[rrd-users] Re: To DERIVE or not to DERIVE

[Alex van den Bogaerdt]

Michael MacFaden wrote:

> >The best way of dealing with this, if possible, is to detect real
> >counter resets and tell RRDtool about them.  At time of a reboot,
> >the counter value is known to be unknown.  Just after the reboot,
> >the value is known to be zero.  
> 
> Not if you are talking about counters polled from SNMP agent.
> 
> >From RFC 2578, page 23:
>    "Counters have no defined "initial" value..."

Oops.  OK, so the counter is unknown.  Indeed one needs to discard
the first interval between reboot time and poll time in that case.

> Any snmp poller should include sysUpTime in the get pdu. 
> If between any two polls the value is less than the previous, 
> throw out the poll value and insert an uknown entry for that 
> time period. 

But this would also make the next interval unknown.  My suggestion:
insert an unknown at that time minus one second, enter the fetched
value at that time.

cheers,
-- 
   __________________________________________________________________
 / alex at slot.hollandcasino.nl                  alex at ergens.op.het.net \
| work                                                         private |
| My employer is capable of speaking therefore I speak only for myself |
+----------------------------------------------------------------------+
| Technical questions sent directly to me will be nuked. Use the list. | 
+----------------------------------------------------------------------+
| http://faq.mrtg.org/                                                 |
| http://rrdtool.eu.org  --> tutorial                                  |
+----------------------------------------------------------------------+

--
Unsubscribe mailto:rrd-users-request at list.ee.ethz.ch?subject=unsubscribe
Help        mailto:rrd-users-request at list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/rrd-users
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi