Date: Tue, 11 Jan 2000 03:26:00 -0800 From: "Fabian Thylmann" <fthylmann@stats.net> To: "Scott Hess" <scott@avantgo.com>, <questions@freebsd.org> Subject: Re: Problem with not disappearing sockets. Message-ID: <002a01bf5c26$a4227ee0$0593e289@oih.rwthaachen.de> References: <004301bf58c6$9f33bc40$0593e289@oph.rwthaachen.de> <0ad901bf587b$6fde13f0$1e80000a@avantgo.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi Scott, sorry for taking so long to reply, I normally reply faster, but I had too much stuff to do.. Your sysctl thing helped wonders :) Thanks a LOT for this! The way we handle the high socket numbers is just a very high maxusers setting in the kernel and on top of that also directly setting the max nmbclusters to a high number (even higher than the calculated one that results from the maxusers) ..... This helped keeping the box from kernel panicing because of mbuf cluster overruns, which I guess your box probably crashes on. Check the netstat -m numbers, that will give you info on max and current mbuf usage. If thats very close, its probably mbuf clusters running out and the kernel panicing on that. The easiest way to get around this is to just increase the maxusers number in the kernel. Fabian Thylmann STATSnet sprl ----- Original Message ----- From: Scott Hess <scott@avantgo.com> To: Fabian Thylmann <fthylmann@stats.net>; <questions@freebsd.org> Sent: Thursday, January 06, 2000 11:22 AM Subject: Re: Problem with not disappearing sockets. > [...] > > [BTW, could you please make suggestions as to how you managed tens of > thousands of sockets in this state? Our machines always rebooted at around > 1500.] > > [...] To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?002a01bf5c26$a4227ee0$0593e289>