NCV RRD weirdness 🔗 link

OK, so I've finally delved into the murky waters (ok so I had to make
that sound scary and all fearsome and what not LOL) of writing my own
Hobbit server extension script.  And as part of the script, it collects
data.  I want to graph that data on hobbit.  So I set up the NCV
stuff... problem is that it doesn't create the RRD file correctly. 
Here's the relevant entries from hobbitserver.cfg:

TEST2RRD="cpu=la,disk,inode,qtree,memory,$PINGCOLUMN=tcp,http=tcp,dns=tcp,dig=tcp,time=ntpstat,vmstat,iostat,netstat,temperat
ure,apache,bind,sendmail,mailq,nmailq=mailq,socks,bea,iishealth,citrix,bbgen,bbtest,bbproxy,hobbitd,qstat=ncv"

# This defines which RRD files to include on the "trends" column webpage,
# and the order in which they appear.
GRAPHS="la,disk,inode,qtree,memory,users,vmstat,iostat,tcp.http,tcp,netstat,mrtg::1,temperature,ntpstat,apache,bind,sendmail,
mailq,socks,bea,iishealth,citrix,bbgen,bbtest,bbproxy,hobbitd,qstat"

#
# Ok these columns are from qstat and are used to graph (I HOPE!) the
columns via rrd.
#
NCV_qstat="avgping:GAUGE,curplayers:GAUGE"

And that status message that gets sent is this:

qstat status for host 123.123.123.123
curplayers : 0
avgping : 0

IE my ext script does this:

$BB $BBDISP "status $MACHINE.$COLUMN $COLOR `date`

${MSG}
curplayers : $cur_players
avgping : $avg_ping"

Where $MSG can be changed depending on circumstances.

However, when I dump the RRD file, I get this info:

<!-- Round Robin Database Dump -->
<rrd>
        <version> 0001 </version>
        <step> 300 </step> <!-- Seconds -->
        <lastupdate> 1133107889 </lastupdate> <!-- 2005-11-27 10:11:29
CST -->

        <ds>
                <name> qstatstatusforhost6 </name>
                <type> DERIVE </type>
                <minimal_heartbeat> 600 </minimal_heartbeat>
                <min> 0.0000000000e+00 </min>
                <max> NaN </max>

                <!-- PDP Status -->
                <last_ds> 0 </last_ds>
                <value> 0.0000000000e+00 </value>
                <unknown_sec> 0 </unknown_sec>
        </ds>

Any thoughts as to what I'm doing wrong?  This is on hobbit 4.1.2p1.

(1038:38)$ rpm -q -a | grep hobbit
hobbit-client-4.1.2p1-1.fc3.re
hobbit-4.1.2p1-1.fc3.re

-- 
Jeff Stuart
Network Admin 
MyInternetServices.com
1-800-300-HOST


--

Jeff Stuart wrote:

Ok... well it seems that I wasn't allowed to have my message as the
first line of the status report.  Once I removed that, it collected the
first data point.  IE curplayers.  But it's not collecting avgping.  Is
this correct?  Or should I send each of these data points as a data
message instead of in status?

-- 
Jeff Stuart
Network Admin 
MyInternetServices.com
1-800-300-HOST


--

Jeff Stuart wrote:

AH HAH... my problem is solved!!!

Seems that I had to have an extra carriage return (or make the value(s)
show up first in my status) for it to find TWO data points!

Now I get this in my rrd file:
        <version> 0001 </version>
        <step> 300 </step> <!-- Seconds -->
        <lastupdate> 1133148517 </lastupdate> <!-- 2005-11-27 21:28:37
CST -->

        <ds>
                <name> curplayers </name>
                <type> GAUGE </type>
                <minimal_heartbeat> 600 </minimal_heartbeat>
                <min> 0.0000000000e+00 </min>
                <max> NaN </max>

                <!-- PDP Status -->
                <last_ds> UNKN </last_ds>
                <value> 1.0000000000e+02 </value>
                <unknown_sec> 207 </unknown_sec>
        </ds>

        <ds>
                <name> avgping </name>
                <type> GAUGE </type>
                <minimal_heartbeat> 600 </minimal_heartbeat>
                <min> 0.0000000000e+00 </min>
                <max> NaN </max>

                <!-- PDP Status -->
                <last_ds> UNKN </last_ds>
                <value> 4.2000000000e+02 </value>
                <unknown_sec> 207 </unknown_sec>
        </ds>

Which is EXACTLY what I wanted!!!! :)  Might want to stress that Henrik.
:)  I had to do this in my ext script:

                $BB $BBDISP "status $MACHINE.$COLUMN $COLOR `date`

                curplayers : $cur_players
                avgping : $avg_ping
                ${MSG}"

Now if I hadn't sent ${MSG} it would have to look like this:

                $BB $BBDISP "status $MACHINE.$COLUMN $COLOR `date`

                curplayers : $cur_players
                avgping : $avg_ping
"

Now my next question.. LOL  I'd like to have two SEPARATE graphs for
this stuff...  I tried this:

[qstat]
        TITLE Current Number of Players
        YAXIS # of Players
        DEF:curplayers=qstat.rrd:curplayers:AVERAGE
        LINE2:curplayers#00CCCC:Current Players
        COMMENT:\n
        GPRINT:curplayers:LAST: \: %5.1lf (cur)
        GPRINT:curplayers:MAX: \: %5.1lf (max)
        GPRINT:curplayers:MIN: \: %5.1lf (min)
        GPRINT:curplayers:AVERAGE: \: %5.1lf (avg)\n
       
[qstat1]
        TITLE Average Ping
        YAXIS ms
        DEF:avgping=qstat.rrd:avgping:AVERAGE
        LINE2:avgping#00CCCC:Average Ping
        COMMENT:\n
        GPRINT:avgping:LAST: \: %5.1lf (cur)
        GPRINT:avgping:MAX: \: %5.1lf (max)
        GPRINT:avgping:MIN: \: %5.1lf (min)
        GPRINT:avgping:AVERAGE: \: %5.1lf (avg)\n

And that didn't work.  Now if I moved the qstat1 stuff (minus title and
yaxis) up into qstat.. I got one graph with two lines... I GUESS that's
ok but.. ;)

-- 
Jeff Stuart
Network Admin 
MyInternetServices.com
1-800-300-HOST


--

And as a followup.  Is there any way I can add some graphs from other
hosts?  IE a number of our game servers are on the same physical machine. 

IE something like:
123.123.123.123 <--- Main IP
123.123.123.124
123.123.123.125
123.123.123.126

For the game servers on say .124 and .126 I'd also like the CPU/Mem
graphs on the trends page also.  Is this possible?

-- 
Jeff Stuart
Network Admin 
MyInternetServices.com
1-800-300-HOST


--