With the conditions for editing RAID, the edit button of the storage pool is still gray and cannot be clicked?

[TMzethar]

Troubleshooting

Applicable models
All TNAS models

Applicable version
TOS 4.1.x ~ TOS 4.2.x

Cause Analysis
Use a large enough hard disk to replace the failed hard disk or add it to the NAS, but the edit button in the storage pool is grayed out, so the new hard disk cannot be restored or added to the RAID. The reason may be that the capacity of your hard disk is slightly different from other hard disks. Therefore, the system judges that the hard disk does not match the other hard disks in the RAID, and the hard disk cannot be added to the raid.

Solution
1. Go to Control Panel> Storage Pool, and create a new Single type storage pool with your new hard drive
2. Delete the newly created storage pool
3. Log in to the SSH terminal and switch to the root user How to log in to the SSH terminal?
4. Execute the following commands to observe the relationship between RAID and each hard disk:

Code: Select all

lsblk

5. Finally, execute the following command to add the new hard drive to the RAID:
"sd[a-z]4" represents hard disk partition;
"a-z" represents the serial number of the hard disk, and the serial number of the newly added hard disk needs to be determined;
"4" represents the fourth partition, that is, the key data partition;
"md0" stands for storage pool 1. If you want to edit storage pool 2, enter "md1", and so on.

Code: Select all

mdadm -a /dev/md0 /dev/sd[a-z]4

[vlad]

I have the same problem, I cannot add the 5th disk to a 4 disk RAID6. But my drives are the same model and the exact same size, so this is clearly a TOS bug:

Disk /dev/sda: 16.4 TiB, 18000207937536 bytes, 35156656128 sectors
Disk /dev/sdb: 16.4 TiB, 18000207937536 bytes, 35156656128 sectors
Disk /dev/sdc: 16.4 TiB, 18000207937536 bytes, 35156656128 sectors
Disk /dev/sdd: 16.4 TiB, 18000207937536 bytes, 35156656128 sectors
Disk /dev/sde: 16.4 TiB, 18000207937536 bytes, 35156656128 sectors

The first 4 disks form the RAID 6. The 5th I just inserted. The edit button remains grayed out. I even rebooted the system, losing the 150 days of uptime, the edit button is still grayed out.

[vlad]

Also, the instructions are incorrect. When trying to add the 5th disk to a RAID6 pool with that procedure it just added it as a spare, it will not actually reshape the RAID:

Code: Select all

md0 : active raid6 sde4[5](S) sda4[0] sdd4[4] sdc4[2] sdb4[1]
      35149936640 blocks super 1.2 level 6, 64k chunk, algorithm 2 [4/4] [UUUU]
      bitmap: 0/131 pages [0KB], 65536KB chunk

Notice the (S) near sde4, meaning it's a spare. The pool actually still consists of only 4 disks after following these instructions, notice the [4/4] and [UUUU] meaning 4 disks, not 5.

In the end, I fixed the problem myself by running this additional command:

Code: Select all

mdadm --grow /dev/md0 --raid-devices=5

And now it looks like this:

Code: Select all

Personalities : [linear] [raid0] [raid1] [raid10] [raid6] [raid5] [raid4] [multipath]
md0 : active raid6 sde4[5] sda4[0] sdd4[4] sdc4[2] sdb4[1]
      35149936640 blocks super 1.2 level 6, 64k chunk, algorithm 2 [5/5] [UUUUU]
      [>....................]  reshape =  0.0% (335136/17574968320) finish=5243.7min speed=55856K/sec
      bitmap: 0/131 pages [0KB], 65536KB chunk

If anyone else hits this problem, you may need to specify a different number for the mdadm command, it should match the number of disks from your pool.

Also, TerraMaster, why are you not fixing this bug? Not everyone has enough Linux experience to workaround your bugs. While you are at it, you may want to fix the auto-update feature, that claims that 4.2.40 is the latest TOS version, when in fact TOS 5 is a more recent version.

When I bought this F5-422 earlier this year TOS 5 was already available, and I would have installed it, if the auto-update feature would have prompted me to do that. Now, that I already have data on it, I'm reluctant to do a manual update to TOS 5 since the auto-update functionality is not working.

Speaking of bugs, you also have a 200 MB/s speed limit for RAID syncs, dev.raid.speed_limit_max. The problem is that this not configurable anywhere in the user interface to a higher value, and my drives can do up to 300 MB/s during some phases of the RAID sync, so your limit slows down the RAID sync.

I had to increase it manually 400 MB/s by setting dev.raid.speed_limit_max in sysctl.conf:

Code: Select all

dev.raid.speed_limit_min = 200000
dev.raid.speed_limit_max = 400000

then running

Code: Select all

sysctl -p

The problem is that there is some TOS process that keeps resetting the limit back to 200 MB/s periodically. I found a TOS script with those limits,

Code: Select all

tos/scripts/mntdata:    sysctl -w dev.raid.speed_limit_max=200000
tos/scripts/mntdata:    sysctl -w dev.raid.speed_limit_min=100000

I edited those to 200000 min and 400000 max instead, but it still ignores that, and changes the speed back to the lower values every couple of hours or so.

Finally, out of desperation, I added a cron job to set them to the proper values every minute:

Code: Select all

* * * * * /sbin/sysctl -p > /dev/null 2>&1

Only to find out today, after rebooting, that the cron job has been deleted. Now I added it back again.

Needless to say, for a device that is quite expensive, having to do all this manual work, fighting it at each step to make it work properly, is not acceptable.

I lost hours due to the TOS bugs, please fix them so that other users don't have to fight these issues again in the future. Thanks.