freebsd 5.3 have any problem with vinum ?

[Matthias Schuendehuette]

Hello,

Am Mittwoch, 10. November 2004 22:31 schrieben Sie:
> ok, your instructions worked like a charm.  So i'm running my nice 4
> member SCSI gvinum raid5 array (with softupdates turned on), and it's
> zipping along.

Fine! :-)

> Now I need to test just how robust this is.

Ouhh... ;-)

> camcontrol is too nice.  I want to test a more real world failure. 
> I'm running dbench and just pull one of  the drives.  My expectation
> is that  I should see a minor pause, and then the array continue in
> some slower, degraded mode.

That was mine too...

> What I get is a kernel trap 12 (boom!).  
> I reboot, and it will not mount the degraded set till I replace the
> drive.
>
> I turned off softupdates, and had the same thing happen.  Is this a
> bogus test?  Is it reasonable to expect that a scsi drive failure
> should of been tolerated w/o crashing?

No, of course not. I have more or less the same problems here. Once I 
came so far as to delete the crashed subdisk but when I tried to delete 
the (not existing anymore) vinumdrive, my kernel also crashed...

Well, to be honest, I once tried to pull the plug on one of my disks 
with 'classic' vinum and I got a kernel panic as well. OK, this 
happened a few years ago and I never tried that again...

I'm not sure if this is a problem of (g)vinum or if FreeBSD has other 
problems in this area.

And we all have to consider that gvinum is in a relatively early 
development phase (IMHO) - it is basically working, that is, it's 
possible to continue an existing 'classic' vinum installation with 
gvinum but it's still not fully functional in all depth. But I think, 
there's all the potential to get there. I have the impression that 
Lukas is *very* interested in his baby and I have a good overall 
feeling so far. But he's the only one developing gvinum AFAIK...

And - my primary interest is the LVM functionality of (g)vinum. IMHO if 
you *really* have valuable data to protect, you can afford a hardware 
RAID-controller (*and* a tape drive :-). Anything else is wrong 
economy.

But the current disk layout possibilities with up to four slices and at 
max 28 BSD-partitions are far to inflexible for todays large disks.

So from this point of view I'm already fairly happy with gvinum as it is 
today. Which doesn't mean that I'm not trying to help to get gvinum to 
the place and state where it deserves to be... :-)
-- 
Ciao/BSD - Matthias

Matthias Schuendehuette <msch [at] snafu.de>, Berlin (Germany)
PGP-Key at <pgp.mit.edu> and <wwwkeys.de.pgp.net> ID: 0xDDFB0A5F