[Nagiosplug-help] check_http problem with PDF documents
Ham, Gerrit van IA60
g.van.ham at nijmegen.nl
Thu Sep 13 11:42:04 CEST 2012
Hello list members,
We have been using the Nagios plugin 'check_http' for some years now without any problems but yesterday I found something strange with one of our checks involving a page that contains a PDF document. Sometimes this check reports an error and at other times the result is OK. The page is static and some content is delivered in all cases but the text I'm looking for with the '-s' option is not always found. So I did some tests, including the '-v' option and saving the output to a file. If I run this check 4 times I get 4 different output files. If I retrieve the same page with 'wget', the output is identical all the time.
We run two Nagios hosts (prod on RHEL6 with check_http v1.4.15 and test on RHEL5 with v1.104) and both behave the same.
As a further test, I placed the PDF file on my test Nagios host and ran the check again. If I run 'check_http' against the local server, the output file differs from the original PDF file, but the output is consistent during all test. If I run the check from my other Nagios server the output file differs. I also tested this the other way around and again, output is consistent if the page is retrieved from the local server. Is doesn't matter whether I use hostname, FQDN or IP-address.
Next I ran a test against an external PDF document:
check_http -H www.netways.de -u '/uploads/media/Ton_Voon_An_Active_Check_on_the_Status_of_the_Nagios_Plugins01_02.pdf' -v > test.txt
I get a different 'test.txt' on every run.
Has anybody else experienced this problem? Is there a way to get around this problem and retrieve the complete PDF document in all cases using 'check_http'?
Many thanks in advance for all help.
Gerrit van Ham
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Help