Notifications
Clear all

Replace failed drive with larger driver?

(@bry2k)
Active Member Customer

I have a 6x8TB OWC RAID (HGST drives). One of the drives just died after 2 years of use. OWC warrants the drives for 3 years but 8TB HGST are no longer available. Can I replace with a 10TB or some other different/equivalent drive? Would I eventually be able to upgrade the RAID to the 10TB capacity once all the drives were swapped out?

Quote
Topic starter Posted : 04/02/2021 11:45 am
(@bry2k)
Active Member Customer

OWC has an 8TB Toshiba 

https://eshop.macsales.com/item/Toshiba/MG06ACA800E/

Will this work? Will my RAID suffer any performance issues and/or fail to allow this disk to be added for the rebuild due to any discrepancy in available sectors or some other variable I'm not aware of?

ReplyQuote
Topic starter Posted : 04/02/2021 12:31 pm
(@softraid-support)
Member Admin

@bry2k

the Toshiba will work fine. There are no issues with different disks, only if the new one is slightly smaller, then you can resize the volume down a couple GB.

ReplyQuote
Posted : 04/02/2021 2:02 pm
(@bry2k)
Active Member Customer

If drives are successively swapped out on a RAID that started out as 6x8TB with 10TB drives, could the RAID theoretically be resized to use the larger available space once all drives were 10TB ?

ReplyQuote
Topic starter Posted : 04/02/2021 7:50 pm
(@softraid-support)
Member Admin

@bry2k

Yes this is a feature. We have "resize volume" to do this.

ReplyQuote
Posted : 04/02/2021 10:21 pm
(@bry2k)
Active Member Customer

Always a bit terrifying when a RAID5 degrades. I'm pleased to report that the process was fairly straightforward and uneventful. I would note for anyone who reads this in the future that the only point that confused me for a few minutes was:

My RAID5 has two volumes.

I initialized the replacement disk.

I then had to add the replacement disk to BOTH volumes. That wasn't immediately apparent to me.

SoftRAID then did the rebuild on both volumes sequentially.

The initial time estimates reported by SoftRAID for the rebuild varied wildly, but the actual rebuild, once started on each volume, completed much faster than I would have expected.

Bummer than one of my HGST drives failed after only 2 years (approx 17,000 hours - the RAID is on 24/7).

I look forward to trying SoftRAID6. 

Logs below:

Feb 1 1845 - SoftRAID Monitor: Starting SMART test on all disks which support SMART.
Feb 1 1800 - SoftRAID Monitor: Finished SMART test on all disks. No disks failed the SMART test.
Feb 2 1841 - SoftRAID Monitor: Starting SMART test on all disks which support SMART.
Feb 2 1856 - SoftRAID Monitor: Finished SMART test on all disks. No disks failed the SMART test.
Feb 3 1837 - SoftRAID Monitor: Starting SMART test on all disks which support SMART.
Feb 3 1849 - SoftRAID Monitor: The disk at disk4, SoftRAID ID: 07E454145F49A4C0, Location: Thunderbolt, SATA bus, 7.28 TB is predicted to fail. This disk is 20 - 60 times more likely to fail in the next 2 - 6 months than a normal disk. This prediction is based on SMART data retrieved from the disk. This disk should be replaced soon. SMART Attributes used for failure prediction are: ID 5 (Reallocated SectorCount) = 0, ID 196 (Reallocated Event Count) = 0, ID 197 (Reallocation Pending Sector Count) = 8, ID 198 (Uncorrectable Sector Count) = 1.
Feb 3 1849 - SoftRAID Monitor: Finished SMART test on all disks.
Feb 4 0305 - SoftRAID Driver: A disk for the volume "RAID5_2019A" (disk11) encountered a read error (E00002CA). The disk (disk4, SoftRAID ID: 07E454145F49A4C0) was unable to read sectors. The error occurred at volume offset 1325303037952 (i/o block size 4997120). This disk should be replaced.
Feb 4 0319 - SoftRAID Driver: A disk (disk4, SoftRAID ID: 07E454145F49A4C0) for the SoftRAID volume "RAID5_2019A" (disk11) encountered multiple read or write errors. This disk has been marked "failed" and will no longer be used for when reading volume data.
Feb 4 0910 - SoftRAID Application: Launching application: SoftRAID XT, version 5.8.4
Feb 4 0940 - SoftRAID Driver: A disk (disk4, SoftRAID ID: 07E454145F49A4C0) for the SoftRAID volume "RAID5_2019A" (disk11) was removed or stopped responding while the volume was mounted and in use.
Feb 4 0940 - SoftRAID Driver: A disk (disk4, SoftRAID ID: 07E454145F49A4C0) for the SoftRAID volume "RAID5_2019B" (disk10) was removed or stopped responding while the volume was mounted and in use.
Feb 4 0919 - SoftRAID Application: Initializing the disk disk19, SN: VKGZ6WEX, SATA bus 0, id 9 (Thunderbolt) to GPT format.
Feb 4 0931 - SoftRAIDTool: Initializing EFI partition on disk19.
Feb 4 0934 - SoftRAID Application: The disk initialize command for disk disk19, SoftRAID ID: 08CCF68F0F9A2A00, SATA bus 0, id 9 (Thunderbolt) completed successfully.
Feb 4 0951 - SoftRAID Application: Adding a disk to the volume "RAID5_2019B" (disk10).
Feb 4 0951 - SoftRAIDTool: Rebuilding boot caches for volume "RAID5_2019B".
Feb 4 0951 - SoftRAID Application: The volume add disk command for volume "RAID5_2019B" (disk10) completed successfully. The disk disk19, SoftRAID ID: 08CCF68F0F9A2A00, SATA bus 0, id 9 (Thunderbolt) was added to the volume.
Feb 4 0951 - SoftRAID Driver: The RAID volume "RAID5_2019B" (disk10) is out of sync. A rebuild has started automatically.
Feb 4 1033 - SoftRAID Application: Adding a disk to the volume "RAID5_2019A" (disk11).
Feb 4 1053 - SoftRAID Application: The volume add disk command for the volume "RAID5_2019A" failed because a volume on this disk cannot be unmounted (error number = 13). The disk disk19, SoftRAID ID: 08CCF68F0F9A2A00, SATA bus 0, id 9 (Thunderbolt) could not be added to the volume.
Feb 4 1030 - SoftRAID Application: Quitting application
Feb 4 1059 - SoftRAID Application: Launching application: SoftRAID XT, version 5.8.4
Feb 4 1041 - SoftRAID Application: Quitting application
Feb 4 1029 - SoftRAID Driver: SoftRAID driver loaded, version 5.8.4.
Feb 4 1031 - SoftRAID Monitor: SoftRAID XT Monitor loaded, version 5.8.4.
Feb 4 1036 - SoftRAID Monitor: Starting SMART test on all disks which support SMART.
Feb 4 1052 - SoftRAID Monitor: Finished SMART test on all disks. No disks failed the SMART test.
Feb 4 1028 - SoftRAID Monitor: SoftRAID XT Monitor loaded, version 5.8.4.
Feb 4 1029 - SoftRAID Monitor: Starting SMART test on all disks which support SMART.
Feb 4 1031 - SoftRAID Driver: The SoftRAID volume "RAID5_2019A" is missing one or more disks. It has been mounted but will not offer the normal level of data protection.
Feb 4 1031 - SoftRAID Driver: The RAID volume "RAID5_2019B" (disk11) is out of sync. A rebuild has started automatically.
Feb 4 1036 - SoftRAID Monitor: Starting SMART test on all disks which support SMART.
Feb 4 1059 - SoftRAID Monitor: Finished SMART test on all disks. No disks failed the SMART test.
Feb 4 1019 - SoftRAID Application: Launching application: SoftRAID XT, version 5.8.4
Feb 4 1049 - SoftRAID Application: Adding a disk to the volume "RAID5_2019A" (disk10).
Feb 4 1050 - SoftRAID Driver: The rebuild for the volume "RAID5_2019B" (disk11) has been stopped.
Feb 4 1054 - SoftRAID Driver: The RAID volume "RAID5_2019B" (disk11) is out of sync. A rebuild has started automatically.
Feb 4 1001 - SoftRAID Driver: The rebuild for the volume "RAID5_2019B" (disk11) has been stopped.
Feb 4 1001 - SoftRAID Driver: The RAID volume "RAID5_2019B" (disk11) is out of sync. A rebuild has started automatically.
Feb 4 2240 - SoftRAID Driver: The RAID volume "RAID5_2019B" (disk11) has been rebuilt successfully.
Feb 4 2228 - SoftRAID Driver: The RAID volume "RAID5_2019A" (disk10) is out of sync. A rebuild has started automatically.
Feb 5 1027 - SoftRAID Monitor: Starting SMART test on all disks which support SMART.
Feb 5 1038 - SoftRAID Monitor: Finished SMART test on all disks. No disks failed the SMART test.
Feb 5 1233 - SoftRAID Driver: The RAID volume "RAID5_2019A" (disk10) has been rebuilt successfully.
ReplyQuote
Topic starter Posted : 06/02/2021 9:40 am
(@softraid-support)
Member Admin

A difference with hardware and software RAID when replacing disks is every volume can have a different RAID level, so disks must be added to each volume when rebuilding. It would be hard to automate this, but perhaps inside the same enclosure it is possible.

ReplyQuote
Posted : 06/02/2021 12:03 pm
Share:
close
open