Even after upping the VPS nodes the internal server error happens at some intervals
following is the latest config of the vps that we are hosted on
The numbers look normal to me except the last one, the load average. At load average 1.00 the CPU is 100% busy. Your load averages vary from 0.86 to 1.19, i.e. sometimes overloaded.
The question is what makes the CPU work that hard, your users or other customers of the shared server? This is not easy to find out because you don't have full access to a shared server. What is the usage pattern when this happens? See http://docs.moodle.org/22/en/Performance_FAQ#How_do_you_define_.22concurrent_users.22.3F.
P.S. I have created a new thread for your question, splitting it from "Moodle 2.0: Servre RAM momery showing full" http://moodle.org/mod/forum/discuss.php?d=198462.
Thanks for the reply vishwanath, honestly I am a bit confused here too. Have tried out the recommended settings also given in the forums -
query_cache_type = 1
query_cache_size = 36M
query_cache_min_res_unit = 2K
cache with table_cache = 512
also , since its a vps , its a managed server, i don't have the full access but i can ask them to install anything and they will do it.
the main problem started with the database lockup, from the log given to us by the hosting support team we came to know the session table was choking up hence we moved to a file based session.After moving to a file based session, the problem of lockup sorted out but, the 500 error became frequent.
can you help me out with some pointers on where to begin with the debug, if there is something i can check out in the course logs let me know.