Date: Sat, 19 Sep 1998 23:14:55 -0400 From: David Holland <dholland@cs.toronto.edu> To: current@FreeBSD.ORG Subject: Re: softupdates & fsck Message-ID: <98Sep19.231458edt.37911-17305@qew.cs.toronto.edu> In-Reply-To: <199809192317.QAA11186@salsa.gv.tsc.tdk.com> from "Don Lewis" at Sep 19, 98 07:17:37 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> } My suspicion is that the first fsck run is getting the link counts wrong
> } when it repairs the filesystem. I've taked a look at the fsck code, but
> } haven't gotten too far, mostly because the code is so well commented -- NOT!
>
> Yup, fsck is definitely screwing up the filesystem when it tries to repair
> it. The following transcripts were all taken while running single-user
> after a panic.
> :
> :
> ** Phase 4 - Check Reference Counts
> LINK COUNT DIR I=139883 OWNER=root MODE=40755
> SIZE=1024 MTIME=Sep 19 11:47 1998 COUNT 2 SHOULD BE 1
> ADJUST? no
FWIW (probably not much) I was seeing almost the same behavior out of
NetBSD's fsck about a month ago after using some broken fs code. This
had messed up reference counts, so that when you removed a file, it
thought it was still open forever and so it never cleared the inode.
This made fsck unhappy on files, but *very* unhappy if you removed a
directory.
The symptoms were almost identical, down to fsck marking the fs clean
when it actually wasn't.
So I'm going to hypothesize that there's a long-standing bug in fsck
here that something in softupdates is tickling.
--
- David A. Holland | (please continue to send non-list mail to
dholland@cs.utoronto.ca | dholland@hcs.harvard.edu. yes, I moved.)
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?98Sep19.231458edt.37911-17305>