Re: HOWTO monitor Apache with Monit.

[David Fletcher]

Hi Jan-Henrik

> Hi David. I have added your article at  
> http://www.tildeslash.com/monit/doc/apache.php. Please let me know if  
> you would like to change anything, such as linking to your homepage.  
> I'll be happy to do that. You can also send me a patch for the article  
> in cvs  
> http://savannah.nongnu.org/cgi-bin/viewcvs/monit/monit/web/doc/ 
> apache.php

Thanks for your email. The article looks good on the website, and I can see
that the changes (and -> or, etc) make it match the CVS version better.

Please can you link to my home page at www.megapico.co.uk

I have not yet tried the CVS version. My servers are running the patch as I
applied it before Christmas, and everything seems fine. I'll try the new code
out when I get time to update things.

I noticed from the CVS that you have included changes so that the user can
decided to trigger action using either "<" or ">", whereas I had hard-coded
these for the different tests. This is good since it allows the user to decide
when to take action, however, I think the HOWTO is still based on the
assumption of fixed "<" or ">" for the different tests.

A patch is below to update the HOWTO to the way the code should now work.

The WAP interface I submitted to the contrib directory was very useful
over Christmas. I was able to check everything was running while I was away
from home and without any normal internet connection.

Best wishes,

David.

-----------------------------------------------------------
--- apache_orig.php     2005-01-11 21:32:17.000000000 +0000
+++ apache.php  2005-01-11 21:38:45.000000000 +0000
@@ -149,15 +149,17 @@
 
 <li><h3>How the limits work</h3>
 
-<p>With one exception all of the percentage limits are the level
-above which action is triggered. As in the example above, an
-alert is sent when 25% or more of the Apache child processes
-are performing a DNS lockup.
-
-<p>The exception to this rule is when monitoring idle servers waiting
-for a connection. In this case it is much more useful to be alerted if
-there are too few free servers, so the action is triggered when the
-measured level is below the percentage limit.
+<p>Action can be triggered when each measured quantity rises above or 
+falls below the percentage limit. However, with one exception all of the 
+percentage limits are likely to be most useful if taken as the level 
+above which action is triggered. As in the example above, an alert is 
+sent when greter than 25% of the Apache child processes are performing 
+a DNS lockup.
+
+<p>The exception to this rule is when monitoring idle servers waiting 
+for a connection. In this case it is much more useful to be alerted 
+if there are too few free servers, so the action is best triggered 
+when the measured level is less than the percentage limit.
 
 <li><h3>What happens when the server is close to collapse?</h3>


-------------------------------------------------
Email: address@hidden
-------------------------------------------------