From owner-freebsd-stable Mon Jul 15 00:06:25 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id AAA08065 for stable-outgoing; Mon, 15 Jul 1996 00:06:25 -0700 (PDT) Received: from acc0.elvisti.kiev.ua (acc0.elvisti.kiev.ua [193.125.28.132]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id AAA07874 for ; Mon, 15 Jul 1996 00:04:57 -0700 (PDT) Received: from office.elvisti.kiev.ua (office.elvisti.kiev.ua [193.125.28.129]) by acc0.elvisti.kiev.ua (8.7.5/8.7.3) with ESMTP id KAA26260; Mon, 15 Jul 1996 10:06:17 +0300 (EET DST) Received: (from stesin@localhost) by office.elvisti.kiev.ua (8.6.12/8.ElVisti) id KAA00739; Mon, 15 Jul 1996 10:06:17 +0300 From: "Andrew V. Stesin" Message-Id: <199607150706.KAA00739@office.elvisti.kiev.ua> Subject: Re: Memory leak in 1.0.1 / FreeBSD 2.1R & BSDI To: stefan@netscafe.rotterdam.luna.net (Stefan Arentz) Date: Mon, 15 Jul 1996 10:06:16 +0300 (EET DST) Cc: squid-users@nlanr.net, stable@freebsd.org In-Reply-To: <199607142159.XAA01121@netscafe.rotterdam.luna.net> from "Stefan Arentz" at Jul 14, 96 11:59:53 pm X-Mailer: ELM [version 2.4 PL24alpha5] Content-Type: text Sender: owner-stable@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi people, that's me who did an "official" FreeBSD port, and we don't see any such problems here. BUT: we are running Squid on a FreeBSD-stable machine with malloc() taken from FreeBSD-current (so called phkmalloc). I'll be grateful to see the results of the testing. # Eric, # # Danny ter Haar send me the following private message: # # >From: Danny ter Haar # >Subject: Re: Memory leak in 1.0.1 / FreeBSD 2.1R # > # >An other provider running BSD and squid had the same problem: # >After compiling with the linux malloc everything worked like a charm # > # >Danny # # So i installed libdlmalloc-2.5 from the FreeBSD packages collection. I'm now # running a squid binary linked with that library. I'll post the results on # tuesday I think. # # Regards, # - Stefan # # Your message dated: Sun, 14 Jul 1996 23:24:41 +0200 # >At 22:05 14/07/1996 +0200, Stefan Arentz wrote: # > # >>> top -b | grep squid # >>> 23351 root 2 0 13M 6365K sleep 9:40 0.00% 0.00% squid # >>> # >>> and Total Accounted was 10092KB. # >>> # >>> This is on a squid-1.0.1 system running on FreeBSD-2.1-stable. # >>> # >>> Personally I wouldn't be worried about the discrepencies in your # >>> figures until they affect your system in some manor. # >> # >>If it's broken then it should be fixed imho. I'm running squid on a web # >>server and cant afford to let squid crash this machine... Four days ago # >>squid was using 42MB of memory while there was just 20MB cached on disk.. # >> # >>I'll track this down. It looks like a memory leak somewhere... # >> # > # >Since the Jul 8th, 1996: # >I run squid 1.0.1 on a 128 MB central server running BSDI. # >Process squid grows steadily in memory (we have over 17.000 users) until it # >reaches Datasize memory limit then "crashes" (xxalloc can't alloc 4096 bytes # >of memory). # > # >I run squid 1.0.1 on a neighbor server handling requests, from the above # >server, that are related to US sites: So far, so good. # > # >Since the "squid" source has been passed through "purify", I would think the # >memory leak is real but doesn't show up until some error condition occurs. # > # >Most errors reported in "cache.log" are related to the following source # >lines: # > # >* comm.c : 372 # >* comm.c : 1194 # > # >ftpget.c: 652 # > # >ipc.c : 1158 # >ipc.c : 1375 # >* ipc.c : 1887 # >ipc.c : 1862 # > # >* outlines the most common errors. # > # >Hope it will help. I'll try to spend some time around those lines. # >Best regards, # >Eric Salome # > # > # -- With best regards -- Andrew Stesin. Phones/fax: +380 (44) { 244-0122, 276-0188, 271-3457, 271-3560 } "You may delegate authority, but not responsibility." Frank's Management Rule #1.