Graphs based on SNMP

s24sean · Post by **s24sean** » Tue Feb 14, 2006 2:07 pm

I am having a bit of an issue and assuming im doing something im missing something stupid.

I have used mrtg in the past so i am just trying to make a graph with the data collected from a particular snmpget, OID 1.3.6.1.4.1.2021.8.1.101.1

I have created a data template and graph template but i am still getting big gaps in my graphs...

I have screenshots of my Data & Graph templates.

thanks to anyone who can tell me what im doin wrong

Post by **gandalf** » Tue Feb 14, 2006 2:19 pm

You entered a MAX value of 1000. This is not exceeded? Please switch Settings->Logging Level to DEBUG for one polling cycle and have a look at log/cacti.log. Find the OID and post the results including a few lines above and below
Reinhard

s24sean · Post by **s24sean** » Tue Feb 14, 2006 2:28 pm

oddly enough it worked fine that time, I will continue to tail the cacti.log in debug mode and next time it errors will post the results...

thanks for the help..

s24sean · Post by **s24sean** » Tue Feb 14, 2006 2:33 pm

cacti is going to make me a liar now...

seems to be working...

s24sean · Post by **s24sean** » Tue Feb 14, 2006 2:51 pm

ok... back to gaps...

02/14/2006 02:46:36 PM - CMDPHP: Poller[0] Host[117] DS[1175] SNMP: v1: x.x.x.x, dsname: barracuda_mail_in, oid: 1.3.6.1.4.1.2021.8.1.101.1, output: 1
02/14/2006 02:46:36 PM - CMDPHP: Poller[0] DEBUG: SQL Exec: "insert into poller_output (local_data_id,rrd_name,time,output) values (1175,'barracuda_mail_in','2006-02-14 14:45:02','U')"

s24sean · Post by **s24sean** » Tue Feb 14, 2006 2:52 pm

the odd part is, it stops making all graphs for that host

Post by **gandalf** » Tue Feb 14, 2006 3:44 pm

s24sean wrote:ok... back to gaps...

02/14/2006 02:46:36 PM - CMDPHP: Poller[0] Host[117] DS[1175] SNMP: v1: x.x.x.x, dsname: barracuda_mail_in, oid: 1.3.6.1.4.1.2021.8.1.101.1, output: 1
02/14/2006 02:46:36 PM - CMDPHP: Poller[0] DEBUG: SQL Exec: "insert into poller_output (local_data_id,rrd_name,time,output) values (1175,'barracuda_mail_in','2006-02-14 14:45:02','U')"

That is weird, indeed. SNMP returns 1 and poller_output gets U
Reinhard

Post by **rony** » Tue Feb 14, 2006 4:51 pm

Counter roll was detected, poller is going to pass 'U' to RRDtool to prevent a spike from occuring.

How is the availability rating for this host on the device list?

Post by **rony** » Tue Feb 14, 2006 4:53 pm

What type of values come over that SNMP OID? Does it a number of transactions at that time, or an increasing counter?

s24sean · Post by **s24sean** » Wed Feb 15, 2006 6:51 pm

it returns and integer between 0 and 30, representing the number of messages in the outgoing mail queue.

For example, if i would having mrtg do this i would use

TargetBarracudaINOUT]:1.3.6.1.4.1.2021.8.1.101.1&1.3.6.1.4.1.2021.8.1.101.2:public@[YourSERVERIP]
Title[BarracudaINOUT]: Barracuda IN / Out Queue
PageTop[BarracudaINOUT]: <H1>Barracuda IN / Out Queue</H1>
MaxBytes[BarracudaINOUT]: 20000000
ShortLegend[BarracudaINOUT]: messages
YLegend[BarracudaINOUT]: messages
Options[BarracudaINOUT]: growright, gauge, unknaszero

What i did that actually works, for some reason unknown to myself, is create a separate device that just does the monitoring of the queue.

I have a second barracuda box i am about to start testing on fresh,

Thanks for the help

s24sean · Post by **s24sean** » Wed Feb 15, 2006 7:07 pm

I am getting this in the log now

02/15/2006 07:01:43 PM - CMDPHP: Poller[0] Host[124] RECACHE: Processing 2 items in the auto reindex cache for 'x.x.x.x'.
02/15/2006 07:01:43 PM - CMDPHP: Poller[0] ASSERT: '10144<10037' failed. Recaching host 'x.x.x.x', data query #1
02/15/2006 07:01:43 PM - CMDPHP: Poller[0] Host[124] NOTICE: Spike Kill in Effect for 'x.x.x.x'.
02/15/2006 07:01:43 PM - CMDPHP: Poller[0] ASSERT: '10164<10041' failed. Recaching host 'x.x.x.x', data query #2
02/15/2006 07:01:43 PM - CMDPHP: Poller[0] Host[124] NOTICE: Spike Kill in Effect for 'x.x.x.x'.

Cacti

Graphs based on SNMP

Graphs based on SNMP

Who is online