February 11, 20242 yr Hello everyone, I have the following problem: All copy or write operations to and from the encrypted formatted ZFS cache pool are incredibly slow. Maximum 60MB/s. Even, as you can see in this screenshot, when I copy from SSD to SSD. The IOwait is then also high. I would like to stay with ZFS because of the storage efficiency, but the speed is ridiculous. I've read about a problem of Unraid when using ZFS for the array itself but not when using it for pools. Could someone give me a hint? Thanks in advance! dringenet-ms-diagnostics-20240211-1113.zip
February 11, 20242 yr Community Expert 1 hour ago, pappaq said: I've read about a problem of Unraid when using ZFS for the array itself but not when using it for pools. That doesn't affect pools: NAME STATE READ WRITE CKSUM cache ONLINE 0 0 0 raidz1-0 ONLINE 0 0 0 sdb1 ONLINE 0 0 32 sdd1 ONLINE 0 0 32 sdi1 ONLINE 0 0 32 sdh1 ONLINE 0 0 32 Both btrfs and zfs are detecting data corruption, start by running memtest.
February 11, 20242 yr Author I've noticed data corruption on the ZFS pool but where did you see the corruption for the BTRFS pool? In the UI it does not show any errors. I will run memtest as soon as I've got the time for it...the system is completely new 😑
February 11, 20242 yr Author Scrubbed the pool again and the errors are gone, for now...the scrub speed was very good though. The speed is slow anyways and the iowait is super high when there is traffic on the zpool...
February 11, 20242 yr Community Expert 2 hours ago, pappaq said: where did you see the corruption for the BTRFS pool? Feb 10 18:43:33 dringenet-MS kernel: BTRFS info (device dm-15): bdev /dev/mapper/23392J803394 errs: wr 0, rd 0, flush 0, corrupt 7, gen 0 You can take a look below for more info: https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=700582
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.