[Nagiosplug-devel] Check_Http Plugin BUG ?

[Stefan Meier]

Hi Developer,

we use the Nagios “Check_Http” for Performance Tests of our Webapplication.
The Webapplication is parted over 4 Computer Centers. 

One Computer Center want to prove a fault programming with the used asprintf
Function in your "check_http.c" plugin Sourcecode. 

They changed the Value "MAX_INPUT_BUFFER" from 1024 to 8196 on the Headerfile
common.h. They got on a 800 MHZ Server with modified Check_Http Plugin a 120 ms 
shorter Response Time from our Webapplication. 

They told me the Response Html Data of a query will be parted into pieces (MAX_INPUT_BUFFER) 
and processed by the asprintf Function to get the elapsed Time of the query.

The Opinion is that the asprintf Function under Linux used a lot of Time
with 1024 Byte Pieces (allocating Memory) to process the asprintf Function.
Using the 8196 Byte Pieces the calling of the asprintf Function will be 
lower as with 1024 Pieces and the Response Time (Performance Data)
of the Webapplication will appropriately shorter.

This Problem only occurs when the Response Html Data is greater than 100 Kb.

Which "MAX_INPUT_BUFFER" Value should be used in the check_http Plugin
to get real performance Data on a Web Query? Should the Check_Http Plugin
be reprogrammed?

What do you think?
Is the Opinion of the Computer Center correct, or is this a dubious workaround?

Thanks for your Response

Greets
Stefan
_______________________________________________________________
SMS schreiben mit WEB.DE FreeMail - einfach, schnell und
kostenguenstig. Jetzt gleich testen! http://f.web.de/?mc=021192