raid 5 to raid 6 conversion when adding drive

    • OMV 4.x

    This site uses cookies. By continuing to browse this site, you are agreeing to our Cookie Policy.

    • raid 5 to raid 6 conversion when adding drive

      Hello,

      I am looking to convert my current raid 5 array with 5x8TB drives into a raid 6 with 6x8TB drives. I found this guide in a post from a few years ago and was wondering if this would work with OMV and would show in the gui after it is complete?

      ewams.net/?date=2013/05/02&vie…g_RAID5_to_RAID6_in_mdadm

      Thanks for any assistance, I know a lot of you guys dont like RAID but I would like to add an additional parity disk to help against physical drive failures while maintaining the data in the existing raid.
    • chuckado wrote:

      I found this guide in a post from a few years ago and was wondering if this would work with OMV and would show in the gui after it is complete?
      That should work.

      chuckado wrote:

      I know a lot of you guys dont like RAID
      I don't mind raid at all. I just think most OMV users don't need it because uptime is not critical for a home user (or even some small businesses). We also push against it when people think it is backup.
      omv 4.1.17 arrakis | 64 bit | 4.15 proxmox kernel | omvextrasorg 4.1.13
      omv-extras.org plugins source code and issue tracker - github

      Please read this before posting a question and this and this for docker questions.
      Please don't PM for support... Too many PMs!
    • ryecoaaron wrote:

      chuckado wrote:

      I found this guide in a post from a few years ago and was wondering if this would work with OMV and would show in the gui after it is complete?
      That should work.

      chuckado wrote:

      I know a lot of you guys dont like RAID
      I don't mind raid at all. I just think most OMV users don't need it because uptime is not critical for a home user (or even some small businesses). We also push against it when people think it is backup.
      awesome, thanks now just waiting for a drives to be on sale then I will give it a go.
    • chuckado wrote:

      When it is completed, will omv web gui be updated to show the new version of raid?
      yes
      omv 4.1.17 arrakis | 64 bit | 4.15 proxmox kernel | omvextrasorg 4.1.13
      omv-extras.org plugins source code and issue tracker - github

      Please read this before posting a question and this and this for docker questions.
      Please don't PM for support... Too many PMs!
    • instead of adding my 2 new disks to my raid 5 array via cli, using the following command as shown in the guide.

      mdadm --add /dev/md0 /dev/sda

      I instead used the gui and used the grow button on the raid array which I assume makes them functioning members of the raid instead of spares. Is there any way for me to reverse or undo this or will I have to buy another disk and follow the instructions correctly. Correctly the raid array is reshaping.
    • chuckado wrote:

      I instead used the gui and used the grow button on the raid array which I assume makes them functioning members of the raid instead of spares. Is there any way for me to reverse or undo this or will I have to buy another disk and follow the instructions correctly. Correctly the raid array is reshaping.
      Just to be clear, you added both with the grow button and now you want to use one of the disks added to create raid 6 instead? What is the output of: cat /proc/mdstat
      omv 4.1.17 arrakis | 64 bit | 4.15 proxmox kernel | omvextrasorg 4.1.13
      omv-extras.org plugins source code and issue tracker - github

      Please read this before posting a question and this and this for docker questions.
      Please don't PM for support... Too many PMs!
    • Brainfuck Source Code

      1. cat /proc/mdstat
      2. Personalities : [raid1] [raid6] [raid5] [raid4] [linear] [multipath] [raid0] [raid10]
      3. md127 : active raid5 sdi[6] sdh[5] sdc[0] sde[2] sdg[4] sdf[3] sdd[1]
      4. 31255580672 blocks super 1.2 level 5, 512k chunk, algorithm 2 [7/7] [UUUUUUU]
      5. [=========>...........] reshape = 48.9% (3823605824/7813895168) finish=1465.7min speed=45371K/sec
      6. bitmap: 6/59 pages [24KB], 65536KB chunk
      7. md0 : active raid1 sda[0] sdb[1]
      8. 937561408 blocks super 1.2 [2/2] [UU]
      9. bitmap: 1/7 pages [4KB], 65536KB chunk
      10. unused devices: <none>
      Display All
      Above is the output, and you are correct. I would like to use one of the disks that I added via grow to convert the raid 5 to a raid 6.
    • Unfortunately, using the grow button to add both disks is doing exactly what it is supposed to. You will have to wait for the array to get done reshaping and then you will have to reduce it by one disk (not something I like to do) to make it raid 6. I really hope you have backup if you try to reduce it.
      omv 4.1.17 arrakis | 64 bit | 4.15 proxmox kernel | omvextrasorg 4.1.13
      omv-extras.org plugins source code and issue tracker - github

      Please read this before posting a question and this and this for docker questions.
      Please don't PM for support... Too many PMs!
    • New

      So i followed the process and everything looked to be going as expected. I got an email that the spare drive has failed. Is this just part of the process of mdadm adding it to the raid?

      Source Code

      1. Version : 1.2
      2. Creation Time : Sat Feb 3 15:52:55 2018
      3. Raid Level : raid6
      4. Array Size : 46883371008 (44711.47 GiB 48008.57 GB)
      5. Used Dev Size : 7813895168 (7451.91 GiB 8001.43 GB)
      6. Raid Devices : 8
      7. Total Devices : 8
      8. Persistence : Superblock is persistent
      9. Intent Bitmap : Internal
      10. Update Time : Mon Jan 14 10:32:12 2019
      11. State : clean, degraded, reshaping
      12. Active Devices : 7
      13. Working Devices : 7
      14. Failed Devices : 1
      15. Spare Devices : 0
      16. Layout : left-symmetric-6
      17. Chunk Size : 512K
      18. Reshape Status : 0% complete
      19. New Layout : left-symmetric
      20. Name : omv2:Raid5 (local to host omv2)
      21. UUID : a0e694f5:f6d6bf81:34c60a50:6b934b20
      22. Events : 128257
      23. Number Major Minor RaidDevice State
      24. 0 8 64 0 active sync /dev/sde
      25. 1 8 80 1 active sync /dev/sdf
      26. 2 8 96 2 active sync /dev/sdg
      27. 3 8 112 3 active sync /dev/sdh
      28. 4 8 128 4 active sync /dev/sdi
      29. 6 8 176 5 active sync /dev/sdl
      30. 5 8 160 6 active sync /dev/sdk
      31. 7 8 144 7 faulty /dev/sdj
      Display All
    • New

      The new disk sdj, is also no longer showing up in the disks tab. Should I power down the server and confirm it is cabled firmly or do i have to wait for something to happen with the reshaping?

      Also adding what seems to be the relevant data from the syslog. Those messages continue for another minute or so then stop. It would go over the character limit if I add them all.

      Source Code

      1. Jan 14 10:25:38 omv2 kernel: [ 947.158095] md: could not open unknown-block(8,144).
      2. Jan 14 10:25:38 omv2 kernel: [ 947.158099] md: md_import_device returned -16
      3. Jan 14 10:25:38 omv2 kernel: [ 947.158126] md: could not open unknown-block(8,144).
      4. Jan 14 10:25:38 omv2 kernel: [ 947.158126] md: md_import_device returned -16
      5. Jan 14 10:27:08 omv2 kernel: [ 1037.964676] md/raid:md127: device sdh operational as raid disk 3
      6. Jan 14 10:27:08 omv2 kernel: [ 1037.964677] md/raid:md127: device sdg operational as raid disk 2
      7. Jan 14 10:27:08 omv2 kernel: [ 1037.964678] md/raid:md127: device sdf operational as raid disk 1
      8. Jan 14 10:27:08 omv2 kernel: [ 1037.964678] md/raid:md127: device sdl operational as raid disk 5
      9. Jan 14 10:27:08 omv2 kernel: [ 1037.964679] md/raid:md127: device sde operational as raid disk 0
      10. Jan 14 10:27:08 omv2 kernel: [ 1037.964680] md/raid:md127: device sdi operational as raid disk 4
      11. Jan 14 10:27:08 omv2 kernel: [ 1037.964680] md/raid:md127: device sdk operational as raid disk 6
      12. Jan 14 10:27:08 omv2 kernel: [ 1037.985561] md/raid:md127: raid level 6 active with 7 out of 8 devices, algorithm 18
      13. Jan 14 10:27:19 omv2 kernel: [ 1048.798967] ata6.00: exception Emask 0x10 SAct 0x20000 SErr 0x400100 action 0x6 frozen
      14. Jan 14 10:27:19 omv2 kernel: [ 1048.798971] ata6.00: irq_stat 0x08000000, interface fatal error
      15. Jan 14 10:27:19 omv2 kernel: [ 1048.798973] ata6: SError: { UnrecovData Handshk }
      16. Jan 14 10:27:19 omv2 kernel: [ 1048.798975] ata6.00: failed command: WRITE FPDMA QUEUED
      17. Jan 14 10:27:19 omv2 kernel: [ 1048.798979] ata6.00: cmd 61/01:88:08:00:00/00:00:00:00:00/40 tag 17 ncq dma 512 out
      18. Jan 14 10:27:19 omv2 kernel: [ 1048.798979] res 40/00:00:08:00:00/00:00:00:00:00/40 Emask 0x10 (ATA bus error)
      19. Jan 14 10:27:19 omv2 kernel: [ 1048.798982] ata6.00: status: { DRDY }
      20. Jan 14 10:27:19 omv2 kernel: [ 1048.798984] ata6: hard resetting link
      21. Jan 14 10:27:20 omv2 kernel: [ 1049.114017] ata6: SATA link up 3.0 Gbps (SStatus 123 SControl 320)
      22. Jan 14 10:27:20 omv2 kernel: [ 1049.130767] ata6.00: configured for UDMA/133
      23. Jan 14 10:27:20 omv2 kernel: [ 1049.130775] ata6: EH complete
      24. Jan 14 10:27:23 omv2 systemd[1]: Created slice system-mdadm\x2dgrow\x2dcontinue.slice.
      25. Jan 14 10:27:23 omv2 systemd[1]: Started Manage MD Reshape on /dev/md127.
      26. Jan 14 10:27:23 omv2 kernel: [ 1052.959039] md: reshape of RAID array md127
      27. Jan 14 10:27:23 omv2 systemd[1]: mdadm-grow-continue@md127.service: Main process exited, code=exited, status=2/INVALIDARGUMENT
      28. Jan 14 10:27:23 omv2 systemd[1]: mdadm-grow-continue@md127.service: Unit entered failed state.
      29. Jan 14 10:27:23 omv2 systemd[1]: mdadm-grow-continue@md127.service: Failed with result 'exit-code'.
      30. Jan 14 10:27:23 omv2 kernel: [ 1053.050852] ata6.00: exception Emask 0x10 SAct 0x18000000 SErr 0x400100 action 0x6 frozen
      31. Jan 14 10:27:23 omv2 kernel: [ 1053.050856] ata6.00: irq_stat 0x08000000, interface fatal error
      32. Jan 14 10:27:23 omv2 kernel: [ 1053.050859] ata6: SError: { UnrecovData Handshk }
      33. Jan 14 10:27:23 omv2 kernel: [ 1053.050861] ata6.00: failed command: WRITE FPDMA QUEUED
      34. Jan 14 10:27:23 omv2 kernel: [ 1053.050865] ata6.00: cmd 61/08:d8:40:77:5f/00:00:20:01:00/40 tag 27 ncq dma 4096 out
      35. Jan 14 10:27:23 omv2 kernel: [ 1053.050865] res 40/00:00:80:fe:59/00:00:2f:02:00/40 Emask 0x10 (ATA bus error)
      36. Jan 14 10:27:23 omv2 kernel: [ 1053.050868] ata6.00: status: { DRDY }
      37. Jan 14 10:27:23 omv2 kernel: [ 1053.050869] ata6.00: failed command: WRITE FPDMA QUEUED
      38. Jan 14 10:27:23 omv2 kernel: [ 1053.050873] ata6.00: cmd 61/08:e0:80:fe:59/00:00:2f:02:00/40 tag 28 ncq dma 4096 out
      39. Jan 14 10:27:23 omv2 kernel: [ 1053.050873] res 40/00:00:80:fe:59/00:00:2f:02:00/40 Emask 0x10 (ATA bus error)
      40. Jan 14 10:27:23 omv2 kernel: [ 1053.050876] ata6.00: status: { DRDY }
      41. Jan 14 10:27:23 omv2 kernel: [ 1053.050878] ata6: hard resetting link
      42. Jan 14 10:27:24 omv2 kernel: [ 1053.364240] ata6: SATA link up 3.0 Gbps (SStatus 123 SControl 320)
      43. Jan 14 10:27:24 omv2 kernel: [ 1053.381076] ata6.00: configured for UDMA/133
      44. Jan 14 10:27:24 omv2 kernel: [ 1053.381084] ata6: EH complete
      45. Jan 14 10:27:24 omv2 kernel: [ 1053.406833] ata6.00: exception Emask 0x10 SAct 0x4000003c SErr 0x400100 action 0x6 frozen
      46. Jan 14 10:27:24 omv2 kernel: [ 1053.406838] ata6.00: irq_stat 0x08000000, interface fatal error
      47. Jan 14 10:27:24 omv2 kernel: [ 1053.406840] ata6: SError: { UnrecovData Handshk }
      48. Jan 14 10:27:24 omv2 kernel: [ 1053.406842] ata6.00: failed command: WRITE FPDMA QUEUED
      49. Jan 14 10:27:24 omv2 kernel: [ 1053.406846] ata6.00: cmd 61/08:10:88:fe:59/00:00:2f:02:00/40 tag 2 ncq dma 4096 out
      50. Jan 14 10:27:24 omv2 kernel: [ 1053.406846] res 40/00:00:48:77:5f/00:00:20:01:00/40 Emask 0x10 (ATA bus error)
      51. Jan 14 10:27:24 omv2 kernel: [ 1053.406849] ata6.00: status: { DRDY }
      52. Jan 14 10:27:24 omv2 kernel: [ 1053.406850] ata6.00: failed command: WRITE FPDMA QUEUED
      53. Jan 14 10:27:24 omv2 kernel: [ 1053.406854] ata6.00: cmd 61/28:18:00:f8:a3/01:00:6f:02:00/40 tag 3 ncq dma 151552 out
      54. Jan 14 10:27:24 omv2 kernel: [ 1053.406854] res 40/00:00:48:77:5f/00:00:20:01:00/40 Emask 0x10 (ATA bus error)
      55. Jan 14 10:27:24 omv2 kernel: [ 1053.406857] ata6.00: status: { DRDY }
      56. Jan 14 10:27:24 omv2 kernel: [ 1053.406858] ata6.00: failed command: WRITE FPDMA QUEUED
      57. Jan 14 10:28:34 omv2 kernel: [ 1124.026509] print_req_error: I/O error, dev sdj, sector 16
      58. Jan 14 10:28:34 omv2 kernel: [ 1124.026515] print_req_error: I/O error, dev sdj, sector 16
      59. Jan 14 10:28:34 omv2 kernel: [ 1124.026517] md: super_written gets error=10
      60. Jan 14 10:28:34 omv2 kernel: [ 1124.026520] md/raid:md127: Disk failure on sdj, disabling device.
      61. Jan 14 10:28:34 omv2 kernel: [ 1124.026520] md/raid:md127: Operation continuing on 7 devices.
      62. Jan 14 10:28:34 omv2 kernel: [ 1124.026532] ata6: EH complete
      Display All

      The post was edited 1 time, last by chuckado ().

    • New

      So at this point the reshape has stayed at 0% for the past 8 hours. Is it safe to assume that because the disk failed, it will not complete? If it will not complete, am I safe to power down and confirm the cabling of the drive to see if I can restore its functionality? Anything special I need to do on the OMV side to try to revive this brand new drive?
    • New

      Hope you have a backup but this is unfamiliar territory for me. It *seems* that the reshape has not started. So, I would imagine you could power down to check cables, etc.
      omv 4.1.17 arrakis | 64 bit | 4.15 proxmox kernel | omvextrasorg 4.1.13
      omv-extras.org plugins source code and issue tracker - github

      Please read this before posting a question and this and this for docker questions.
      Please don't PM for support... Too many PMs!
    • New

      chuckado wrote:

      Is there someone else who might know more?
      google :) I do know there is no "good" way to stop a reshape You can suspend it with something like echo "idle" > /sys/block/md127/md/sync_action followed by a reboot but I don't think anyone is going to give you a solution that is 100% safe.
      omv 4.1.17 arrakis | 64 bit | 4.15 proxmox kernel | omvextrasorg 4.1.13
      omv-extras.org plugins source code and issue tracker - github

      Please read this before posting a question and this and this for docker questions.
      Please don't PM for support... Too many PMs!
    • New

      I rebooted and the "faulty" drive is showing up but the raid will not come up. Any ideas on what to try to bring the raid back up?

      Source Code

      1. root@omv2:~# mdadm --detail /dev/md127
      2. /dev/md127:
      3. Version : 1.2
      4. Raid Level : raid0
      5. Total Devices : 8
      6. Persistence : Superblock is persistent
      7. State : inactive
      8. New Level : raid6
      9. New Layout : left-symmetric
      10. New Chunksize : 512K
      11. Name : omv2:Raid5 (local to host omv2)
      12. UUID : a0e694f5:f6d6bf81:34c60a50:6b934b20
      13. Events : 302753
      14. Number Major Minor RaidDevice
      15. - 8 64 - /dev/sde
      16. - 8 32 - /dev/sdc
      17. - 8 144 - /dev/sdj
      18. - 8 112 - /dev/sdh
      19. - 8 80 - /dev/sdf
      20. - 8 48 - /dev/sdd
      21. - 8 128 - /dev/sdi
      22. - 8 96 - /dev/sdg
      23. root@omv2:~# mdadm --assemble /dev/md127
      24. mdadm: /dev/md127 not identified in config file.
      25. root@omv2:~# mdadm --assemble /dev/md127 /dev/sd[ecjhfdig]
      26. mdadm: /dev/sdc is busy - skipping
      27. mdadm: /dev/sdd is busy - skipping
      28. mdadm: /dev/sde is busy - skipping
      29. mdadm: /dev/sdf is busy - skipping
      30. mdadm: /dev/sdg is busy - skipping
      31. mdadm: /dev/sdh is busy - skipping
      32. mdadm: /dev/sdi is busy - skipping
      33. mdadm: /dev/sdj is busy - skipping
      34. root@omv2:~# mdadm --stop /dev/md127
      35. mdadm: stopped /dev/md127
      36. root@omv2:~# mdadm --assemble /dev/md127 /dev/sd[ecjhfdig]
      37. mdadm: Failed to restore critical section for reshape, sorry.
      38. Possibly you needed to specify the --backup-file
      39. root@omv2:~# mdadm --assemble /dev/md127 /dev/sd[ecjhfdig] --backup-file=raid5ba ckup
      40. mdadm: /dev/sdc is busy - skipping
      41. mdadm: /dev/sdd is busy - skipping
      42. mdadm: /dev/sde is busy - skipping
      43. mdadm: /dev/sdf is busy - skipping
      44. mdadm: /dev/sdg is busy - skipping
      45. mdadm: /dev/sdh is busy - skipping
      46. mdadm: /dev/sdi is busy - skipping
      47. mdadm: /dev/sdj is busy - skipping
      48. root@omv2:~#
      Display All