Yes, you are right, -Cr2 is extremely slow. What I meant here is, I am not able to get the time down to less than ~50sec, when playing with different -Cr values, betting it -Cr20 (default), -Cr60 or -Cr100.

No -Cr60 - ifEntry :-
real    1m0.221s

No -Cr60 - ifXEntry :-
real    0m49.692s

-Cr2 - ifEntry :-
real    4m27.568s

-Cr2 - ifXEntry :-
real    4m10.640s

Is it possible to tell the script not to poll the unwanted data?

Thanks. 

On Fri, Jan 30, 2015 at 1:56 PM, Adam Armstrong <adama@memetic.org> wrote:
I find that very hard to believe.

-Cr2 would be unbearably slow, since the default is 20.

adam.


On 2015-01-30 21:55, Jaynee Shah wrote:
I have tried it with all different values, no -Cr60, -Cr2, -Cr100, and
even more. There is absolutely no difference.

On Fri, Jan 30, 2015 at 1:11 PM, Mike Stupalov <mike@observium.org>
wrote:

try same commands without '-Cr60' option
and second one with snmpwalk (instead snmpbulkwalk).

On Sat, Jan 31, 2015 at 12:00 AM, Jaynee Shah
<jayneejshah@gmail.com> wrote:

Hi,

We have heavily loaded Brocade MLXe routers, and polling time is
very high for these devices. The same is seen for few Arista
switches (at longer distances) as well.

This certainly is related to IF-MIB, trying to poll lots of data
statistics for all interfaces, resulting in ports.inc.php module
run time of ~150sec+. Ideally, we want to reduce this polling time
to <60sec.

time /usr/bin/snmpbulkwalk -Cr60 -v2c -c 'jtv' -OQUs -m IF-MIB -M
/home/jtv/observium//mibs/rfc:/home/jtv/observium//mibs/net-snmp
'udp':'br02.lhr02':'161' ifEntry
real 0m57.616s
user 0m0.044s
sys 0m0.012s

time /usr/bin/snmpbulkwalk -Cr60 -v2c -c 'jtv' -OQUs -m IF-MIB -M
/home/jtv/observium//mibs/rfc:/home/jtv/observium//mibs/net-snmp
'udp':'br02.lhr02':'161' ifXEntry
real 0m52.763s
user 0m0.032s
sys 0m0.024s

We do not care about some of the interface statistics, for
example, mtu, speed, duplex, promiscuous mode, trunk, vlan, etc.
We only want interface traffic in/out and error statistics.
Is there a way to selectively modify any discovery/polling/MIB
files, and only keep the statistics we are interested in?

Thanks,
Jaynee
_______________________________________________
observium mailing list
observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium [1]

--

Mike Stupalov
http://observium.org/ [2]
_______________________________________________
observium mailing list
observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium [1]



Links:
------
[1] http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
[2] http://observium.org/

_______________________________________________
observium mailing list
observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

_______________________________________________
observium mailing list
observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium