ZFS pool is empty after crash – Managing your servers can streamline the performance of your team by allowing them to complete complex tasks faster. Plus, it can enable them to detect problems early on before they get out of hand and compromise your business. As a result, the risk of experiencing operational setbacks is drastically lower.
But the only way to make the most of your server management is to perform it correctly. And to help you do so, this article will share nine tips on improving your server management and fix some problem about linux, zfs, , , .
I’m on Linux 3.6.11 with ZFS on Linux 0.6.0rc11.
I have a large ZFS pool with three subpools:
zpool1 zpool1/backups zpool1/graphite zpool1/log_backups
The machine rebooted last night, for no obvious reason. After I was paged and got a shell it was already back up.
zfs mount -a to remount the pools.
zpool1/log_backups has data in it, but
zpool1/backups are completely empty, and
df reports that there’s no space in use on those partition.
Unfortunately I didn’t have snapshots set up, so that’s a no go. I’ve tried exporting and reimporting, but no change.
What other things could I try? And what data recovery options exist for ZFS? I have seen a few scripts around that dig through fs data structures, but they’re all a bit more unpolished and potentially destructive than I’d like at this point.
zpool status has something to say. Try running
zpool import (without arguments) to list out what pools ZFS found that you can likely re-import.
The file that describes what pools exist on the system should be “/etc/zfs/zpool.cache”. If that’s corrupted, it would explain why the pools aren’t automatically importing.
You can also try the old
zpool import <pool name> and see if that helps you out.
That’s an older release of ZFS on Linux. I’d hate for that to happen to anyone. Did you post to the mailing list?
I would try to understand the nature of the crash, as that’s not normal. What type of hardware (server, controller) did you have, and what was your pool setup?