From owner-freebsd-stable@FreeBSD.ORG Mon Jun 9 05:04:42 2003 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 95A0337B401; Mon, 9 Jun 2003 05:04:42 -0700 (PDT) Received: from onion.ish.org (onion.ish.org [219.118.161.2]) by mx1.FreeBSD.org (Postfix) with ESMTP id BC80843F3F; Mon, 9 Jun 2003 05:04:41 -0700 (PDT) (envelope-from ishizuka@ish.org) Received: from localhost (ishizuka@localhost [IPv6:::1]) h59C4eaO049669; Mon, 9 Jun 2003 21:04:40 +0900 (JST) (envelope-from ishizuka@ish.org) Date: Mon, 09 Jun 2003 21:04:40 +0900 (JST) Message-Id: <20030609.210440.112307207.ishizuka@ish.org> To: das@freebsd.org From: Masachika ISHIZUKA In-Reply-To: <20030609085454.GA5633@HAL9000.homeunix.com> References: <20030609041942.GA4029@HAL9000.homeunix.com> <20030609.163252.71109503.ishizuka@ish.org> <20030609085454.GA5633@HAL9000.homeunix.com> X-PGP-Fingerprint20: 276D 697A C2CB 1580 C683 8F18 DA98 1A4A 50D2 C4CB X-PGP-Fingerprint16: C6 DE 46 24 D7 9F 22 EB 79 E2 90 AB 1B 9A 35 2E X-PGP-Public-Key: http://www.ish.org/pgp-public-key.txt X-URL: http://www.ish.org/ X-Mailer: Mew version 3.2 on Emacs 21.2 / Mule 5.0 (SAKAKI) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit cc: stable@freebsd.org Subject: Re: system slowdown - vnode related X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 09 Jun 2003 12:04:42 -0000 >>>> I have still vnodes problem in 4.8-stable with /sys/kern/vfs_subr.c >>>> 1.249.2.30. >>>> >>>> % sysctl kern.maxvnodes >>>> kern.maxvnodes: 17979 >>>> % sysctl vm.zone | grep VNODE >>>> VNODE: 192, 0, 18004, 122, 18004 >>> >>> This looks pretty normal to me for a quiescent system. >> >> I think the used(18004) exceeds maxvnodes(17979), isn't it ? > > Only by a little bit. maxvnodes isn't a hard limit, since making > it a hard limit would lead to deadlocks. Instead, the system > garbage collects vnodes to keep the number roughly in line with > maxvnodes. Judging by the numbers above, it's doing a pretty good > job, but that's probably because, from the looks of it, you > just booted the system. Hi, David-san. Thank you for mail. I understood. > The reason it might make sense to increase maxvnodes is that > having vnlru work overtime to keep your vnode count low may > result in vnodes being freed that are still needed, e.g. by the > buffer cache. This would cause the slowdown you were mentioning. I will try to increase kern.maxvnodes when the machine is slowdown. But I can not reproduce slowdown in experimental environment, yet. >>> Did you get a backtrace from the panics? >> >> It's too hard for me. Is there any way to do it ? > > The panics might be unrelated to the number of vnodes, so it's > important that we have additional information. See: > > http://www.freebsd.org/doc/en/books/developers-handbook/kerneldebug.html I'll try. Thank you very much. -- ishizuka@ish.org