Re: RDBMS reliability on DEC Alpha OSF/1

cwakins@leia.alloys.rmc.com (Clem Akins) writes: <snip> >more and more of our operation to the Alpha box. If there's a >problem (perhaps associated with a tuning parameter at 4 cpus?) then we >really want to know. >Testify, brother! Well, OK, I'll toss in my 2 bits :) I'm really interested in this thread too, we are hoping to go live in December. (fortunately not a real-time system). We had problems where the indexes are corrupted on largeish (>300 meg or so) tables under 7.10UD1. Setting NUMCPUVPs to 1 was a workaround for this problem, but cost us a bit of performance. We also always had a lot of messages about "out of KAIO resources: errno 35" in the online.log, but couldn't prove any real problems from this. This is on a 4cpu Alpha 8200, OSF/1 3.2B. We *just* got 7.11UC1, and are checking to see if this helps or hurts :) I'd be interested to know if others are having the same problems. We also got it on an HP K400, 4cpu with 7.10UD3. (shows up in oncheck -cI, consistently after creating any index on the big tables). jeffb -- Microsoft Network is prohibited from redistributing this work in any form, in whole or in part. Copyright, Jeff Bartoszewicz, 1995. License to distribute this post is available to Microsoft for $1000. Redistribution without permission constitutes an agreement to these terms. Please send notices of violation to Postmaster@microsoft.com ------------------------------------------------------------------------------- Jeff Bartoszewicz jeffb@netcom.com Give me some slack... JFB Associates, Inc. Canter & Siegal Must Be Annoyed or KILL me. fnord.