From nobody@hyperreal.com Fri Jun 20 02:43:50 1997 Received: (from nobody@localhost) by hyperreal.com (8.8.5/8.8.5) id CAA18963; Fri, 20 Jun 1997 02:43:50 -0700 (PDT) Message-Id: <199706200943.CAA18963@hyperreal.com> Date: Fri, 20 Jun 1997 02:43:50 -0700 (PDT) From: Josef Karthauser Reply-To: joe@pavilion.net To: apbugs@hyperreal.com Subject: Apache1.2.0 wont start with more than 124 virtual servers on my machine X-Send-Pr-Version: 3.2 >Number: 755 >Category: os-freebsd >Synopsis: Apache1.2.0 wont start with more than 124 virtual servers on my machine >Confidential: no >Severity: critical >Priority: medium >Responsible: apache >State: closed >Class: sw-bug >Submitter-Id: apache >Arrival-Date: Fri Jun 20 02:50:00 1997 >Last-Modified: Sat Jun 28 15:43:08 PDT 1997 >Originator: joe@pavilion.net >Organization: >Release: 1.2.0 >Environment: FreeBSD dougal.pavilion.net 2.2-STABLE FreeBSD 2.2-STABLE #0: Thu Jun 19 17:56:11 BST 1997 joe@core.pavilion.net:/usr/src/sys/compile/PAV2_2X96d i386 >Description: 1.2.0 complains of: httpd: cannot determine the IP address of the specified ServerName (www.pavilion.co.uk). It does this when the number of virtual servers is > 124. At 124 it runs but doesn't service connections. >How-To-Repeat: >Fix: I assume that this is an O/S limit, but I don't know which one. I've run the 'httpd' from an entirely unlimited shell: dougal# unlimit dougal# limit cputime unlimited filesize unlimited datasize 131072 kbytes stacksize 65536 kbytes coredumpsize unlimited memoryuse unlimited memorylocked unlimited I've also recompiled the FreeBSD kernel to include: maxusers 256 # joe/pavilion/19970529: Our machines have 96Mb of memory options "MAXMEM=96*1024" # joe/pavilion/19970619: jungled the maximum number of connections options SOMAXCONN=1024 This should be plenty, but I still have this problem. %0 >Audit-Trail: State-Changed-From-To: open-analyzed State-Changed-By: marc State-Changed-When: Fri Jun 20 08:49:11 PDT 1997 State-Changed-Why: Talking with submitter on FreeBSD mailing lists about same problem. Probably FD_SETSIZE limit. State-Changed-From-To: analyzed-closed State-Changed-By: dgaudet State-Changed-When: Sat Jun 28 15:43:08 PDT 1997 State-Changed-Why: A workaround and documentation has been committed to 1.2.1-dev and 1.3-dev. Thanks Dean >Unformatted: