Date: Tue, 4 Feb 2003 00:16:23 -0500 From: "aSe" <aSe@SysFail.com> To: "FreeBSD-Questions" <freebsd-questions@freebsd.org> Cc: "Ruben de Groot" <fbsd-q@bzerk.org> Subject: RE: Too many files open / file: table is full Message-ID: <KCECKLBMJCEIIOLIPBPECEEHCGAA.aSe@SysFail.com> In-Reply-To: <20030203135208.GA78364@ei.bzerk.org>
next in thread | previous in thread | raw e-mail | index | archive | help
>This is not a matter of diskspace. The kernel holds a fixed length = table >in memory with all open files. If this table gets full it usually means = >one of two things: > >1) You have a runaway application, opening way too many files. Identify = >the application and fix or disable it. > >2) You're running a kernel with a too low value for maxusers (which,=20 >among other things, determines the maximum amount of open files). The=20 >default in 4.7-RELEASE is 0, which means: optimize according to amount=20 >of memory installed. The default is usually O.K. If not, one option is >to simply install more memory. The machine itself runs several logging applications and things of that nature. I didn't think It was an issue with HD. Nor do I believe its = ram, It has 512mb installed, and 256mb of swap. As it stands right now it has 270mb free and hasn't touched the swap. Right now maxusers is set to 6,=20 I didn't realize it would play a role in this instance. Jack Stone suggested looking up the number of max open files by doing=20 "sysctl kern.maxfiles" It returns only "232" which to me seems like a=20 very small number. He also suggested to change it using=20 "sysctl -w kern.maxfiles=3D4160." My question to you is, does maxusers play more of a role then just the max number of open files. In the long run would it be better to just set maxusers to 0 or just change the kern.maxfiles? Thank you! Gordon Keesler [aSe@SysFail.com] 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?KCECKLBMJCEIIOLIPBPECEEHCGAA.aSe>