jesse
Touchdown! Greaser!
At about 8:57 PM CST the POA website quit responding. I started pinging the server and received consistent responses throughout the event.
That is about an average response time from my computer to the POA server. Based on this, I don't think this is a network issue.
I poked around a little more and at about 9:00 PM CST I had a decent test going. I was trying to download the main page from the POA server with wget. After 10 seconds it would timeout and try again. As you can tell the web server did not respond until 9:01:34. As soon as I was able to download a page the problem disappeared. Here is the test:
Based on this it's pretty obvious that whatever the problem is it is completely maxing out the resources of the server. The instant that the resource max out stops--is the instant POA responds. There isn't much in between--it pretty much either works or doesn't.
I really think it's a waste of time to sit here trying to optimize the database constantly. This would make sense if the problem wasn't at the exact same time every day. This isn't an issue of someone on POA generating too much load. This is an issue of another user on the physical server generating too much load.
I'm available to assist with resolving the issue and tools could be put in place that would allow us to graph when these problems are occurring in relation to other server resource specifications. But his is kind of a waste of effort. POA needs a new home.
Code:
Pinging pilotsofamerica.com [216.69.169.202] with 32 bytes of data:
Reply from 216.69.169.202: bytes=32 time=98ms TTL=40
Reply from 216.69.169.202: bytes=32 time=100ms TTL=40
Reply from 216.69.169.202: bytes=32 time=97ms TTL=40
Reply from 216.69.169.202: bytes=32 time=99ms TTL=40
I poked around a little more and at about 9:00 PM CST I had a decent test going. I was trying to download the main page from the POA server with wget. After 10 seconds it would timeout and try again. As you can tell the web server did not respond until 9:01:34. As soon as I was able to download a page the problem disappeared. Here is the test:
Code:
[root@server1 jangell]# date
Thu Oct 4 21:00:20 CDT 2007
[root@server1 jangell]# wget pilotsofamerica.com --timeout=10 --tries=0
--21:00:22-- http://pilotsofamerica.com/
Resolving pilotsofamerica.com... 216.69.169.202
Connecting to pilotsofamerica.com|216.69.169.202|:80... connected.
HTTP request sent, awaiting response... 302 Found
Location: http://www.pilotsofamerica.com/ [following]
--21:00:22-- http://www.pilotsofamerica.com/
Resolving www.pilotsofamerica.com... 216.69.169.202
Reusing existing connection to pilotsofamerica.com:80.
HTTP request sent, awaiting response... Read error (Connection timed out) in headers.
Retrying.
--21:00:33-- (try: 2) http://www.pilotsofamerica.com/
Connecting to www.pilotsofamerica.com|216.69.169.202|:80... connected.
HTTP request sent, awaiting response... Read error (Connection timed out) in headers.
Retrying.
--21:00:45-- (try: 3) http://www.pilotsofamerica.com/
Connecting to www.pilotsofamerica.com|216.69.169.202|:80... connected.
HTTP request sent, awaiting response... Read error (Connection timed out) in headers.
Retrying.
--21:00:58-- (try: 4) http://www.pilotsofamerica.com/
Connecting to www.pilotsofamerica.com|216.69.169.202|:80... connected.
HTTP request sent, awaiting response... Read error (Connection timed out) in headers.
Retrying.
--21:01:12-- (try: 5) http://www.pilotsofamerica.com/
Connecting to www.pilotsofamerica.com|216.69.169.202|:80... connected.
HTTP request sent, awaiting response... Read error (Connection timed out) in headers.
Retrying.
--21:01:27-- (try: 6) http://www.pilotsofamerica.com/
Connecting to www.pilotsofamerica.com|216.69.169.202|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: unspecified [text/html]
Saving to: `index.html'
[ <=> ] 70,792 258K/s in 0.3s
21:01:34 (258 KB/s) - `index.html' saved [70792]
[root@server1 jangell]#
I really think it's a waste of time to sit here trying to optimize the database constantly. This would make sense if the problem wasn't at the exact same time every day. This isn't an issue of someone on POA generating too much load. This is an issue of another user on the physical server generating too much load.
I'm available to assist with resolving the issue and tools could be put in place that would allow us to graph when these problems are occurring in relation to other server resource specifications. But his is kind of a waste of effort. POA needs a new home.