From: Dan Dubovik <dandubo@gmail.com>
> If you are running into issues with too many httpd processes spinning up,
Why do people call problems "issues"?
> You could run into similar constraints if you hit MaxClients (assuming
> apache here?) Is the cause of the load from the web server spinning up a
bunch
> of processes all of a sudden? Or from the code it is attempting to parse /
> deliver? Increasing the number of servers that are started up could also
> help in preventing a spike of load if it is the former.
256M RAM means there's a hard and relatively low limit on the number of apache
processes you can start. About 10 or 12 seems to be the max here. I tried
adjusting MaxClients, ServerLimit, MaxRequestsPerChild, *buffer_size,
query_cache* , and so forth in httpd.conf and my.cnf according to what other
people said worked for them in the slicehost wiki and forum. Nothing helped.
RAM is AFAICT the limiting factor on a cheap slice, since almost nobody runs
production stuff on boxes with that little RAM.
> it seems that having that 30 second sleep in the middle of a bunch
> of traffic, would go against the (supposed) goal of driving traffic
30 second downtime, or 4 hour downtime? Traffic is secondary to having the
box stay up without manual intervention; if some people get 404s for a while,
oh well. The sleep() could probably be reduced to 15. I know 30 works
though.
If this were some sort of profit-generating site, I'd have sprung for a more
expensive slice with more RAM, and these problems probably never would've come
up. However, there are probably people out there who don't have a lot of
money and are facing similar problems.
--