Date: Tue, 3 Dec 1996 16:53:38 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: imp@village.org (Warner Losh) Cc: terry@lambert.org, hackers@FreeBSD.org Subject: Re: But it didn't work. Message-ID: <199612032353.QAA14902@phaeton.artisoft.com> In-Reply-To: <199612032211.PAA00422@rover.village.org> from "Warner Losh" at Dec 3, 96 03:11:10 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> : You are running UFS on the drive, right?
>
> Yes. Why wouldn't lsof show anything, yet umount think the drive is
> still busy? I don't know the answer to that, but it smells to me like
> some kind of refcount bug somewhere. I don't know the kernel FS code
> well enough to know where or why or how at this point. Nor do I know
> how to recreate it :-(.
>
> umount -f did work, however.
Probably the cache is not clean... ie: one of the vnodes has been
clodes, but is still in the name cache, and so still has buffers
on it (lsof wouldn't show cache contents).
If you wait for a long time, or force the cache to recycle ("find"
on the FS's not on the JAZ drive to force recycle of unused but
still-cache vnodes), and it fixes it, then it's the cache. Otherwise,
it's something else.
Terry Lambert
terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199612032353.QAA14902>