From owner-freebsd-stable Sat Jan 26 20:51:27 2002 Delivered-To: freebsd-stable@freebsd.org Received: from resnet.uoregon.edu (resnet.uoregon.edu [128.223.122.47]) by hub.freebsd.org (Postfix) with ESMTP id E693837B400 for ; Sat, 26 Jan 2002 20:51:23 -0800 (PST) Received: from localhost (dwhite@localhost) by resnet.uoregon.edu (8.11.3/8.10.1) with ESMTP id g0R4pkd18540; Sat, 26 Jan 2002 20:51:47 -0800 (PST) Date: Sat, 26 Jan 2002 20:51:46 -0800 (PST) From: Doug White To: Sergey Gershtein Cc: freebsd-stable@FreeBSD.ORG Subject: Re[4]: Strange lock-ups during backup over nfs after adding 1024M RAM In-Reply-To: <0863311696.20020125105552@ur.ru> Message-ID: <20020126204941.H17540-100000@resnet.uoregon.edu> X-All-Your-Base: are belong to us MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG On Fri, 25 Jan 2002, Sergey Gershtein wrote: > On Friday, January 25, 2002 you wrote: > DW> On Thu, 24 Jan 2002, Sergey Gershtein wrote: > > >> It's a rather heavily loaded web server (average of 25 requests/sec, > >> 100kb/sec), so I guess it's a lot of network. Could you please point > >> me to the right direction where to read about mbuf monitoring? I > >> found some info in the Handbook (6.10 Tuning Kernel Limits), but there > >> are not much unfortunately. > > DW> 'netstat -m', watch these lines: > DW> 131/864/10240 mbufs in use (current/peak/max): > DW> 128/172/2560 mbuf clusters in use (current/peak/max) > > DW> You want to keep the second number about 80% of the third number. If it > DW> gets too close bump the kernel tunables kern.mbufs and kern.nmbclusters > DW> (or set options MBUFS and NMBCLUSTERS in the kernel config). > > Today the lock-up happened again during backup over nfs with no sign > in any log files or on console. It was running 4.4-STABLE kernel > cvsuped yesterday with MAXUSERS 128 and NMBCLUSTERS=8192... Any ideas > on where to look? Get a serial console. Something may be logged but since it doesn't sync to disk you can't see it afterwards. > The strange thins is that we have another server with exactly the same > hardware and amout of RAM which works fine. The only difference is > that its kernel was compiled with MAXUSERS 1024 for some reason. Do I > really need to bump MAXUSERS so high to handle more than 1Gb of RAM? So *high*? You need to *reduce* it! 128 should work; if that is blowing up on wierd VM failures, start up a crontask that runs 'sysctl vm.zone' and 'netstat -m' every so often and logs the output. That will say what is gobbling up all the space. NFS is also suspicious .. this is after you updated to today's -STABLE? Doug White | FreeBSD: The Power to Serve dwhite@resnet.uoregon.edu | www.FreeBSD.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message