I have a TS-h1277AXU-RP currently running QuTS hero h5.2.6.3195 Build 20250715. I mistakenly configured the machine with the HDD pool as the system drive, but I would prefer to use the M.2 SSDs that I have. The HDD pool is already quite full of data, how can I go about making the SSDs the system drive? Could I detach the HDD pool and re-initialize with only the SSDs installed, and then reattach the HDD pool?
Our internal team will try to test and determine the best approach.
However, could you please provide us with more detailed information about your setup, such as how many SSDs or HDDs you have? If possible, please also provide a screenshot of your Storage Manager. Thanks!
5 HDDs, 2 M.2 SSDs currently configured as a cache.
Iâm assuming my best path forward would be to add more drives, create a new pool with the drives, move the data to that pool, and then re-initialize?
In terms of screenshot, does this suffice or do you need more info?
Right, sorry, I wasnât explicit enough. I would move the data to the new pool and then detach the pool so that nothing happens to the data. At that point I can re-initialize with the SSDs and the original HDD pool still installed, then reinstall/reattach the new pool.
We currently donât have backups. The reason why I need to do this is to try and speed up the applications. I read from someone else somewhere that they had better replication/restore performance when using SSDs as the system drive, and I need that performance boost with 17TB of data to move.
Messing around with the drive configuration without backups is not a terribly great plan. Even if it takes a long time, Iâd suggest performing this critical task first.
Remember the OS is on the drives, so I would not do anything with internal disks (hence the backups!)
Otherwise you might import your problems right back into the new system upon insertion of that detached volume (it âshouldâ not, but it has happened)
The problem is that when you insert drives/create a pool, the OS gets written on to them. They have memory of the configuration. You need to backup to something completely off the NAS (ie: an external HD, a different NAS or the cloud.
As soon as you insert those drives back in after you set everything back up, itâs going to cause an issue. When you âstart overâ, you need to completely start over. Itâs a pain and it takes a long time. I have done it more than once.
Ok, so to clarify, any new pool (Pool2) I create to act as a backup target could create issues when I restore my data from it after the re-initialization? Even though itâs not the original system drive pool (Pool1) which will get re-initialized? Doesnât that kind of negate the point of the NAS migration feature?
Maybe this will clear it up a bit: This is a 12-bay NAS. The current system drive is on a pool made up of 5 10TB drives (Pool1). I have some 16TB drives I can use in other bays for a new pool (Pool2) which I can then copy my data to. I can then detach Pool2 and physically remove those drives. Then re-initialize with the SSDs and Pool1 still installed, and re-attach/install Pool2 once thatâs complete. Then move the data back over to the new Pool1.
Thatâs why I said it âshouldâ work, but these things have gone wrong in the past, so doing it via and external disk (to also keep your backups) would be the best solution
And I think itâs just easier to have all the data external to the NAS. That way, you can set up everything the way you want it at once and then start moving your data back into place. It just works much better that way and itâs also easier.
Iâm paranoid enough that I went and pulled another NAS from almost-production. I will install the 16TB drives in that one, get everything backed up, and then perform the re-initialization. Relatedly, Iâve had some issues setting up RTRR, hopefully it works in this case (theyâre patched into the same switch now). I saw a post from 7 years ago where people were having the same issue I am having where authentication of the RTRR account fails for some reason.