From owner-freebsd-current Tue Jan 29 23:30:15 2002 Delivered-To: freebsd-current@freebsd.org Received: from smtpzilla3.xs4all.nl (smtpzilla3.xs4all.nl [194.109.127.139]) by hub.freebsd.org (Postfix) with ESMTP id CC61D37B404; Tue, 29 Jan 2002 23:30:01 -0800 (PST) Received: from router.hackerheaven.org ([80.126.0.97]) by smtpzilla3.xs4all.nl (8.12.0/8.12.0) with ESMTP id g0U6nUbS070453; Wed, 30 Jan 2002 07:49:31 +0100 (CET) Received: from loki.ipv6.hackerheaven.org (loki.ipv6.hackerheaven.org [10.0.0.12]) by router.hackerheaven.org (Postfix) with ESMTP id B01401C1F; Wed, 30 Jan 2002 07:48:50 +0100 (CET) Date: Wed, 30 Jan 2002 07:49:13 +0100 Subject: Re: many copies of make running while building a port... Content-Type: text/plain; charset=US-ASCII; format=flowed Mime-Version: 1.0 (Apple Message framework v480) Cc: current@freebsd.org, ports@freebsd.org To: Emiel Kollof From: Emiel Kollof In-Reply-To: <86263CF3-1547-11D6-A3B8-000A277C7FDE@hackerheaven.org> Message-Id: <78AB616E-154D-11D6-A3B8-000A277C7FDE@hackerheaven.org> Content-Transfer-Encoding: 7bit X-Mailer: Apple Mail (2.480) Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Emiel Kollof heeft op woensdag 30 januari 2002 om 07:06 het volgende geschreven: > Number of procs in top(1) shoot up into the ranges of 600+ procs (and > usually defunct). Load average sometimes jumps up into the 100+ (I kid > you not). I doubt that that is desired operation. I only started seeing > this behavior when I switched over to CURRENT, but this mail is cc'd to > ports@ too, just in case. It makes using portupgrade(1) as tense as a > good thriller flick (will my system survive? The suspense!) I hate following up on myself, but to show you the scope of it all, here's a snippet from top(1)'s output: last pid: 23113; load averages: 1.88, 1.13, 1.01 up 0+01:02:22 07:41:57 440 processes: 3 running, 414 sleeping, 23 zombie CPU states: 9.3% user, 0.0% nice, 89.8% system, 0.8% interrupt, 0.0% idle Mem: 13M Active, 7656K Inact, 25M Wired, 3220K Cache, 14M Buf, 9468K Free Swap: 396M Total, 92M Used, 304M Free, 23% Inuse, 8K In PID USERNAME PRI NICE SIZE RES STATE TIME WCPU CPU COMMAND 14608 root 98 0 2348K 892K RUN 0:13 5.79% 5.76% top 394 root 99 0 3868K 1472K RUN 2:27 3.47% 3.47% icecast 864 root 96 0 2376K 796K select 0:20 0.29% 0.29% top 22728 root 8 0 644K 524K wait 0:00 0.58% 0.24% make 22690 root 8 0 644K 524K wait 0:00 0.43% 0.20% make 22709 root 8 0 624K 500K wait 0:00 0.46% 0.20% make [snip lots more occurrences of make(1)] And another snippet from the building of the port: ---> Building '/usr/ports/audio/libvorbis' Cannot fork: Resource temporarily unavailable Cannot fork: Resource temporarily unavailable Cannot fork: Resource temporarily unavailable Cannot fork: Resource temporarily unavailable Cannot fork: Resource temporarily unavailable Cannot fork: Resource temporarily unavailable Cannot fork: Resource temporarily unavailable Cannot fork: Resource temporarily unavailable ===> Cleaning for libogg-1.0.r3_1,2 This happens with _all_ ports. Somehow I don't tjink this is right... Any leads? Just built world with fresh sources as of 05:00 CET yesterday. I've been seeing this behavior since monday. Cheers, Emiel To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message