Now that I am back in the USA, I think my problem may be fixed (or helped out at least)... Looking through the posts in this forum, I did several things, I do not know which one did the most to help:
- Through WHM, increased the max amount of RAM cpanel may use to 256 MB (Why is this limit needed? Speaking as a programmer, it sounds like a cheap hack to hide a memory leak in the cpanel daemon, certainly it does not inspire confidence in the cpanel programming).
- Did upcp --force to force a cpanel reinstall.
- Updated RVSkins (I'd be extremely surprised if this had anything to do with the problem at hand).
After doing these things, my server has been up for about 12 hours, which is a vas improvement over 5 minutes. So far, so good. I still get a bunch of these two messages in the error_log file:
Child [22593]: exited with signal 13 [BTW: Does this have to do with the max RAM setting?]
Internal Error: "" Normal Connection Timeout (Probably closed browser on a HTTP/1.1 session) at cpsrvd.pl line 1572.
But my server has not gone down again, even though I get these messages.
BTW: the cpanel logs files are in /usr/local/cpanel/logs, and if your error_log file is empty (as mine was) you have to enable writing to it in WHM.