Date: Tue, 2 Sep 1997 14:42:24 -0300 (EST) From: Joao Carlos Mendes Luis <jonny@mailhost.coppe.ufrj.br> To: questions@freebsd.org Subject: -stable and squid Message-ID: <199709021742.OAA03184@gaia.coppe.ufrj.br>
next in thread | raw e-mail | index | archive | help
Hi, After a make world using 2.2-stable sources from last week, my squid server started to have problems wiht memory, giving these errors: Sep 2 10:59:09 gaia squid[24843]: xmalloc: Unable to allocate 4536 bytes! Sep 2 11:07:41 gaia squid[25430]: xmalloc: Unable to allocate 4096 bytes! Sep 2 11:17:42 gaia squid[25727]: xmalloc: Unable to allocate 8192 bytes! Sep 2 11:36:50 gaia squid[25926]: xmalloc: Unable to allocate 8192 bytes! Sep 2 11:54:48 gaia squid[26496]: xmalloc: Unable to allocate 8192 bytes! Sep 2 12:12:52 gaia squid[27198]: xmalloc: Unable to allocate 4096 bytes! Sep 2 12:36:16 gaia squid[28344]: xmalloc: Unable to allocate 8192 bytes! Sep 2 12:57:45 gaia squid[29784]: xmalloc: Unable to allocate 4536 bytes! Sep 2 13:46:36 gaia squid[1054]: xmalloc: Unable to allocate 4536 bytes! Very soon after this, it dies with SIGIOT: Sep 2 11:01:26 gaia /kernel: pid 24843 (squid), uid 76: exited on signal 6 Sep 2 11:09:53 gaia /kernel: pid 25430 (squid), uid 76: exited on signal 6 Sep 2 11:19:52 gaia /kernel: pid 25727 (squid), uid 76: exited on signal 6 Sep 2 11:38:59 gaia /kernel: pid 25926 (squid), uid 76: exited on signal 6 Sep 2 11:57:05 gaia /kernel: pid 26496 (squid), uid 76: exited on signal 6 Sep 2 12:15:03 gaia /kernel: pid 27198 (squid), uid 76: exited on signal 6 Sep 2 12:38:37 gaia /kernel: pid 28344 (squid), uid 76: exited on signal 6 Sep 2 12:59:57 gaia /kernel: pid 29784 (squid), uid 76: exited on signal 6 Sep 2 13:48:56 gaia /kernel: pid 1054 (squid), uid 76: exited on signal 6 It was a 1.1.12 squid, and upgrading to 1.1.14 did not solve the problem. Since it started with the FreeBSD upgrade, I think it's a FreeBSD problem, not a squid problem. At the squid-users list, they suggested to compile a new kernel with options "MAXDSIZ=(256*1024*1024)" options "DFLDSIZ=(256*1024*1024)" but I've never used these. Instead, I use sysctl and ulimit to get this in the squid startup script. Also, since this seems to be a problem with resource management, maybe this is related: Also after the FreeBSD upgrade, I'm receiving these messages: Sep 2 12:20:00 gaia /kernel: calcru: negative time: -2133 usec Sep 2 12:29:03 gaia /kernel: calcru: negative time: -909 usec Sep 2 12:30:00 gaia /kernel: calcru: negative time: -1207 usec Sep 2 12:40:00 gaia /kernel: calcru: negative time: -2156 usec Sep 2 12:46:20 gaia /kernel: calcru: negative time: -584 usec Sep 2 12:46:25 gaia /kernel: calcru: negative time: -343 usec Sep 2 12:46:26 gaia /kernel: calcru: negative time: -171 usec Sep 2 12:50:26 gaia /kernel: calcru: negative time: -2087 usec Sep 2 13:30:00 gaia /kernel: calcru: negative time: -2096 usec Sep 2 14:01:46 gaia /kernel: calcru: negative time: -353 usec Does someone know what's happening ? Jonny -- Joao Carlos Mendes Luis jonny@gta.ufrj.br +55 21 290-4698 jonny@coppe.ufrj.br Universidade Federal do Rio de Janeiro UFRJ/COPPE/CISI PGP fingerprint: 29 C0 50 B9 B6 3E 58 F2 83 5F E3 26 BF 0F EA 67
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199709021742.OAA03184>