[hobbit] Custom RRD graphs

Hubbard, Greg L greg.hubbard at eds.com
Thu Nov 30 20:55:03 CET 2006


Why bother when Hobbit will use your echo output to do that?
 
But there is more to graphing than simply creating the RRD files.  Next
you have to create graph definitions that will use your new files, and
these graphs have to work.  Then you can add your new graphs to the
Hobbit server so they will show up with the associated test (TEST2RRD
variable) and/or in the trends display.
 
You say you see the data being echoed -- are the RRD files not getting
created/updated?  Did you add your rrd-ext to the right place on the
server?  Did you add it to hobbit-rrd for both data and status messages?
 
(or whatever, I am going from dim memory)
 
GLH


________________________________

	From: Ger Maguire [mailto:ger at fwi.ie] 
	Sent: Thursday, November 30, 2006 1:44 PM
	To: hobbit at hswn.dk
	Subject: RE: [hobbit] Custom RRD graphs
	
	
	Guys, When creating custom scripts, has anyone just used
rrdcreate and rrdtool update to directly output the data to the rrd
files ?
	 
	If the the tests are all residing on the one machine and not
polling data from other clients is it an easier way to update the graphs
as I just can't get the status script to update for me ?
	 
	Thanks,
	Ger.

________________________________

	From: Ger Maguire [mailto:ger at fwi.ie] 
	Sent: 29 November 2006 20:13
	To: hobbit at hswn.dk
	Subject: [hobbit] Custom RRD graphs
	
	
	Hi folks,
	 
	I have created a custom script for capturing CPU utilisation
from a device and updated all the relevant files but I cannot get my rrd
files to update correctly.
	 
	An initial rrd file gets created but the data does not get input
into it.
	 
	I have created a rrd-ext sript which is running and is capturing
the incomming status data and by looking at the /tmp files I can see
that the echo outputs are being generated correctly every 5 mintes.
	 
	----> rrd-ext.sh
	 
	HOSTNAME="$1"
	TESTNAME="$2"
	FNAME="$3"
	 
	if [ "$TESTNAME" = "acpu" ]
	then
	        echo "`date`" >> /tmp/hobbit.$HOSTNAME.$TESTNAME
	        echo "$FNAME" >> /tmp/hobbit.$HOSTNAME.$TESTNAME
	        # Analyze the message we got
	        # ACPU=`grep "connected" $FNAME | awk '{print $14}'`
	        ACPU=`grep "The 5" $FNAME | cut -d"(" -f2 | cut -d"."
-f1`
	        # The RRD dataset definitions
	        echo "DS:acpu:GAUGE:600:0:100"
	        # The filename
	        echo "acpu.rrd"
	        # The data
	        echo "$ACPU"
	        echo "$ACPU"  >> /tmp/hobbit.$HOSTNAME.$TESTNAME
	fi
	 
	 
	The value I want to graph is simply a cpu % counter, have I got
the above values and setting correct for this ?
	 
	Any help or pointers appreciated.
	 
	 
	Output that this produces is below showing that it is getting
the data correctly from the status input.
	 
	---> /tmp/hobbit.$HOSTNAME.$TESTNAME
	 
	Wed Nov 29 19:54:07 GMT 2006
	/usr/lib/hobbit/server/tmp/rrd_msg_13879
	14
	Wed Nov 29 19:59:05 GMT 2006
	/usr/lib/hobbit/server/tmp/rrd_msg_26569
	25
	Wed Nov 29 20:04:07 GMT 2006
	/usr/lib/hobbit/server/tmp/rrd_msg_2788
	17
	Wed Nov 29 20:09:12 GMT 2006
	/usr/lib/hobbit/server/tmp/rrd_msg_11944
	9
	
	 
	 
	Ger.
	 
	 
	 
	 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.xymon.com/pipermail/xymon/attachments/20061130/5b420c6a/attachment.html>


More information about the Xymon mailing list