Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 15 Mar 1999 07:31:46 +1030
From:      Greg Lehey <grog@lemis.com>
To:        vallo@matti.ee, freebsd-stable@freebsd.org
Subject:   Re: 3.1-stable crash: is it Vinum or what..?
Message-ID:  <19990315073146.G429@lemis.com>
In-Reply-To: <19990314204444.A67104@matti.ee>; from Vallo Kallaste on Sun, Mar 14, 1999 at 08:44:44PM %2B0200
References:  <19990314204444.A67104@matti.ee>

next in thread | previous in thread | raw e-mail | index | archive | help
On Sunday, 14 March 1999 at 20:44:44 +0200, Vallo Kallaste wrote:
>
> Hello !
>
> We plan to move some machines to 3.1-STABLE soon and use now-in-beta
> status Vinum for striping. The first shot was to install test machine
> and try the Vinum. Well, I can crash the test machine within 15
> minutes doing make -j6 -DCLOBBER -DNOGAMES -DNOAOUT buildworld with
> MAKEOBJDIRPREFIX pointing to striped volume. I hooked up another
> machine for serial console and got traceback from DDB and dump also.
> Here's results using guidelines from man 4 vinum:
>
> vinum: loaded
> vinum: reading configuration from /dev/wd2s1h
> vinum: updating configuration from /dev/wd1s1h
> rm: /dev/vinum/drive/drive1: Read-only file system
> (many more of same)
> Can't create /dev/vinum/control: File exists
> (many more of same)

These messages now no longer appear.  You should get a newer version
of vinum (ftp://ftp.lemis.com/pub/vinum/).

> #9  0xf01efe0a in trap (frame={tf_es = 16, tf_ds = 16, tf_edi = -559038242,
>       tf_esi = -2147483648, tf_ebp = -194888692, tf_isp = -194888728,
>       tf_ebx = -255801832, tf_edx = 0, tf_ecx = 0, tf_eax = -16162,
>       tf_trapno = 12, tf_err = 0, tf_eip = -257838208, tf_cs = 8,
>       tf_eflags = 66182, tf_esp = -257904640, tf_ss = -242003000})
>     at ../../i386/i386/trap.c:437
> #10 0xf0a1b380 in launch_requests (rq=0xf0cdde40, reviveok=0)
>     at /usr/src/sys/modules/vinum/../../dev/vinum/vinumrequest.c:368
> #11 0xf0a1b235 in vinumstart (bp=0xf19353c8, reviveok=0)
>     at /usr/src/sys/modules/vinum/../../dev/vinum/vinumrequest.c:300
> #12 0xf0a1b00a in vinumstrategy (bp=0xf19353c8)
>     at /usr/src/sys/modules/vinum/../../dev/vinum/vinumrequest.c:151
> #13 0xf0163832 in spec_strategy (ap=0xf4623c88)
>     at ../../miscfs/specfs/spec_vnops.c:540

Yup, I've seen this before.  I *hope* I fixed it yesterday (revision
1.14 of vinumstrategy.c).  I'm *very* interested in taking a look at
the dump, though; can you give me access to the machine where it's
located?

> All filesystems except the vinum volume are mounted with
> softupdates.  The filesystem on the volume is async mounted, no
> softupdates. Swap is spread over the three disks, 64MB each. I'm not
> clever enough to analyse this situation, is it caused by Vinum or
> something else? 

It's definitely a Vinum problem.

> In fact the machine doesn't crash with inactive volume. Yet another
> note, after crashing the filesystem which resides on the vinum
> volume never get fixed by fsck. it complains: fsck: cannot find
> inode XXXXXX

I haven't seen that one before, but it's conceivable, depending on
what you were doing at the time of the crash.  That in itself isn't
Vinum-specific.

Greg
--
See complete headers for address, home page and phone numbers
finger grog@lemis.com for PGP public key


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19990315073146.G429>