Search results

hmm ... 120M or so. how about "ulimit -a" ?

pmap pid-of-lsphp5-process

how about set both to 768M

lsphp itself has soft/hard limit: External App->lsphp5: Memory Soft Limit (bytes) Memory Hard Limit (bytes) increase these limit and see if it helps.

looks like only 3 variable available: SERVER_ROOT, VH_ROOT, VH_NAME can you be more specific on how you load balance web servers. will forward your question to other staff.

add --with-libdir=lib64 to build php

I'd give a quick guess: xcache caused the problem. can you try disable xcache and see if the issue still exist.

please check: lsof -p 9492

assume php in suExec mode? since lsws 4.0.2: maybe just what you want.

this is the problem ... most likely it's related to bug in php script. we can login your server to investigate, in case it really a bug of lsphp.

create a phpinfo() page, at the bottom, you can find all the server variables.

what's result of pstree -p if lsphp5 process has no parent, then it's a runaway process, just kill it. you can strace the runaway process, what it is doing. a bug php script may cause dead lock and waiting for mutex etc forever.

you can use php_flag php_admin_flag php_value php_admin_value to customize default system-wide php.ini for each virtual host. http://php.net/manual/en/configuration.changes.php this is a lsphp special feature, comparing with other php fastcgi implementation.

how much memory of your VPS ? wordpress may need lsphp5. you are running lighttpd instead of apache ? if you only run your own wordpress, even no need to install kloxo+lighttpd,that will same memory. only lsws is needed. pstree -p also a utility to show what processes are running in...

can use rewrite rule to prevent it. in .htaccess at that directory: RewriteEngine on RewriteRule (.*)php$ - [F]

1>just check if apache process is running ps -ef|grep httpd on centOS, service httpd stop chkconfig httpd off 3>CentOS is better than Debian as a server. Debian may be better than centOS as a desktop --- please google the difference of them to get the accurate answer. 4>there is a...

not necessary. please try to add custom error page under apache + cPanel. if it works, should work under litespeed + cPanel.

we have one: admin console-->Actions-->Compiler PHP

yeah, network disk not recommended. We experienced very slow disk access when /tmp is NFS in a few customer's servers. using local disk, maybe a dedicated partition to store /tmp or for cache_dir only.

then set cache_dir to /tmp/ea you can clean up /tmp/ea daily