On 20 Aug 2009, at 21:03, Shane Hathaway wrote:
Wade Preston Shearer wrote:My server goes down from time to time. When it does, it is still on but none of the services respond (no response from SSH, Apache, ping, etc). The server is just a web server. Each time, I reboot it and it comes back up just fine. I'm trying to figure out what keeps taking it down. I've wondered if it was a bad hard drive in the RAID, but after looking at the Apache logs just now and seeing a plethora of memory leak notices, I'm wondering if it's a memory issue. Would a bad script that is leaking memory cause the machine to run out of memory and then freeze up or cause all of the services to become non-responsive?No. A program that simply eats memory can cause thrashing and activatethe kernel's OOM killer, but it can't kill the network stack unless there's also a kernel or hardware bug.
I received the following in my nightly LogWatch report:
--------------------- Kernel Begin ------------------------ WARNING: Kernel Errors Present serial: probe of 00:06 failed with error -16...: 2 Time(s) ---------------------- Kernel End -------------------------
smime.p7s
Description: S/MIME cryptographic signature
/* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
