Replies: 1 comment 4 replies
-
|
Beta Was this translation helpful? Give feedback.
4 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hardware: Raspberry Pi 4 Model B Rev 1.4 - 8 GB RAM
OS: Ubuntu Server 20.04.3 LTS
Kernel: Linux fs 5.4.0-1045-raspi #49-Ubuntu SMP PREEMPT Wed Sep 29 17:49:16 UTC 2021 aarch64 aarch64 aarch64 GNU/Linux
The currently installed zfsutils-linux version is 0.8.3
I'm setting up ZFS on a fresh install of Ubuntu Server for Raspberry Pi. It's a currently just a test setup. There's no production data at risk, so I can completely wipe and restart with a fresh install as needed.
I'm setting up an 8 x 6 TB raidz2 pool using 6 TB SATA drives in (2) four-bay SATA/USB3 docks.
I have no problem creating the pool using /dev/sda .. /dev/sdh
Once created, I am able to read/write some test data - disk benchmarking with sysbench, for example.
The physical drives don't get consistently assigned to /dev/sd* device files across reboots, so I want to have disks assigned to the pool by serial number, so I have a custom udev rule:
/lib/udev/rules.d/10-local.rules
KERNEL=="sd[a-h]", SUBSYSTEM=="block", PROGRAM="get_disk_serial.sh %k", SYMLINK+="disk/by-serial/%c"
At boot /dev/disk/by-serial looks like:
I should be able to export pool1 and then reimport using the links in /dev/disk/by-serial. I get the following:
Any ideas on how to troubleshoot/solve this. At first glance, this looks like it might be a problem with the zpool command handling the -d option, rather than some issue with how the physical disks are linked in /dev/disk/by-serial. I could be wrong.
Beta Was this translation helpful? Give feedback.
All reactions