Search results

You can create crush rules to make localized reads

My opinion: If you use system without data checksum - it is not critical system.

If you use ZFS try snapshots. If you want to use "mountpoint" in VM you can try 9p mount

If your `zpool status` report "Some supported and requested features are not enabled on the pool." then your pool is not upgraded. Or you can check in `zpool history` did upgrade was initialized. Or you can check in `zpool get all pool` enabled features and compare to documentation what was new...

Try to load pool in read-only mode with ZFS 2.2 and check the version.

Did you upgraded pool version ?

I use 12 disks in 2x Raidz2 of 6 combination. Doing copy 40GB file from ramfs to ZFS `zpool iostat` report as 600-1300 MiB/s. Using pv tool report as 622 MiB/s average. Using dd with bs=200M : 828 MB/s I suggest you to troubleshoot with tools like atop or others who could show disk activity...

Try to use lower version of ZFS. Maybe it will help. I stuck to kernel 6.2.16-18-pve and ZFS 2.1.13-pve1 and I`m not sure then I`ll upgrade to newer kernel.

As I was told enterprise version is the same as old community version. To keep track of package version go to https://pve.proxmox.com/wiki/Downloads and find '# pveversion -v' list But you will have to update it manually to needed version.

One I had a problem / unknow error with ZFS. It started then I wanted to clean old data, old snapshots. ZFS started to hang up. Nothing helped. Tried to import the pool with -N - still the same. I did not know what ZFS was trying to do. Only importing pool in read-only mode allowed to see...

ZFS is not suitable to run on single disk then problems. Specially for system OS partition. If you could put 2 disks for ZFS mirror to handle this type of errors.

My server did the same ( or at least I think so ). I use Corsair Commander Pro to control the speed of ventilators. After speeding it up CPU is not getting hot and do not do auto system reboot by hardware. Try to limit CPU speed if you cannot cool your NUC.

If you want to know then yes, keep it on. In the same situation others FS will continue to operate. But keep in mind your VM will be broken or you movie will be half way watchable.

I will ask a question for thought. If recordsize or volblocksize is set to 1M, what is the minimum read IO for that block even need part of it only? For example MariaDB with 16K database storage on recordsize/volblocksize with 1M. Will ZFS read 16K of request or all 1M of the block before...

This thread started as performance comprising between native ZFS raid and ZFS on top of H.Raid and half leg is it good to run ZFS on H.Raid. I can tell you this - How you will set up that's way it will work. If you want H.Raid to monitor HDD, I can suggest to set ZFS checksum=off otherwise ZFS...

How HWRAID protect from data distortion or bit rot ?

I start this software #top -f -F 1 And then with "L" I change to see all disks and CPU. In the benchmark time you can see disk activity and you can compare do all disk are busy equally and what is CPU activity.

ZFS was not designed to allow remove disk. It is very complex system. Yes you can remove disk from the mirror. Maybe one day it will be possible.

You cannot remove disk in your situation right now. Plan how to backup your data and recreate ZFS.

ZFS is very complex system. It is COW system. If you don't care compression, encryption, snapshot, data integrity then use old file systems. Use #atop to see CPU and disk usage. Maybe it will show something interesting.