ZFS "problem"
- Thread starter ivans89
- Start date
It depends on how you do it. With enough free space you could copy every dataset/zvol one after another and delete the old one afterwards. That way you don't lose your data. Or you copy all data to another pool that can fit your 120TB, destroy everything and later move it back.
There are multiple ways to do this, but you will have to rewrite everything.
puh...i have 50TB free on my 120TB pool.
So this will take a few days?
I think maybe a better way is just to download/unpack on a fast ssd and then move the files to hdd.
So this will take a few days?
I think maybe a better way is just to download/unpack on a fast ssd and then move the files to hdd.
Yes, probably.
Always better to spend some time researching before ordering hardware and start using a pool. That saves a lot of money and time.
Could be the same pool. But needs to be a new dataset/zvol so it gets copied and not simply moved.
Last edited:
One thing for understanding.
I have mounted my pool unter /home/media
In this media folder I have this:
Folder1 - in this folder 20TB
Folder2 - in this folder 20TB
Folder3 - in this folder 20TB
Now I create the special disk. I can just copy each folder into a new folder in the same pool and after that, I can move it back? That’s it or is there a special command that I have to use?
You need to move them between Datasets. In case you didn't create datasets for each folder via CLI this wont help.
Reorganizing your pool and start working with datasets. Then moving files from your folders to your datasets.
You should really have a look at the ZFS basics and try to undestand the concepts like datasets/zvols/vdevs/recordsize/compression/replication/snapshots and so on to be able to make best use of your storage.
You should really have a look at the ZFS basics and try to undestand the concepts like datasets/zvols/vdevs/recordsize/compression/replication/snapshots and so on to be able to make best use of your storage.
Last edited:
Ok thank you.
I start adding the special device to the pool.
After that i creating datasets (now i know what it is
).
After that i will move my folders into the datasets and check if special devices works properly.
I start adding the special device to the pool.
After that i creating datasets (now i know what it is
). After that i will move my folders into the datasets and check if special devices works properly.
I deleted my Pool and want to start a fresh one. Now im thinking of raidz2 or raid10. How about special devices for raid10? how many i need minimum? 2?
And what if i run raidz1 with 2vdevs?
Last edited:
I have created now raidz1 with 2 vdevs (2x4) and special device with 2x1TB (will add here 2x 1TB next week when new ssd arriving). And will add a extra vdev if my space run out.
For now i can tell, it's MUUUUUUUUCH better. Before that my disk io was all the time at 95-100%. Now at the same tasks it's at around 15-20%.
For now i can tell, it's MUUUUUUUUCH better. Before that my disk io was all the time at 95-100%. Now at the same tasks it's at around 15-20%.
Last edited:
This is very crucial. You defenitely want to have a lot of datasets to organize different data you store so that you have fine-granular control over snapshots.
Raid10 for performance and faster disk replacement and raidz2 if you care about your data and capacity. Striped raidz1 is somewhere in between.
2/4/6/... for a raid10 and 3/6/9/... for a raidz2. When a raidz2 allows you to lose any 2 HDDs it would be bad if you would only got two special device SSDs in a mirror so the whole pool would be lost as soon as the second SSD starts failing.
Last edited:
a not so fancy way is to clear up you ZFS cache when you know you need them or per script. But then would hurt your read speed when reading you data from the pool and its not in the ram... But you can add NVME storage as read cache. I guess better than reading from HDD.
No need to delete any cache.
With this setting its running without any problems anymore.
With this setting its running without any problems anymore.
Code:
Plex 132T 13.1T 119T - - 0% 9% 1.00x ONLINE -
raidz1-0 65.5T 5.86T 59.6T - - 0% 8.95% - ONLINE
sde 16.4T - - - - - - - ONLINE
sdf 16.4T - - - - - - - ONLINE
sdg 16.4T - - - - - - - ONLINE
sdh 16.4T - - - - - - - ONLINE
raidz1-1 65.5T 7.23T 58.2T - - 0% 11.0% - ONLINE
sdi 16.4T - - - - - - - ONLINE
sdj 16.4T - - - - - - - ONLINE
sdk 16.4T - - - - - - - ONLINE
sdl 16.4T - - - - - - - ONLINE
special - - - - - - - - -
mirror-2 888G 14.4G 874G - - 1% 1.62% - ONLINE
sdn 894G - - - - - - - ONLINE
sdo 894G - - - - - - - ONLINE