NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
Solved
Volume data health changed from Redundant to Degraded on adding disk 7
Hi to the Netgear community (I am new here) I upgraded my RN104 to an RN528 (went well). Have been adding in identical disks, which went perfectly until it got to Disk7 (I had already marked disk...
You must turn off XRAID before you do this to insure it doesn't try to do anything you don't want.
The command is specific to the configuration of your NAS, but it's general form is MDADM --grow mdxxx --level=5 --raid-devices=z where mdxxx is your volume (typically md127) and z is the number of drives to include (typically, all of them). Note that if you screw this up and specify too few devices, it could kill your volume because the BTRFS volume is too big for it. If you have a multi-layer RAID, then you would have to do the same for each level, though I've never done so
When I did this the first time, I didn't specify the number of devices, and my 12-drive RAID6 became an 11-drive RAID5 with a spare and I had to add the spare after the RAID5 sync. I'm not sure if that's a safer approach, but it definately took longer with the two re-syncs.
Normally, the NAS will automatically expand the BTRFS volume when the RAID grows. But if it doesn't, follow with btrfs filesystem resize max /data (assuming volume name data).
As with any case of making changes with SSH, be sure you have a backup first. If something goes sideways, you may have to destroy and re-create the volume.
ericrobotham wrote:
I upgraded my RN104 to an RN528 (went well). Have been adding in identical disks, which went perfectly until it got to Disk7 (I had already marked disk 8 as a global spare). DIsk 7 did not automaticall get added when disk 6 finished its Sync, so I reformatted it and then it did add.
However as soon as it started to add the volume got marked as Degraded (which did not happen with any other drives). There are no disk or ATA errors.
Nov 03, 2021 09:03:09 AM Disk: Disk in channel 2 (Internal) changed state from RESYNC to ONLINE. Nov 03, 2021 08:59:56 AM Volume: Volume data health changed from Redundant to Degraded. Nov 03, 2021 08:59:54 AM Volume: Resyncing started for Volume data. Nov 03, 2021 08:59:45 AM Volume: Disk format completed on channel 2. Nov 03, 2021 02:41:14 AM Volume: Volume data is resynced.
It looks like you formatted disk 2 (not sure what you mean by disk 7). That was in the array already, so the NAS changed the status to degraded (which means a disk has dropped out of the array) and resynced.
ericrobotham wrote:
2nd question: The disks are in the wrong order (sda-d in slots 5-8, sde-h in 4-1), should I shut down and redorder disks in sda-h in 1-8)
That doesn't mean the disks are in the wrong order. When it's convenient, reboot the NAS and see what the order is. Don't reboot if the volume is syncing.
StephenB wrote:
ericrobotham wrote:I upgraded my RN104 to an RN528 (went well). Have been adding in identical disks, which went perfectly until it got to Disk7 (I had already marked disk 8 as a global spare). DIsk 7 did not automaticall get added when disk 6 finished its Sync, so I reformatted it and then it did add.
However as soon as it started to add the volume got marked as Degraded (which did not happen with any other drives). There are no disk or ATA errors.
Nov 03, 2021 09:03:09 AM Disk: Disk in channel 2 (Internal) changed state from RESYNC to ONLINE. Nov 03, 2021 08:59:56 AM Volume: Volume data health changed from Redundant to Degraded. Nov 03, 2021 08:59:54 AM Volume: Resyncing started for Volume data. Nov 03, 2021 08:59:45 AM Volume: Disk format completed on channel 2. Nov 03, 2021 02:41:14 AM Volume: Volume data is resynced.
It looks like you formatted disk 2 (not sure what you mean by disk 7). That was in the array already, so the NAS changed the status to degraded (which means a disk has dropped out of the array) and resynced.
I meant disk 7 as in SDG (the 7th in the array, which is in slot 2, confusing bcause of the order the array is built, see not on disk order), normally a new disk chnges from formatted, reysnc stats and disk gets marked online (not marking the array as degraded, no disks have failed and no errors), here are the rest of the logs (with the snapshots removed)
Nov 03, 2021 09:03:09 AM
Disk: Disk in channel 2 (Internal) changed state from RESYNC to ONLINE.
Nov 03, 2021 08:59:56 AM
Volume: Volume data health changed from Redundant to Degraded.
Nov 03, 2021 08:59:54 AM
Volume: Resyncing started for Volume data.
Nov 03, 2021 08:59:45 AM
Volume: Disk format completed on channel 2.
Nov 03, 2021 02:41:14 AM
Volume: Volume data is resynced.
Nov 02, 2021 10:31:33 AM
Volume: Successfully made disk on channel 1 a global spare disk.
Nov 02, 2021 10:29:58 AM
Disk: Disk Model:WDC WD30EFRX-68AX9N0 Serial:WD-WCC1T0820766 was added to Channel 1 of the head unit.
Nov 02, 2021 06:58:02 AM
Disk: Disk in channel 3 (Internal) changed state from RESYNC to ONLINE.
Nov 01, 2021 10:07:09 AM
Disk: Disk in channel 4 (Internal) changed state from RESYNC to ONLINE.
Nov 01, 2021 10:02:33 AM
Volume: Disk format completed on channel 3.
Nov 01, 2021 10:02:12 AM
Volume: Resyncing started for Volume data.
Nov 01, 2021 10:02:02 AM
Volume: Disk format completed on channel 4.
ericrobotham wrote:
2nd question: The disks are in the wrong order (sda-d in slots 5-8, sde-h in 4-1), should I shut down and redorder disks in sda-h in 1-8)
That doesn't mean the disks are in the wrong order. When it's convenient, reboot the NAS and see what the order is. Don't reboot if the volume is syncing.
I meant moving the disks so SDA in slot 1, SDB in slot 2, so the disk numbers and identifiers are in the same order.
ericrobotham wrote: I meant moving the disks so SDA in slot 1, SDB in slot 2, so the disk numbers and identifiers are in the same order.I know.
The drive names (SDA, etc) are more dynamic than you are thinking. When you add/remove disks, the system will assign different drive names to the added drives. So if the old drive in slot 1 is SDA (for example), the system might label the new drive in that slot as SDH and not use SDA at all. When you reboot the NAS, the drive names will revert to using SDA for slot 1, etc.
That is why I said the drive names you are seeing don't mean that the array is out-of-order.
FWIW, "out of order" generally refers to the RAID array itself (since each disk in the array is striped differently). The array will normally be assembled correctly if the drives are out-of-order in this sense, but some troubleshooting becomes more complicated. So I don't normally recommend shuffling the drives.
Thanks StephenB, I will leave the drives in the current order.
Leaves the question, Why did it marked as degraded though, as an expansion / reysnc shouldnt do that, from my previous experience (and didnt with the other disk additions on this system, or on any others I have done on other readynas systems)
When you add the 7th drive and are in XRAID, it converts to RAID6. So, your volume was a degraded RAID6 because it only had one of the two drives for RAID6 redundancy. Unfortunately, I think you expected it to expand as RAID5, which would have required you to switch to FlexRAID and specify that. That it will do that is in the documentation, but I have recommended that there be some kind of GUI pop-up that at least warns you and lets you proceed or abort, or better yet, choose RAID5 expansion or RAID6 conversion.
There is a way to convert back to RAID5 via SSH if you feel comfortable doing that. You'll have no redundancy during the conversion, so you should have a backup of the data before you try. Let me know if you need the commands.
Just found this in the logs (search on Raid Level in rn-expand.log)... {its called Eric's_NAS104 because otherwise I would have to redo evey MAC timemachine backup, as it does not allow you to create volume named drive, with an existing backup on it}
Nov 01 10:02:03 Erics-NAS104 rn-expand[5996]: Reshaping array to RAID level 5, 5 disks...
Nov 02 06:47:08 Erics-NAS104 rn-expand[5996]: Reshaping array to RAID level 5, 6 disks...
Nov 03 08:59:46 Erics-NAS104 rn-expand[5996]: Reshaping array to RAID level 6, 7 disks...So it looks like the 7th disk caused a RAID strategy change from RAID 5 to RAID 6 (meaning, I guess, that when fully working it has a 2 disk redundancy rather than 1, so degraded for a RAID 6 [but not a RAID 5]).
Doesnt this also mean I have wasted disk 8 by making it a global spare, as there is now a 3 disk redundancy in the system...
ericrobotham wrote:
Doesnt this also mean I have wasted disk 8 by making it a global spare, as there is now a 3 disk redundancy in the system...
Another option (easier than what Sandshark offers) is to add disk 8 to the array.
FWIW, you don't have more redundancy with a spare (that is, you do not have 3 disk redundancy now). The spare just speeds up the disk replacement process. Personally I'd rather not have the system automatically deploy a spare - I'd rather take care of that manually. Though opinions on that likely vary.
NETGEAR Academy
Boost your skills with the Netgear Academy - Get trained, certified and stay ahead with the latest Netgear technology!
Join Us!
