Hoping for some assistance with some Groupwise Graphs

[zesh]

Hello,

I have been using Cacti for a while now in a very generic capacity (read built in templates only) but was reading the forums an stumbled accross templates to graph Groupwise statistics.

The importing went well and graphs started to appear, but I suddenly realised they made no sense.

The messages routed since startup is an ever increasing number, but the graphs generating were going up and down all the time, with a scale that seems huge.

I verified through snmpwalk and through the debug log that the queries are returning the real numbers the MTA has relayed, so why would it be graphic completly wrong restults?

When the templates seemed to be playing up I tried using the generic OID template to create the graph, same result, although attempt 2 seems to close to the truth though all results should be whole numbers.

Debug log:
(Groupwise template found in this forum)
04/10/2006 10:12:06 AM - CACTID: Poller[0] Host[13] DS[198] SNMP: v1: hostname.domain.name.com, dsname: grpws_mta_total_msg, oid: .1.3.6.1.4.1.23.2.37.1.1.1.9.1, value: 41
(Generic SNMP OID Template)
04/10/2006 10:12:06 AM - CACTID: Poller[0] Host[13] DS[199] SNMP: v1: hostname.domain.name.com, dsname: snmp_oid, oid: .1.3.6.1.4.1.23.2.37.1.1.1.9.1, value: 41

SNMP Get/Walk performed shortly after cacti debug run:
hostname ~# snmpget -cpublic -v1 hostname.domain.name.com:161 .1.3.6.1.4.1.23.2.37.1.1.1.9.1
SNMPv2-SMI::enterprises.23.2.37.1.1.1.9.1 = Counter32: 70
snmpwalk -cpublic -v1 hostname.domain.name.com:161 .1.3.6.1.4.1.23.2.37.1.1.1.9.1
SNMPv2-SMI::enterprises.23.2.37.1.1.1.9.1 = Counter32: 72


Version 2c results in a timeout.

I would love to work out what is going on here, plenty of other graphs are working well and the results returned from the queries are real whole numbers so I have no idea how it has gone wrong.

Thanks for any assistance you can offer.

Zesh

[gandalf]

Not really sure if my statements will help you, but
- the m in the graphs denote milli. This results from calculating per second values while measuring 5 min interval data (this is: dividing by 300)
- stricly increasing numbers are denoted COUNTERs in rrdtool speech. rrdtool works by subtracting two consequtive numbers to get the difference related to that very time interval. See link for more info, if this matches your question.
As I have no Groupwise System, I can't follow your questions exactly
Reinhard