Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 13 Apr 2010 13:20:34 -0400 (EDT)
From:      Benjamin Kaduk <kaduk@MIT.EDU>
To:        Jan Henrik Sylvester <me@janh.de>
Cc:        afs-list freebsd <freebsd-afs@freebsd.org>
Subject:   Re: AFS on FreeBSD 8?
Message-ID:  <alpine.GSO.1.10.1004131314070.29136@multics.mit.edu>
In-Reply-To: <4BC4A5F1.6050406@janh.de>
References:  <558186334.2619.1271175586732.JavaMail.root@thunderbeast.private.linuxbox.com> <alpine.GSO.1.10.1004131220540.29136@multics.mit.edu> <4BC4A5F1.6050406@janh.de>

next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 13 Apr 2010, Jan Henrik Sylvester wrote:
>
> With a kernel build, I was able to build OpenAFS. First experience:
>
> afsd was complaining about /usr/local/etc/openafs/cacheinfo missing. After I

Yes, that needs to be specified; I either forgot to or didn't feel like 
packaging a default setup.

> copied that from a Linux OpenAFS install (and modified it to something 
> suitable), I was able to start afsd, but /afs was missing. Creating that and

Arguably I should create /afs in the packaging.

> starting afsd again locked the machine completely (for some reason, fsck on

I have seen a page fault on startup like that (once); I have a coredump 
but haven't had time to dig through it.

> reboot took very long even though I got journaling). Having rebooted, I was 
> able to start afsd and browse to some public afs directory -- 'ls' took quite 
> some time, but eventually, I was able to list that directory.

Is the CellServDB up-to-date for your cell of choice?  Needing to do DNS 
queries could make that take a while.

>
> About the port: pkg-plist differs quite a bit from what actually got 
> installed. I will try to create "my pkg-plist" later.

Hm, I am pretty sure it was correct (or very nearly so) for me.
Please do send me your differences (private mail is fine).

> I will to more testing tomorrow -- especially with Kerberos.

I note that with recent opeanfs snapshots from git, I've been getting heap 
corruption in aklog, so that it dumps core after getting tokens for the 
first cell in the list.  It's annoying, but it has still been getting 
tokens for me.  Whether or not you see that same behaviour would be good 
to know.

-Ben



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?alpine.GSO.1.10.1004131314070.29136>