Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

I've happened accross a odd issue this morning and I'm not sure where to go from here.

We use a couple of site up time monitoring services (pingdom, siteuptime) that essentially ping our server to check that its 'up'. Over the weekend both of these tools reported fairly large periods of outages.

Having looked through the Apache Access log for the reported down times I can see the requests coming in from both of these services. There are no errors reported in the error log against these times so I can only assume that the request was served as per normal.

I have already checked with our server hosts and they confirm that they had no downtime.

Is there such a thing as a sent log that details the actual pages that Apache has served? Can you think of any other avenues that I can check to see where this problem may have come from?

Many thanks

Rob

[EDIT] Extract from access_log

85.17.156.11 - - [12/May/2012:17:31:00 +0100] "GET / HTTP/1.0" 200 12056 "-" "Pingdom.com_bot_version_1.4_(http://www.pingdom.com/)"

share|improve this question

3 Answers 3

up vote 2 down vote accepted

there is a field on pingdom with the details of the previous error;

pingdom

you can get the content of the error message;

enter image description here

that should give you a clue as to whether the requests are 4xx/5xx errors, or whether it was some network problem for your ISP or Hosting company

share|improve this answer
    
Pingdom (and possibly siteuptime) can be configured to require a certain string to be in the response. This can cause them to say your site is down, even when you are serving them 200 responses. –  Ladadadada May 14 '12 at 10:39
    
yep, but they log the reason... –  Tom H May 14 '12 at 10:39
    
Precisely... that's why Pingdom's web interface is the best place to check for errors. –  Ladadadada May 14 '12 at 10:42
    
I've managed to get the account details for Pingdom and as you say they've got a nice log. However it simply says "Timeout (> 30s)" which doesn't really help me that much, any ideas? –  Rob Forrest May 14 '12 at 10:45
1  
If pingdom log shows timeouts, then its unlikely that there are corresponding hits in the CustomLog file. This is generally a network issue. If you have pingdom checking a few sites or multiple servers on your network, and they are all down. At this point you have to escalate your your network admin. –  Tom H May 14 '12 at 11:25

You can see or setup access log with LogFormat and CustomLog config directives.

See CustomLog

share|improve this answer
    
Which directives do you think I should add into our access log? Would '%X' or '%O' do the job? –  Rob Forrest May 14 '12 at 9:47
    
Probably. Also, you can enable forensic logs and whether your requests are completing successfully with check_forensic. –  Vadym S. Khondar May 14 '12 at 10:02

Apache logs the exact requests served at the time the request was completed, by default in a CLF format access_log file.

In newer versions of apache there are increasingly more detailed log modules available (mod_log_io and mod_log_forensic) that allow you to log more detail about the request process.

See their respecitive documentation entries for more details.

share|improve this answer
    
I've been looking through the access logs already and have got entries like the one that I've added to my question. As far as I can tell, the request has been processed ok (status 200) and 12056 bytes was sent back. Clearly at some point pingdom simply didn't get this response. Is there anything that I can do see if this is our problem or theirs? Thanks –  Rob Forrest May 14 '12 at 9:56
    
If apache returns the result you intended, it's not a problem on your side :) –  adaptr May 14 '12 at 16:08

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.