Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 23 Aug 2024 18:00:01 +0300
From:      Anthony Pankov <anthony.pankov@yahoo.com>
To:        freebsd-fs@FreeBSD.org
Subject:   scheduled run of background fsck
Message-ID:  <967684385.20240823180001@yahoo.com>
References:  <967684385.20240823180001.ref@yahoo.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Hello,

I use UFS with regular (not journaled) softupdates (SU).

Recently I've been told that it is a good idea to schedule a background fsck at least weekly. (I don't now much about UFS,  the reason was something about "reclaiming blocks". Is it a right idea or am I missing something?)

So I plan to add "fsck -B /data/meshare" to crontab. I have two issue with this:

1.
When I manually run "fsck -B" nothing happen:

# fsck -B /data/meshare
# echo $?
0
and no fsck process in a background.

Is it OK?

2. 
Some years ago I've used classic UFS+SU with background_fsck="YES" in rc.conf. But ended with production server rebooted after every 5-10 minutes. The problem appeared after rebooting after some software crash. The reason I've found in logs. Message was something like:

FSCK: CAN'T WRITE
PANIC

As I understand background fsck was doing it job but when it reach part of filesystem  with inconsistency it discover itself in read-only mode and panic to not leave filesystem with inconsistency. It was very frustrating.

May I be sure for FreeBSD 12/STABLE+ that this will never happen again? Can I rely that 'fsck -B' will never cause system panic but gracefully exit and leave a message in log in case of a problem?



-- 
Best regards,
 Anthony  Pankov                        mailto:anthony.pankov@yahoo.com




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?967684385.20240823180001>