[mrtg] Re: vpdn oids changing from time to time

[Tagle, Michael F.]

Thanks you very much for your response. Your last name is such that familiar
to me, it's the name my university(Adamson University). You've given me a
good idea do study Perl programming in w/c i thinked the solution on most of
my problem in Monitoring. By the way, i know your geek in SNMP...i got one
more serious problem. I found OID to monitor MS SQL Database disk allocation
and I can browse that w/ GETIF MIB Browser and I think the data is right but
when
try using SNMPUTIL.EXE, im getting this error "Error: errorStatus=2,
errorIndex=1".  and when i try to use it directly to MRTG i cant get any
data and of course and im getting also error somewhat like "NoSuch
name...bla bla bla...". Somebody says maybe i did not included the proper
instance. but im what i got from GETIF is complete. what is wrong w/ that?
thanks again...

-----Original Message-----
From: Rich Adamson [mailto:radamson at routers.com]
Sent: Friday, March 01, 2002 7:12 PM
To: 'mrtg at list.ee.ethz.ch'
Subject: [mrtg] Re: vpdn oids changing from time to time




> is anyone there experienced a using variable oids? i thought i'm very
lucky
> because at last i've found oids for our vpdn tunnel w/ cisco. the oid of
> every tunnel is changing from time to time so how we could use that w/
MRTG?
> any ideas regarding this would highly be appreciated.

I've not worked with the vpn stuff, but would imagine it can't be done
realistically with a stock mrtg.

SNMP oid's can be very dynamic (as you are seeing), with a specific oid
instance only being active and available at the time a specific session
(vpn tunnel) is in progress. In many cases, an snmp index is available
elsewhere within the mib variables that indicates what specific instances
are active at a specific time. In order to find that index, one would
have to understand Cisco's mib variable logic comparing the documented
mib to what is seen from a repetitive GetNext mib walk. (Most companies
do not document the inter-relationships of their mib tree elements 
other than the comments included in each mib variable definition.)

Needless to say, mrtg does not have logic built into it to automatically
find Cisco's logic and present you with a "static" oid needed for mrtg 
polling.

Two approaches that might partially work include:
1. write your own perl (or whatever) script to do a GetNext mib walk of
   each oid occurance matching all but the last index number within the
   oid. Use the results from the script to poll the variables wanted.
2. Find and understand the relationships/logic that Cisco used to design
   the mib tree, and use that logic to find/poll the variables wanted.

In either case you probably won't like the results. The issue with
dynamic oid's is that a single oid can come and go in the middle of your
mrtg five-minute (or whatever) polling cycle and you won't know it.
Second, an oid instance many exist in one polling cycle and disappear
before the next polling cycle; what kind of logic would you implement
to address those cases?  Even if you shorten the polling cycle to 
something less than a minute, you would still need to address those 
two issues within your script. Probably not realistic.


--
Unsubscribe mailto:mrtg-request at list.ee.ethz.ch?subject=unsubscribe
Archive     http://www.ee.ethz.ch/~slist/mrtg
FAQ         http://faq.mrtg.org    Homepage     http://www.mrtg.org
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi


--
Unsubscribe mailto:mrtg-request at list.ee.ethz.ch?subject=unsubscribe
Archive     http://www.ee.ethz.ch/~slist/mrtg
FAQ         http://faq.mrtg.org    Homepage     http://www.mrtg.org
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi