Jump to content

Cache Pool Upgrade


Recommended Posts

I currently have 2 x 240gb SSDs as a RAID 1 cache pool. I am trying to replace these with 2 x 480gb SSDs, and I have tried to follow the instructions here:

 

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/#comment-480419

 

Upon selecting the 480gb drive as one of the 2 cache drives and starting the array, the pool devices look like this:

image.thumb.png.fade3e16fc7cef4991e8349bdb4ec6bf.png

 

and there is an option at the bottom to format all unmountable disks (which seems to be both devices currently selected as cache drives, sdg and sdi)

 

I'm not sure what I'm doing wrong. I also really don't want to copy all the data from my cache pool to the array if i can help it since there is a lot of very small files (plex media library info and minecraft servers, etc.) which would take an extremely long time to copy.

 

EDIT: upon noticing the disclaimer that this isn't working in unraid 6.9.x, i read through that post in which someone suggested instead removing one drive without adding a new one first, however i am unable to start the array unless I have 2 drives selected (start button is greyed out).

any other suggestions?

Edited by CmdrKeen
Link to comment
5 minutes ago, jonathanm said:

Follow the directions for completely replacing the cache, by using the parity array as a temporary location for the data using the mover.

I would prefer to not transfer everything to the array as the last time i needed to transfer my plex media library info to/from spinning disk storage it took over a day due to the large number of small files (even though the total size isn't that large). ideally going from SSD -> SSD would be much faster.

Link to comment
17 minutes ago, jonathanm said:

CA Backup / Restore?

 

I suppose you could set up another pool since you are on 6.9.2, if everything is using /mnt/user paths you wouldn't need to change anything, just copy from one pool to the other.

I am using default mounting locations (i believe this should be /mnt/user/...)

so the process would be;

- stop docker & VM services

- stop array

- create a 2nd cache pool of my 2 x 480gb SSDs

- start array

- copy all files from original cache pool to 2nd cache pool (I'll need to figure out how to do this. I guess i can use the command line since i can't use krusader)

- remove original cache pool...?

 

I'm relatively confident in the first part but I'm a little unsure of the last few steps.

 

My appdata share is set to be only on my cache drives for performance reasons. this is where my plex data relies. I'm not entirely sure how to copy data from one cache pool to the other since the /mnt/user/ directory only has shares in it.

Edited by CmdrKeen
Link to comment

If you copy you don't technically need to do the CA backup, but it would be a good idea to have set up anyway.

 

You can use mc at the console for copying, it's almost as easy as krusader, it's a double pane file explorer, just text based.

 

I haven't played with the multi pool feature yet, but I suppose you could probably remove all the pool drives after you made the copy, then add back just the 2 new drives to the original pool and it should work. IIRC, as soon as you start the array with no pool devices, you can then add previously configured devices into the pool and they will retain their data. As long as you don't see the ominous

1 hour ago, CmdrKeen said:

format all unmountable disks

warning you should be ok.

Link to comment

So i copied all data from /mnt/cache to /mnt/cachenew

then i stopped the array and renamed "cache" -> "cacheold" and renamed "cachenew" -> "cache"

then i started the array and went to the chares tab. it auto-selected the new "cache" as what it was using for the appdata share (which is my only share set to remain on cache).

i confirmed by starting a new blank minecraft server just to add a new folder through a docker. it is indeed using this new cache

 

my only issue is there seems to be a descrepency of about 20gb between the old cache and the new one, and i can't figure out why this is.

Link to comment
9 minutes ago, CmdrKeen said:

my only issue is there seems to be a descrepency of about 20gb between the old cache and the new one, and i can't figure out why this is.

 

How did you do the file copying? Perhaps your docker.img file lost its sparseness during the process and got expanded to its full size. If I'm copying sparse files I always use cp with the --sparse=always option or rsync with the --sparse option.

 

Link to comment
57 minutes ago, John_M said:

 

How did you do the file copying? Perhaps your docker.img file lost its sparseness during the process and got expanded to its full size. If I'm copying sparse files I always use cp with the --sparse=always option or rsync with the --sparse option.

 

I used "cp -R /mnt/cache/* /mnt/cachenew/"

when i noticed the descrepency i went over it again with rsync -ah but got the same result

 

I'm not sure what sparseness means

Link to comment

No. Sparse files grow as they fill up, but they never become smaller if you delete their contents. If it did lose it's sparseness and inflated to it's true size you can re-sparsify it by copying it again, with the --sparse=always option (cp is supposed to be good at automatically detecting spare files but I've known it get it wrong on several occasions), or in the case of a docker.img, just delete it and re-build it. The Previous Apps section of Community Apps makes that very easy.

Link to comment
8 hours ago, CmdrKeen said:

I currently have 2 x 240gb SSDs as a RAID 1 cache pool. I am trying to replace these with 2 x 480gb SSDs, and I have tried to follow the instructions here:

 

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/#comment-480419

Didn't you see the warning in red letters that's it's broken in v6.9.x?

 

You just need to stop array, unassign both cache devices, start array so Unraid "forgets" cache assignments, re-assign both the OLD (original) cache devices, start array and and the original pool should mount.

Link to comment
15 hours ago, JorgeB said:

Didn't you see the warning in red letters that's it's broken in v6.9.x?

 

You just need to stop array, unassign both cache devices, start array so Unraid "forgets" cache assignments, re-assign both the OLD (original) cache devices, start array and and the original pool should mount.

you evidently didn't read my post. I DID see the warning message as i said at the bottom. 

additionally, the array would not start (the START button was greyed out) unless 2 cache drives were selected. not sure why this was the case.

 

I did manage to get everything moved over and working great thanks to jonathanm

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...