Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 11 Apr 2000 10:11:05 +0930
From:      Greg Lehey <grog@lemis.com>
To:        "Jim C. Nasby" <jim@nasby.net>
Cc:        tomb <tomb@cgf.net>, freebsd-questions@FreeBSD.ORG
Subject:   Re: vinum causes fatal double fault kernel panic
Message-ID:  <20000411101105.A26373@freebie.lemis.com>
In-Reply-To: <38F1DE8A.502F0672@nasby.net>
References:  <38F0359F.E45937FC@cgf.net> <38F0C358.D120908@cgf.net> <38F0359F.E45937FC@cgf.net> <20000410104141.C17984@freebie.lemis.com> <38F16ACE.F93E576E@cgf.net> <20000410152151.I19079@freebie.lemis.com> <38F17425.985C1DFD@cgf.net> <38F1DE8A.502F0672@nasby.net>

next in thread | previous in thread | raw e-mail | index | archive | help
[Format recovered--see http://www.lemis.com/email/email-format.html]

On Monday, 10 April 2000 at  9:00:42 -0500, Jim C. Nasby wrote:
> X-Mailer: Mozilla 4.7 [en] (WinNT; U)

Please keep your text lines to less than 80 characters; it makes it
very difficult to read otherwise.

> tomb wrote:
> <SNIP>
>> The major crash hasn't happened since I formatted the disks.  The only other
>> change was my refference to the disks.  The first time around (double fault) I
>> refferenced  them as.
>>
>> drive a device /dev/da0s1e
>> drive b device /dev/da1s1e
>>
>> Which judging by the man page on disk labeling is totaly wrong, hence I guess
>> this is the source of the kernel panics.
>
> Is this true Greg?

Guessing is guesswork.  I need to look at the problem first, and so
far we don't have much evidence for anything.  On older versions of
Vinum I didn't support this kind of naming, but it's been there since
3.3.  There were some problems in 3.3-RELEASE, but they've been fixed,
and are documented in http://www.lemis.com/vinum/known-bugs.html.

> I was having frequent crashes and reboots on my vinum raid5 system,
> which was had it's drives configured as
>
> drive vinum0 device /dev/da0s1h
> drive vinum1 device /dev/da1s1h
> drive vinum2 device /dev/da2s1h
> drive vinum3 device /dev/da3s1h
>
> The system lunched the drives recently, so I can't show what
> disklabel used to read, but in each case, the h partition was a
> vinum partition

Tht's the important part.

> and each drive had at least one other partition for a minimum FBSD
> install,

That doesn't matter.

> which is how I bootstrapped FBSD onto vinum (the box now has a 5th
> HD for the OS and I don't currently have vinum running).

If the disk label is wrong, you won't get the system to work at all.
It should just refuse, not panic, but it won't work.

> Is da#s#h a valid device for vinum,

Yes.  This is described in the man page.

> or do I have to use the entire drive? Or do I have to use an entire
> slice (ie: da0s1)?

No, that's wrong.

> Also, don't know if it makes a difference or not, but I also had
> swap on a raid5 plex..

I've seen somebody do that on NetBSD (with RAIDframe) recently.  I
don't understand why you'd want to do that.  RAID-5 is very slow on
writes, and swap is usually small enough that the additional hardware
cost of RAID-1 would be negligible.

> in fact, everything but root (and the old partitions which generally
> weren't used, but were mounted rw) was on a raid5 plex. I can send
> you my old config file if you're interested. Sorry, but I don't have
> any debug files.

If you have any more problems, take a look at
http://www.lemis.com/vinum/how-to-debug.html and give me the
information I ask for there.

Greg
--
When replying to this message, please copy the original recipients.
For more information, see http://www.lemis.com/questions.html
Finger grog@lemis.com for PGP public key
See complete headers for address and phone numbers


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




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000411101105.A26373>