Slow performance with Btrfs RAID 10 with a failed disk
Hey,
I'm using RAID 10, and one of the disks has recently failed [1], and I'm seeing plenty of warning and errors in the dmesg output [2].
What kind of performance should be expected from Btrfs when a disk has failed? [3] At the moment, the system seems very slow. One contributing factor may be that all the logging that Btrfs is generating is being written to the btrfs filesystem that's degraded, probably causing more log messages to be produced.
I guess that replacing the failed disk is the long term solution to get the filesystem back in to proper operation, but is there anything else that can be done to get it back operating until then?
Also, is there anything that can stop btrfs logging so much about the failures, now that I know that a disk has failed?
Thanks,
Chris
1:
Nov 26 19:20:56 localhost vmunix: [5117520.484302] sd 0:1:0:5: [sdf] Unaligned partial completion (resid=52, sector_sz=512) Nov 26 19:20:56 localhost vmunix: [5117520.525506] sd 0:1:0:5: [sdf] tag#360 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE Nov 26 19:20:56 localhost vmunix: [5117520.525525] sd 0:1:0:5: [sdf] Unaligned partial completion (resid=24384, sector_sz=512) Nov 26 19:20:56 localhost vmunix: [5117520.566649] sd 0:1:0:5: [sdf] tag#360 Sense Key : Hardware Error [current] Nov 26 19:20:57 localhost vmunix: [5117520.597829] sd 0:1:0:5: [sdf] tag#363 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE Nov 26 19:20:57 localhost vmunix: [5117520.637610] sd 0:1:0:5: [sdf] tag#360 Add. Sense: Logical unit failure Nov 26 19:20:57 localhost vmunix: [5117520.668134] sd 0:1:0:5: [sdf] tag#363 Sense Key : Hardware Error [current] Nov 26 19:20:57 localhost vmunix: [5117520.668136] sd 0:1:0:5: [sdf] tag#363 Add. Sense: Logical unit failure Nov 26 19:20:58 localhost vmunix: [5117520.707347] sd 0:1:0:5: [sdf] tag#360 CDB: Write(10) 2a 00 46 86 12 00 00 00 80 00 Nov 26 19:20:58 localhost vmunix: [5117520.736962] sd 0:1:0:5: [sdf] tag#363 CDB: Write(10) 2a 00 47 1e 0e 00 00 02 00 00 Nov 26 19:20:58 localhost vmunix: [5117520.774569] print_req_error: critical target error, dev sdf, sector 1183191552 flags 100001 Nov 26 19:20:59 localhost vmunix: [5117520.774573] BTRFS error (device sda3): bdev /dev/sdf errs: wr 2, rd 0, flush 0, corrupt 0, gen 0 Nov 26 19:20:59 localhost vmunix: [5117520.803740] print_req_error: critical target error, dev sdf, sector 1193152000 flags 4001 Nov 26 19:20:59 localhost vmunix: [5117520.803746] BTRFS error (device sda3): bdev /dev/sdf errs: wr 3, rd 0, flush 0, corrupt 0, gen 0 Nov 26 19:20:59 localhost vmunix: [5117520.840559] sd 0:1:0:5: [sdf] Unaligned partial completion (resid=52, sector_sz=512) Nov 26 19:20:59 localhost vmunix: [5117520.868966] BTRFS error (device sda3): bdev /dev/sdf errs: wr 4, rd 0, flush 0, corrupt 0, gen 0 Nov 26 19:21:00 localhost vmunix: [5117520.869037] sd 0:1:0:5: [sdf] Unaligned partial completion (resid=52, sector_sz=512) Nov 26 19:21:00 localhost vmunix: [5117520.869042] sd 0:1:0:5: [sdf] tag#385 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
2:
[5168107.359619] BTRFS error (device sda3): error writing primary super block to device 6 [5168107.932712] BTRFS warning (device sda3): lost page write due to IO error on /dev/sdf [5168108.091827] BTRFS error (device sda3): error writing primary super block to device 6 [5168108.155217] BTRFS warning (device sda3): lost page write due to IO error on /dev/sdf [5168108.288296] BTRFS error (device sda3): error writing primary super block to device 6 [5168108.972431] BTRFS warning (device sda3): lost page write due to IO error on /dev/sdf [5168109.204083] BTRFS error (device sda3): error writing primary super block to device 6 [5168109.595413] btrfs_dev_stat_print_on_error: 296 callbacks suppressed [5168109.595422] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071725, rd 408586, flush 0, corrupt 0, gen 0 [5168109.639670] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071726, rd 408586, flush 0, corrupt 0, gen 0 [5168109.664981] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071727, rd 408586, flush 0, corrupt 0, gen 0 [5168109.689197] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071728, rd 408586, flush 0, corrupt 0, gen 0 [5168109.728189] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071729, rd 408586, flush 0, corrupt 0, gen 0 [5168109.744894] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071730, rd 408586, flush 0, corrupt 0, gen 0 [5168109.755457] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071731, rd 408586, flush 0, corrupt 0, gen 0 [5168109.831763] BTRFS warning (device sda3): lost page write due to IO error on /dev/sdf [5168109.848128] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071732, rd 408586, flush 0, corrupt 0, gen 0 [5168109.849445] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071733, rd 408586, flush 0, corrupt 0, gen 0 [5168109.917277] BTRFS error (device sda3): error writing primary super block to device 6 [5168109.941132] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071734, rd 408586, flush 0, corrupt 0, gen 0 [5168110.009785] BTRFS warning (device sda3): lost page write due to IO error on /dev/sdf
3:
Label: none uuid: 620115c7-89c7-4d79-a0bb-4957057d9991
Total devices 6 FS bytes used 1.08TiB devid 1 size 72.70GiB used 72.70GiB path /dev/sda3 devid 2 size 72.70GiB used 72.70GiB path /dev/sdb3 devid 3 size 931.48GiB used 555.73GiB path /dev/sdc devid 4 size 931.48GiB used 555.73GiB path /dev/sdd devid 5 size 931.48GiB used 555.73GiB path /dev/sde *** Some devices missing
Quoted 11 lines · Christopher Baines
Hey,
I'm using RAID 10, and one of the disks has recently failed [1], and I'm seeing plenty of warning and errors in the dmesg output [2].
What kind of performance should be expected from Btrfs when a disk has failed? [3] At the moment, the system seems very slow. One contributing factor may be that all the logging that Btrfs is generating is being written to the btrfs filesystem that's degraded, probably causing more log messages to be produced.
Quoted 6 lines
I guess that replacing the failed disk is the long term solution to get the filesystem back in to proper operation, but is there anything else that can be done to get it back operating until then?
Also, is there anything that can stop btrfs logging so much about the failures, now that I know that a disk has failed?
You can solve both problems by replacing the disc, or if possible, just removing it from the array. You should, in theory, be able to convert to regular raid1 and then remove the failed disc, though it will likely take a while. Given your output below, I'd actually drop /dev/sdb as well, and look at replacing both with a single 1TB disc like your other three.
The issue here is that BTRFS doesn't see the disc as failed, so it keeps trying to access it. That's what's slowing things down (because it eventually times out on the access attempt) and why it's logging so much (because BTRFS logs every IO error it encounters (like it should)).
Quoted 59 lines
Thanks,
Chris
1:
Nov 26 19:20:56 localhost vmunix: [5117520.484302] sd 0:1:0:5: [sdf] Unaligned partial completion (resid=52, sector_sz=512) Nov 26 19:20:56 localhost vmunix: [5117520.525506] sd 0:1:0:5: [sdf] tag#360 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE Nov 26 19:20:56 localhost vmunix: [5117520.525525] sd 0:1:0:5: [sdf] Unaligned partial completion (resid=24384, sector_sz=512) Nov 26 19:20:56 localhost vmunix: [5117520.566649] sd 0:1:0:5: [sdf] tag#360 Sense Key : Hardware Error [current] Nov 26 19:20:57 localhost vmunix: [5117520.597829] sd 0:1:0:5: [sdf] tag#363 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE Nov 26 19:20:57 localhost vmunix: [5117520.637610] sd 0:1:0:5: [sdf] tag#360 Add. Sense: Logical unit failure Nov 26 19:20:57 localhost vmunix: [5117520.668134] sd 0:1:0:5: [sdf] tag#363 Sense Key : Hardware Error [current] Nov 26 19:20:57 localhost vmunix: [5117520.668136] sd 0:1:0:5: [sdf] tag#363 Add. Sense: Logical unit failure Nov 26 19:20:58 localhost vmunix: [5117520.707347] sd 0:1:0:5: [sdf] tag#360 CDB: Write(10) 2a 00 46 86 12 00 00 00 80 00 Nov 26 19:20:58 localhost vmunix: [5117520.736962] sd 0:1:0:5: [sdf] tag#363 CDB: Write(10) 2a 00 47 1e 0e 00 00 02 00 00 Nov 26 19:20:58 localhost vmunix: [5117520.774569] print_req_error: critical target error, dev sdf, sector 1183191552 flags 100001 Nov 26 19:20:59 localhost vmunix: [5117520.774573] BTRFS error (device sda3): bdev /dev/sdf errs: wr 2, rd 0, flush 0, corrupt 0, gen 0 Nov 26 19:20:59 localhost vmunix: [5117520.803740] print_req_error: critical target error, dev sdf, sector 1193152000 flags 4001 Nov 26 19:20:59 localhost vmunix: [5117520.803746] BTRFS error (device sda3): bdev /dev/sdf errs: wr 3, rd 0, flush 0, corrupt 0, gen 0 Nov 26 19:20:59 localhost vmunix: [5117520.840559] sd 0:1:0:5: [sdf] Unaligned partial completion (resid=52, sector_sz=512) Nov 26 19:20:59 localhost vmunix: [5117520.868966] BTRFS error (device sda3): bdev /dev/sdf errs: wr 4, rd 0, flush 0, corrupt 0, gen 0 Nov 26 19:21:00 localhost vmunix: [5117520.869037] sd 0:1:0:5: [sdf] Unaligned partial completion (resid=52, sector_sz=512) Nov 26 19:21:00 localhost vmunix: [5117520.869042] sd 0:1:0:5: [sdf] tag#385 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE2:
[5168107.359619] BTRFS error (device sda3): error writing primary super block to device 6 [5168107.932712] BTRFS warning (device sda3): lost page write due to IO error on /dev/sdf [5168108.091827] BTRFS error (device sda3): error writing primary super block to device 6 [5168108.155217] BTRFS warning (device sda3): lost page write due to IO error on /dev/sdf [5168108.288296] BTRFS error (device sda3): error writing primary super block to device 6 [5168108.972431] BTRFS warning (device sda3): lost page write due to IO error on /dev/sdf [5168109.204083] BTRFS error (device sda3): error writing primary super block to device 6 [5168109.595413] btrfs_dev_stat_print_on_error: 296 callbacks suppressed [5168109.595422] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071725, rd 408586, flush 0, corrupt 0, gen 0 [5168109.639670] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071726, rd 408586, flush 0, corrupt 0, gen 0 [5168109.664981] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071727, rd 408586, flush 0, corrupt 0, gen 0 [5168109.689197] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071728, rd 408586, flush 0, corrupt 0, gen 0 [5168109.728189] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071729, rd 408586, flush 0, corrupt 0, gen 0 [5168109.744894] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071730, rd 408586, flush 0, corrupt 0, gen 0 [5168109.755457] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071731, rd 408586, flush 0, corrupt 0, gen 0 [5168109.831763] BTRFS warning (device sda3): lost page write due to IO error on /dev/sdf [5168109.848128] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071732, rd 408586, flush 0, corrupt 0, gen 0 [5168109.849445] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071733, rd 408586, flush 0, corrupt 0, gen 0 [5168109.917277] BTRFS error (device sda3): error writing primary super block to device 6 [5168109.941132] BTRFS error (device sda3): bdev /dev/sdf errs: wr 5071734, rd 408586, flush 0, corrupt 0, gen 0 [5168110.009785] BTRFS warning (device sda3): lost page write due to IO error on /dev/sdf3:
Label: none uuid: 620115c7-89c7-4d79-a0bb-4957057d9991Total devices 6 FS bytes used 1.08TiB devid 1 size 72.70GiB used 72.70GiB path /dev/sda3 devid 2 size 72.70GiB used 72.70GiB path /dev/sdb3 devid 3 size 931.48GiB used 555.73GiB path /dev/sdc devid 4 size 931.48GiB used 555.73GiB path /dev/sdd devid 5 size 931.48GiB used 555.73GiB path /dev/sde *** Some devices missing
Quoted 26 lines · Austin S. Hemmelgarn
Quoted 11 lines · Christopher Baines
Hey,
I'm using RAID 10, and one of the disks has recently failed [1], and I'm seeing plenty of warning and errors in the dmesg output [2].
What kind of performance should be expected from Btrfs when a disk has failed? [3] At the moment, the system seems very slow. One contributing factor may be that all the logging that Btrfs is generating is being written to the btrfs filesystem that's degraded, probably causing more log messages to be produced.
Quoted 6 lines
I guess that replacing the failed disk is the long term solution to get the filesystem back in to proper operation, but is there anything else that can be done to get it back operating until then?
Also, is there anything that can stop btrfs logging so much about the failures, now that I know that a disk has failed?
You can solve both problems by replacing the disc, or if possible, just removing it from the array. You should, in theory, be able to convert to regular raid1 and then remove the failed disc, though it will likely take a while. Given your output below, I'd actually drop
/dev/sdbas well, and look at replacing both with a single 1TB disc like your other three.
Quoted 5 lines
The issue here is that BTRFS doesn't see the disc as failed, so it keeps trying to access it. That's what's slowing things down (because it eventually times out on the access attempt) and why it's logging so much (because BTRFS logs every IO error it encounters (like it should)).
Thanks for the tips :)
I've now remounted the filesystem with the degraded flag.
However, I haven't managed to remove the disk from the array yet.
$ sudo btrfs filesystem show / Label: none uuid: 620115c7-89c7-4d79-a0bb-4957057d9991 Total devices 6 FS bytes used 1.08TiB devid 1 size 72.70GiB used 72.70GiB path /dev/sda3 devid 2 size 72.70GiB used 72.70GiB path /dev/sdb3 devid 3 size 931.48GiB used 530.73GiB path /dev/sdc devid 4 size 931.48GiB used 530.73GiB path /dev/sdd devid 5 size 931.48GiB used 530.73GiB path /dev/sde *** Some devices missing
$ sudo btrfs device delete missing / ERROR: error removing device 'missing': no missing devices found to remove
So Btrfs knows at some level that a device is missing, from the output of the first command, but it won't delete the missing device.
Am I missing something?
Thanks,
Chris