Soon™️ 6.12 Series


starbetrayer

Recommended Posts

29 minutes ago, jonathanselye said:

what is the difference of automatically created zfs share vs manually created zfs with the same name on the disk

Do you mean creating a share using the GUI or manually with for example mkdir? If yes the former will create a zfs dataset and the latter not, though both will work as shares, but recommend using the GUI, datasets can be snapshoted, etc.

Link to comment
43 minutes ago, JorgeB said:

Do you mean creating a share using the GUI or manually with for example mkdir? If yes the former will create a zfs dataset and the latter not, though both will work as shares, but recommend using the GUI, datasets can be snapshoted, etc.

Oops sorry for the confusion, i mean create the ZFS using ZFS Master plugin(does that count as GUI?) then put data into it will UNraid treat it the same as the auto created ZFS when making a new share? i plan on migrating to zfs from xfs moving data from disk one by one by creating the zfs via ZFS Master then moving data into it i tested on a test server when making a share unraid automatically creates a ZFS share and i tried also making a ZFS via ZFS master in my test server on another disk where the share does not exist yet, the SHFS seems to work fine on the auto created and the manually created ZFS i just want to make sure the data integrity and if it is fine specially, im migrating a lot of data.

Edited by jonathanselye
Link to comment
52 minutes ago, jonathanselye said:

Oops sorry for the confusion, i mean create the ZFS using ZFS Master plugin(does that count as GUI?)

I've never used the plugin, but creating any top level dataset should create the respective share, nested datasets won't be treated as shares, but unless there's a good reason recommend creating the shares with the GUI instead.

Link to comment
1 minute ago, JorgeB said:

I've never used the plugin, but creating any top level dataset should create the respective share, nested datasets won't be treated as shares, but unless there's a good reason recommend creating the shares with the GUI instead.

apologies for the dumb question, but how does one create on a certain disk via GUI?(when creating via share it only exist on one disk)

Link to comment
3 minutes ago, jonathanselye said:

apologies for the dumb question, but how does one create on a certain disk via GUI?(when creating via share it only exist on one disk)

It will be automatically created when needed, i.e., when writes for that share are first allocated to that disk.

Link to comment
14 minutes ago, JorgeB said:

It will be automatically created when needed, i.e., when writes for that share are first allocated to that disk.

I wish to transfer data from eg. disk1 to disk2 on xfs this can be simply done via unbalance in the new zfs array how can this be done if the share(zfs folder) still does not exist on the destination disk? is there a way to somehow force gui to create the zfs empty share on the destination disk?

Link to comment
3 minutes ago, JorgeB said:

If you use Unbalance it *should* also create the new datasets as needed, you could force by setting the share to temporarily include only the disk you want, then copy something to that share or just do

touch /mnt/user/share_name/a

then delete a.

Tried with unbalance copy/move the folder created seems to be not recognized as a ZFS file system, but now i understand it how dumb of me to forgot to force the share to only the disk i want thanks its all clear to me now.

One last question i know you are not allowed to release more info yet on the public but is it safe to migrate my data now? or better wait for the stable release will there be more changes to the zfs? i will understand if you are not allowed to release information yet.

Thanks for the great help you answered my question perfectly!

Link to comment
17 minutes ago, jonathanselye said:

or better wait for the stable release will there be more changes to the zfs?

There *shouldn't* be any changes to how zfs is implemented that would make any zfs storage created after v6.12-beta5 incompatible with future releases, and even zfs created with beta5 or earlier still works, just won't have the optimal settings.

  • Like 1
Link to comment
22 minutes ago, JorgeB said:

There *shouldn't* be any changes to how zfs is implemented that would make any zfs storage created after v6.12-beta5 incompatible with future releases, and even zfs created with beta5 or earlier still works, just won't have the optimal settings.

Sweet Thanks! I'm so excited with the stable release.

Link to comment

Hello regarding ZFS i want to ask a performance related concern

What would be the best cache-pool setup:

I have two samsung 980 pro 2tb, should i go with, i prefer to get the most performance and data protection is not the priority

a.) seperate the load, 1st 980 dedicated to VM, Docker 2nd 980 as  array cache for *Arrs downloads

b.) put them both and raid0

i am afraid that maybe b.) scenario will increase my acess time and latency so i am currently using a.) i want to be enlightened 

Thanks!,
Dhen

Link to comment
13 minutes ago, JorgeB said:

This

 

 

 

I'm sorry i forgot to mention i am transcoding 24/7 using TDarr and it seems after around 24 hours i max out the second ssd's cache(the ssd itselfs cache not unraids) and put the performance of the second ssd to a crawl will the raid0 system suffer the same? or will it solve this issue?

 

Thanks!,

Dhen

Edited by dhendodong
Link to comment
  • 2 weeks later...

Just wanted to check is this scenario enabled in the new 6.12?

 

Can we configure mover to work differently for different shares?

 

- For example I'm still working out a way to have new folders on cache copied to the pool drives everyday day or once a week.

- But maintain the data downloads cache pool (difference cache pool called TWO) to remain on the SSD cache as much as possible (to avoid spinning up disks where possible)

 

So I want data downloads to keep say 1tb of data on the 2tb cache pool TWO. If I use 'prefer' it will fill up the drive and thus not be able to download any new data.

 

If I set mover to move everyday it will move all the new cache folders from my other shares but leave my data downloads cache pool empty which is a waste of the SSD pool.

Edited by dopeytree
Link to comment
3 hours ago, dopeytree said:

'when share on cache/s (primary) reach 75% capacity' = move 'anything over 90days old' to array (secondary). { but never move: APPDATA, DOMAINS, SYSTEM } 

 

What happens if cache is 90%+ capacity and everything is less than 90 days old?

Link to comment
19 hours ago, dopeytree said:

Perhaps a script to do this.. 

e.g

'when share on cache/s (primary) reach 75% capacity' = move 'anything over 90days old' to array (secondary). { but never move: APPDATA, DOMAINS, SYSTEM } 

 

Have you looked at the mover tuning plugin? That offers a lot of options like you are talking about

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.