Re: kernel panic when mounting root vinum volume

On Saturday,  6 March 2004 at 23:00:55 -0500, George D. Gal wrote:
> Hello Greg,
>
> Well, unfortunately I wasn't able to find the dot files you mention in the
> documentation under /usr/src/sys/modules/vinum/.gdb*

They're there in 4.9.  For 5.2-CURRENT, read gdb(4).

> I've tried this particular configuration under FreeBSD 5.2.1-REL and
> 5.2-CURRENT and have the same problems under both. I'm using this
> exact config on FreeBSD 4.9 without problems. I have not made
> changes to the system source.
>
> My vinum list output is as follows:
> 2 drives:
> D d1                    State: up       /dev/ad1s1d     A: 258/37122 MB (0%)
> D d0                    State: up       /dev/ad0s1d     A: 258/37122 MB (0%)
>
> 4 volumes:
> V rootvol               State: up       Plexes:       2 Size:       2048 MB
> V usrvol                State: up       Plexes:       2 Size:         12 GB
> V varvol                State: up       Plexes:       2 Size:         10 GB
> V datavol               State: up       Plexes:       2 Size:         12 GB
>
> 8 plexes:
> P rootvol.p1          C State: up      Subdisks:     1 Size:       2048 MB
> P usrvol.p1           C State: up      Subdisks:     1 Size:         12 GB
> P varvol.p1           C State: up      Subdisks:     1 Size:         10 GB
> P datavol.p1          C State: up      Subdisks:     1 Size:         12 GB
> P rootvol.p0          C State: up       Subdisks:     1 Size:       2048 MB
> P usrvol.p0           C State: up       Subdisks:     1 Size:         12 GB
> P varvol.p0           C State: up       Subdisks:     1 Size:         10 GB
> P datavol.p0          C State: up       Subdisks:     1 Size:         12 GB
>
> I've tried to also detach individual plexes to see if it would
> eliminate the problem, without luck.

What was the problem?  Yes, you've said it, but you shouldn't expect
people to go back and piece together the evidence.  There's obviously
a panic, but what else?

> Unfortunately my vinum history and /var/log/messages has no error
> messages or anything out of the ordinary for vinum or any other
> errors for that matter.

Let me decide whether things are out of the ordinary.

> Since I have not yet setup the remote debug the details of the crash
> are as follows.

Where's the dump?

> Right before the crash I see a message stating, configuration already read
> from d1.
>
> running where in the debugger gives me the last few func calls:
>
> devsw
> initdrive
> read_drive_label
> check_drive
> vinum_scandisk
> vinum_super_ioctl
> vinum_ioctl

What happened to the details?

It's not easy to debug these problems.  The least I could expect is
your cooperation.

Greg
--
Note: I discard all HTML mail unseen.
Finger grog_at_FreeBSD.org for PGP public key.
See complete headers for address and phone numbers.