From owner-freebsd-questions Sat Apr 27 15:41:49 2002 Delivered-To: freebsd-questions@freebsd.org Received: from Princeton.EDU (postoffice.Princeton.EDU [128.112.129.120]) by hub.freebsd.org (Postfix) with ESMTP id A271437B41D for ; Sat, 27 Apr 2002 15:39:56 -0700 (PDT) Received: from smtpserver1.Princeton.EDU (smtpserver1.Princeton.EDU [128.112.129.65]) by Princeton.EDU (8.12.3/8.12.3) with ESMTP id g3RMa1d3021467 for ; Sat, 27 Apr 2002 18:36:01 -0400 (EDT) Received: from cs.princeton.edu (aegis.CS.Princeton.EDU [128.112.152.6]) (authenticated bits=0 netid=yruan) by smtpserver1.Princeton.EDU (8.12.2/8.12.2) with ESMTP id g3RMa0g5029316 (version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT) for ; Sat, 27 Apr 2002 18:36:01 -0400 (EDT) Message-ID: <3CCB2792.FA96FDB0@cs.princeton.edu> Date: Sat, 27 Apr 2002 18:34:58 -0400 From: Yaoping Ruan X-Mailer: Mozilla 4.76 [en] (Windows NT 5.0; U) X-Accept-Language: en MIME-Version: 1.0 To: freebsd-questions@FreeBSD.ORG Subject: FreeBSD4.5 kernel hang up & cputime limit exceeded Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Greeting: Recently I did an http server performance test on FreeBSD4.5 and met kernel hang problem. The simultaneous connections I had were not too high, at most 100 connections. But some requests do need disk activity and transfer large files (500 - 900K). It happened especially doing large file requests even when no disk access was involved. Here's more information about hardware: Intel Giganet ethernet network adaptor Intel SBT2 SMP motherboard (but I didn't use FreeBSD for SMP) Pentium III Xeon Processor 1 GHz Promise Ultra DMA 66 disk controller (for IDE disk) 1 GB memory And I configured the /boot/loader.conf with: kern.ipc.maxsockets="131072" kern.ipc.nmbclusters="40960" kern.ipc.nmbufs="81920" net.inet.tcp.tcbhashsize="32768" net.inet.tcp.delayed_ack="0" A few days ago, I changed numbclusters="20480" nmbufs="40960" this time, kernel hang didn't happen but I was told "cputime limit exceeded" after couple of hours' test. While "limit" command shows CPUtime is unlimited. Confused. And advise? Thanks - Randy To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message