From owner-freebsd-stable Sat Jan 1 11:58:30 2000 Delivered-To: freebsd-stable@freebsd.org Received: from mail2.uniserve.com (mail2.uniserve.com [204.244.156.7]) by hub.freebsd.org (Postfix) with ESMTP id 3198815007; Sat, 1 Jan 2000 11:58:26 -0800 (PST) (envelope-from tom@uniserve.com) Received: from shell.uniserve.ca ([204.244.186.218]) by mail2.uniserve.com with smtp (Exim 3.03 #4) id 124UfH-00084E-00; Sat, 01 Jan 2000 11:58:15 -0800 Date: Sat, 1 Jan 2000 11:58:12 -0800 (PST) From: Tom X-Sender: tom@shell.uniserve.ca To: Matthew Dillon Cc: Peter Wemm , freebsd-stable@FreeBSD.ORG, freebsd-hackers@FreeBSD.ORG Subject: Re: softupdates and debug.max_softdeps In-Reply-To: <199912311950.LAA85337@apollo.backplane.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Fri, 31 Dec 1999, Matthew Dillon wrote: > : How many instances of postmark are you running? I used 4 separate > :instances (you must run them in separate directories). > > Well, you didn't say that! :-) I'm running one. I'll start up > another couple to match your test. Actually, I did. Postmark as a single process won't really stress your system much. Multiple instances should be similar to the load that postfix can put on a system. > : How fast are your disks? I used an external RAID-5 array that appears > :as single disk to FreeBSD. It is RAID-5, so the performance isn't that > :great for writing. I think my virtual disk is quite slow. If your system > :has fast disks, run more instances of postmark. > > Fast. Two striped 18G seacrates on each system 40 MB/sec SCSI bus on > one system, 80 MB/sec SCSI bus on the other. > > : > : Well, I lost last nights panic because of fsck complaining about several > :million unreferenced files. I'm going to log to disk next time. > : > :Tom > :Uniserve > > Ach. I think fsck verbosity is a problem. My capture file was 24 MB and growing. Almost all of it was fsck UNREF errors. Anyhow, the following is the panic. It isn't hard to see where it is going wrong. "inodedep" increases steadily. panic: kmem_malloc(4096): kmem_map too small: 213647360 total allocated mp_lock = 01000001; cpuid = 1; lapic.id = 00000000 boot() called on cpu#1 > -Matt > Matthew Dillon > > > Tom Uniserve To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message