On 9/23/06, Dave Mabe <davemabe@xxxxxxxxx> wrote:
OK, the problem happened again and I was able to pull some information from server-status. First of all, here's the appropriate part of my config: MinSpareServers 3 MaxSpareServers 5 StartServers 3 MaxClients 50 MaxRequestsPerChild 30
MaxRequestsPerChild is extremely low. Are you having severe memory leakage problems?
As you can see from process.list.txt, Apache is completely ignoring MaxSpareServers and creating a ton of child processes until it completely runs out of memory. Sometimes it gets in this state a couple times a day, other times it will go for 10 days without the problem occuring.
I don't see any evidence of MaxSpareServers being ignored. I see evidence of a bunch of processes serving requests, which means they aren't "spare". Then the only limit is MaxClients, which is 50 and is not being exceeded. The server-status displays do look weird. But this could be caused by any number of things. For example, it is possible that your perl scripts are just not able to keep up with the load because of some holdup on backend database connections or something like that. The main difference I see between the "good" state and the "bad" state is that there are many more active requests in the "bad" state, and they don't seem to be making any progress. To figure out what a particular apache process is up to, you can use the instructions here: http://httpd.apache.org/dev/debugging.html#backtrace Joshua. --------------------------------------------------------------------- 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
|