb1cc65c0>] ? insert_kthread_work+0x40/0x40
Steps To Reproduce	after each run of my program for compute. A have try to run it for 10-times and allways it hung up
Tags	xfs

abrt_hash
URL

Date Modified	Username	Field	Change
2020-07-27 13:46	petrilak.m	New Issue
2020-07-27 13:46	petrilak.m	Tag Attached: xfs
2020-07-27 13:48	TrevorH	Note Added: 0037419
2020-07-27 13:50	petrilak.m	Note Added: 0037420