Bug 241117 - zfs_unlinked_drain takes "forever" for over-quota ZFS filesystems
Summary: zfs_unlinked_drain takes "forever" for over-quota ZFS filesystems
Status: New
Alias: None
Product: Base System
Classification: Unclassified
Component: kern (show other bugs)
Version: 11.2-RELEASE
Hardware: Any Any
: --- Affects Some People
Assignee: freebsd-fs (Nobody)
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2019-10-07 14:36 UTC by Peter Eriksson
Modified: 2019-10-08 06:22 UTC (History)
0 users

See Also:

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Peter Eriksson 2019-10-07 14:36:43 UTC 
When rebooting a server "zfs mount -a" can take "forever" (or atleast a silly amount of time) if there is a ZFS filesystems that are at or over 'refquota' when the kernel tries to execute "zfs_unlinked_drain" (and that filesystem has unlinked stuff to drain).

This effective allows a user to cause a denial-of-service if the server happens to reboot due to a bug...

Workaround: 

Make sure "zfs mount" is done in the background in the /etc/rc.d/zfs startup script - so that you can get a login prompt, and then give the filesystem more quota so progress can continue.