Hi, I discovered what was wrong and thought I'd summarize it here. We were seeing "[warn] (128)Network is unreachable: connect to listener" in our logs whenever we did a restart of any kind. I changed our Listen from: Listen 80 to: Listen 0.0.0.0:80 This disabled IPv6 listening code and it ended our errors in our logs. It also fixed the issue with SIGUSR1 and the children not exiting. So it would seem that 'Listen 80' on a Solaris 8 platform (with IPv6 disabled on the O/S) would not allow the children to exit normally after a SIGUSR1. Yet a SIGHUP would. -----Original Message----- From: Houle, Michael Sent: Thursday, March 03, 2005 1:47 PM To: 'users@xxxxxxxxxxxxxxxx' Subject: SIGUSR1 handling broken ? Hi, I just noticed that whenever I give the parent httpd a SIGUSR1 to refresh our config, I get 5 more spare servers. I have one server that has over 95 children running. We give a USR1 once each week during logfile rotation. This is not an intermittent problem. We are noticing no other problems so it appears that the children are coming home to reload config as they should. Our Maxspareservers is set to 10 so this setting is not being respected. A HUP functions normally. We are running: php 4.3.10 httpd 2.0.53 solaris 8 gcc 3.4.3 Anybody see this behaviour before ? Thanks in advance. --------------------------------------------------------------------- The official User-To-User support forum of the Apache HTTP Server Project. See <URL:http://httpd.apache.org/userslist.html> for more info. To unsubscribe, e-mail: users-unsubscribe@xxxxxxxxxxxxxxxx " from the digest: users-digest-unsubscribe@xxxxxxxxxxxxxxxx For additional commands, e-mail: users-help@xxxxxxxxxxxxxxxx
|