From owner-freebsd-scsi Fri Aug 11 17:53: 2 2000 Delivered-To: freebsd-scsi@freebsd.org Received: from northlink.com (northlink.com [216.19.0.216]) by hub.freebsd.org (Postfix) with ESMTP id 21CEB37BC4A for ; Fri, 11 Aug 2000 17:52:58 -0700 (PDT) (envelope-from jmodjeski@ms1.northlink.com) Received: from ms1.northlink.com (ms1.northlink.com [216.19.4.28]) by northlink.com (8.9.1/8.9.1) with ESMTP id RAA21482; Fri, 11 Aug 2000 17:52:48 -0700 (MST) X-Real-From: jmodjeski@ms1.northlink.com X-Real-To: freebsd-scsi@freebsd.org Received: by MS1 with Internet Mail Service (5.5.2650.21) id ; Fri, 11 Aug 2000 17:40:26 -0700 Message-ID: <00101B7A7FDDD311A89500A0CC56C79048B8@MS1> From: Joe Modjeski To: "'Greg Lehey'" Cc: "'freebsd-scsi@freebsd.org'" Subject: RE: to Vinum or not to Vinum Date: Fri, 11 Aug 2000 17:40:26 -0700 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2650.21) Content-Type: text/plain; charset="iso-8859-1" Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org FreeBSD 4.0-RELEASE #0 The default vinum version that came with 4.0. no source changes. I use a minimal install and then map the /usr /var /tmp directories to the Vinum drive once setup. Then extract the kernel source and install ports to the Raid5 volume vinum list: 5 drives: D Raid51 State: up Device /dev/da1s1e Avail: 0/8675 MB (0%) D Raid52 State: up Device /dev/da2s1e Avail: 0/8675 MB (0%) D Raid53 State: up Device /dev/da3s1e Avail: 0/8675 MB (0%) D Raid54 State: up Device /dev/da4s1e Avail: 0/8675 MB (0%) D Raid55 State: up Device /dev/da5s1e Avail: 0/8675 MB (0%) 1 volumes: V Raid5 State: up Plexes: 1 Size: 33 GB 1 plexes: P Raid5.p0 R5 State: up Subdisks: 5 Size: 33 GB 5 subdisks: S Raid5.p0.s0 State: up PO: 0 B Size: 8675 MB S Raid5.p0.s1 State: up PO: 512 kB Size: 8675 MB S Raid5.p0.s2 State: up PO: 1024 kB Size: 8675 MB S Raid5.p0.s3 State: up PO: 1536 kB Size: 8675 MB S Raid5.p0.s4 State: up PO: 2048 kB Size: 8675 MB I am working on getting the dumps but it is not as easy as it sounds in the handbook or the link. I would like to try the version of vinum that was stated to be patched on the vinum page: 28 September 1999: We have seen hangs when perform heavy I/O to RAID-5 plexes. The symptoms are that processes hang waiting on vrlock and flswai. Use ps lax to display this information. Technical explanation: A deadlock arose between code locking stripes on a RAID-5 plex (vrlock) and code waiting for buffers to be freed (flswai). Status: Fix committed to FreeBSD-CURRENT on 5 January 2000. We're waiting to be reasonably sure that the problem doesn't reoccur, then it will be committed to -STABLE. Actually after looking at the bug report again. I am sure find locks in a state (vrlock) - using ps axl - noticed this while trying to create the crash dump today. I will have to pick this up on Monday. Have a good weekend all. Joe jmodjeski@northlink.com -----Original Message----- From: Greg Lehey [mailto:grog@lemis.com] Sent: Thursday, August 10, 2000 7:36 PM To: Joe Modjeski Cc: freebsd-scsi@FreeBSD.ORG Subject: Re: to Vinum or not to Vinum On Thursday, 10 August 2000 at 12:09:47 -0700, Joe Modjeski wrote: > Currently we have 3 Compaq Proliant 1600R servers with 6 9.1 Ultra3 drives > in each. We are attempting (very unsuccessfully) to do Raid5 with vinum. > We get fatal trap 12 errors very regularly and after a few reboots the vinum > volume is so chewed up that we end up having to rebuild the system. I > tracked down the majority of the problems to the /etc/security script. I > believe it is about the 6th or 7th line down where it starts the find run. > The box starts off fine but after about 1 minute it starts to hit all the > drives at once then BLAM!! It gives me the error. It would be nice to know some details. A dump, for example, or what version of Vinum you're running. You might like to take a look at http://www.lemis.com/vinum/how-to-debug.html. Greg -- 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-scsi" in the body of the message