Upgrade disk size of RAID 1 without loss?

RAID, Volume, storage pool, hard drive, USB, SSD cache and iSCSI LUN
User avatar
RobJVM
Posts: 10
Joined: 16 Jul 2021, 03:25

Re: Upgrade disk size of RAID 1 without loss?

Post by RobJVM »

Watching this. Please post whatever resolution you are able to get.

I have not yet attempted this upgrade because I am wary of this specific issue.
User avatar
mysticusa
Posts: 14
Joined: 30 Aug 2021, 21:20

Re: Upgrade disk size of RAID 1 without loss?

Post by mysticusa »

The instructions made it sound like very simple straightforward process. Which in theory works simple, but in practice it doesnt, and i have a feeling that it has to do with how the commands (service processes that manage the whole thing) to do this are somehow mis-configured and/or not tried/tested properly! Because i was given extra info to do is also messed up with incorrect/typo errors... also (commands) switches do no correspond to scenario...

e.g
The following example is based on "md0, vg0, lv0"
Your "array, pool, Volume" may not be "md0, vg0, lv0".
Please use the following instructions to confirm:
the md0 is relevant, but to confirm the rest? how do you confirm the rest? vg relates to (as i found out) virtual group (there is also pg (physical group i guess real raid card etc?) there is a command that supposed to show this vgdisplay isnt displaying anything... partition type show as ext4 on the system, but i removed the drive (one of the pre-upgrade drives in my case, 4tb, now 6tb s are in place) and attached to a linux, it cant read, attached to windows with linux reader, it also cant read it, i try a disk manager program to find out partition types etc to no available... some kind of trickkery going on? also there are so many partitions in place... in my case (if you see the output), 6tb drive showing 1.9tb partition ~1tb swap partition and 5.5tb data partition. have no idea how this is happening! (maybe partitions are not partitions but mounts? because total comes to 8.5tb? anyhow... i m stuck... i m offloading my media stuff to external drive (i have enclosure to use old drives) will wipe the system start freshh, sucks to do it because i have 3.5tb data will take 24 hrs to copy back and forth... (quicker than getting/arranging help)


it would be have been nice to know about warnings such as any running processes prevent live files from being migrated causing corruptions and hickups... in my case, not only raid is messed up (volume size) but also my plex is corrupted which is what i m using this for! i kind of know how to restore plex db/media works but still a lot of work/time needed!

if you dont have a means to backup your files etc, recommend you to wait... because the process is hit and miss as i found out!
User avatar
RobJVM
Posts: 10
Joined: 16 Jul 2021, 03:25

Re: Upgrade disk size of RAID 1 without loss?

Post by RobJVM »

mysticusa wrote: 31 Aug 2021, 05:17 The instructions made it sound like very simple straightforward process. Which in theory works simple, but in practice it doesnt, and i have a feeling that it has to do with how the commands (service processes that manage the whole thing) to do this are somehow mis-configured and/or not tried/tested properly! Because i was given extra info to do is also messed up with incorrect/typo errors... also (commands) switches do no correspond to scenario...

e.g
The following example is based on "md0, vg0, lv0"
Your "array, pool, Volume" may not be "md0, vg0, lv0".
Please use the following instructions to confirm:
the md0 is relevant, but to confirm the rest? how do you confirm the rest? vg relates to (as i found out) virtual group (there is also pg (physical group i guess real raid card etc?) there is a command that supposed to show this vgdisplay isnt displaying anything... partition type show as ext4 on the system, but i removed the drive (one of the pre-upgrade drives in my case, 4tb, now 6tb s are in place) and attached to a linux, it cant read, attached to windows with linux reader, it also cant read it, i try a disk manager program to find out partition types etc to no available... some kind of trickkery going on? also there are so many partitions in place... in my case (if you see the output), 6tb drive showing 1.9tb partition ~1tb swap partition and 5.5tb data partition. have no idea how this is happening! (maybe partitions are not partitions but mounts? because total comes to 8.5tb? anyhow... i m stuck... i m offloading my media stuff to external drive (i have enclosure to use old drives) will wipe the system start freshh, sucks to do it because i have 3.5tb data will take 24 hrs to copy back and forth... (quicker than getting/arranging help)


it would be have been nice to know about warnings such as any running processes prevent live files from being migrated causing corruptions and hickups... in my case, not only raid is messed up (volume size) but also my plex is corrupted which is what i m using this for! i kind of know how to restore plex db/media works but still a lot of work/time needed!

if you dont have a means to backup your files etc, recommend you to wait... because the process is hit and miss as i found out!
Thanks. Indeed. PLEX. I do keep a backup of the all the actual media files already but rebuilding all the custom meta is NOT something I would like to do if I were forced to start from scratch.
User avatar
mysticusa
Posts: 14
Joined: 30 Aug 2021, 21:20

Re: Upgrade disk size of RAID 1 without loss?

Post by mysticusa »

[root@Terra ~]# lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 0 5.5T 0 disk
|-sda1 8:1 0 285M 0 part
|-sda2 8:2 0 1.9G 0 part
| `-md9 9:9 0 1.9G 0 raid1 /
|-sda3 8:3 0 977M 0 part
| `-md8 9:8 0 976.4M 0 raid1 [SWAP]
`-sda4 8:4 0 5.5T 0 part
`-md0 9:0 0 5.5T 0 raid1 /mnt/md0
sdb 8:16 0 5.5T 0 disk
|-sdb1 8:17 0 285M 0 part
|-sdb2 8:18 0 1.9G 0 part
| `-md9 9:9 0 1.9G 0 raid1 /
|-sdb3 8:19 0 977M 0 part
| `-md8 9:8 0 976.4M 0 raid1 [SWAP]
`-sdb4 8:20 0 5.5T 0 part
`-md0 9:0 0 5.5T 0 raid1 /mnt/md0


something is odd on the outputs.

md0 shows 5.5tb but i think thats the 4tb partition from previous ones, 1.9tb md9 maybe the new space but oddly split with swap partition which is why maybe the expansion is not happening... if this is the case, there is really something wrong in the process that rebuilds the raid... space should be at the end not at the beginning as raid should commence from the beginning to end to mirror the first hdd, then extra space left at the end can be mounted/expanded to rest... just theory...

there is a volume section where it allows you to edit the size to expand to the new hdd size (as the storage pool shows correct size) but the max size it says available is 4500mb which should be ~5500 but even setting 4500 does nothing! it doesnt work! this is where the problem is internally something is not right!
User avatar
mysticusa
Posts: 14
Joined: 30 Aug 2021, 21:20

Re: Upgrade disk size of RAID 1 without loss?

Post by mysticusa »

After re-formatting everything.

[root@TNAS-5E78 /]# lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 0 5.5T 0 disk
|-sda1 8:1 0 285M 0 part
|-sda2 8:2 0 1.9G 0 part
| `-md9 9:9 0 1.9G 0 raid1 /
|-sda3 8:3 0 977M 0 part
| `-md8 9:8 0 976.4M 0 raid1 [SWAP]
`-sda4 8:4 0 5.5T 0 part
`-md0 9:0 0 5.5T 0 raid1
`-vg0-lv0 252:0 0 5.5T 0 lvm /mnt/md0
sdb 8:16 0 5.5T 0 disk
|-sdb1 8:17 0 285M 0 part
|-sdb2 8:18 0 1.9G 0 part
| `-md9 9:9 0 1.9G 0 raid1 /
|-sdb3 8:19 0 977M 0 part
| `-md8 9:8 0 976.4M 0 raid1 [SWAP]
`-sdb4 8:20 0 5.5T 0 part
`-md0 9:0 0 5.5T 0 raid1
`-vg0-lv0 252:0 0 5.5T 0 lvm /mnt/md0

-vg0 parts were missing!!! see below for the previous output. the repair/resync didnt do any good! it made the system locked to previous vol size. anyhow... i started fresh and kind of fixed the issue destructively unfortunately

[root@Terra ~]# lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 0 5.5T 0 disk
|-sda1 8:1 0 285M 0 part
|-sda2 8:2 0 1.9G 0 part
| `-md9 9:9 0 1.9G 0 raid1 /
|-sda3 8:3 0 977M 0 part
| `-md8 9:8 0 976.4M 0 raid1 [SWAP]
`-sda4 8:4 0 5.5T 0 part
`-md0 9:0 0 5.5T 0 raid1 /mnt/md0
sdb 8:16 0 5.5T 0 disk
|-sdb1 8:17 0 285M 0 part
|-sdb2 8:18 0 1.9G 0 part
| `-md9 9:9 0 1.9G 0 raid1 /
|-sdb3 8:19 0 977M 0 part
| `-md8 9:8 0 976.4M 0 raid1 [SWAP]
`-sdb4 8:20 0 5.5T 0 part
`-md0 9:0 0 5.5T 0 raid1 /mnt/md0
User avatar
RobJVM
Posts: 10
Joined: 16 Jul 2021, 03:25

Re: Upgrade disk size of RAID 1 without loss?

Post by RobJVM »

Was there a final conclusion on this? I thought I read to make sure all services (like PLEX) are stopped before the upgrade.
User avatar
RobJVM
Posts: 10
Joined: 16 Jul 2021, 03:25

Re: Upgrade disk size of RAID 1 without loss?

Post by RobJVM »

I'd like to put an end cap on this thread since I started it.

I finally did this upgrade, RAID 1 4TB --> RAID 1 6TB and had no issues whatsoever.

I did a lot of research in this forum before doing it and here are some things I learned which may or may not have helped in my success.

1. Reboot your NAS
2. Stop all services. I only had PLEX installed. Go into Applications --> click on PLEX and click "STOP"
3. After stopping all services except the TOS, DO NOT reboot your NAS again until this whole process is done, otherwise services will be automatically restarted
4. Hot-swap one of the drives in your RAID for the new larger drive. You'll get warnings, don't worry about it. You might also show an IHM warning on the drive health, again, don't worry about it. It took 9hrs for my RAID 1 to resync.
5. When that's finished, hot-swap your next drive for another new larger one. Again, another 9hrs to resync the RAID.
6. Continue until all your drives are the new bigger size (I only had two).
7. When all this was done, The Storage Pool showed the new correct size of 6TB but the Volume still only showed 4TB.
8. Click on Volume --> Edit then find the "Max" button down near the disk size, click that. BOOM. Volume resized to 6TB!
9. After this step, reboot your NAS and that should also clear the IHM warnings if you had them and your drives should show as healthy.
10. Restart your apps and services if they did not automatically start in the reboot.

Hope this helps!
Locked