Jacob Gardiner

1 February 2013 2:09 PM

Hi all,

Firstly - Cheers Adam et all for the development.

I'm looking to see if anybody has had any luck properly monitoring a Juniper SRX cluster. We've got a couple up and running at the moment, in addition to some J-series units that we'll probably replace with another pair of SRXs in the future but we've been running into some issues.

Basically when adding the host to observium, everything works a treat as usual. Devices get detected properly, node0 & node1's routing engines get detected properly and we can see two different graphs for each in relation to processor, memory, temperature etc. However when polling the interfaces, we get a zero response for bps.

This behavior happens on all interfaces except the lo0.X interfaces and fxpX interfaces. (They show graphs and results).

I've run a poll via cmdline with the -d flag and can see that the poller is definitely seeing a 0 response from all of the other interfaces, output looks something like:

<paste>
SQL[INSERT INTO `ports-state` (`port_id`)  VALUES ('2151')] state inserted
SQL[UPDATE `ports-state` set `poll_time` ='1359075465',`poll_period` ='153',`ifInOctets` ='',`ifInOctets_rate` ='0',`ifInOctets_delta` ='0',`ifOutOctets` ='',`ifOutOctets_rate` ='0',`ifOutOctets_delta` ='0',`ifInErrors` ='',`ifInErrors_rate` ='0',`ifInErrors_delta` ='0',`ifOutErrors` ='',`ifOutErrors_rate` ='0',`ifOutErrors_delta` ='0',`ifInUcastPkts` ='',`ifInUcastPkts_rate` ='0',`ifInUcastPkts_delta` ='0',`ifOutUcastPkts` ='',`ifOutUcastPkts_rate` ='0',`ifOutUcastPkts_delta` ='0',`ifInOctets_perc` ='0',`ifOutOctets_perc` ='0',`ifOctets_rate` ='0',`ifUcastPkts_rate` ='0',`ifErrors_rate` ='0' WHERE `port_id` = '2151'] 1 updated
Port reth0(568) VLAN ==
 ifInOctets (0 B) 0 Bps 153 secs

 ifOutOctets (0 B) 0 Bps 153 secs

 ifInErrors (0 B) 0 Bps 153 secs

 ifOutErrors (0 B) 0 Bps 153 secs

 ifInUcastPkts (0 B) 0 Bps 153 secs

 ifOutUcastPkts (0 B) 0 Bps 153 secs
bps(0bps/0bps)bytes(0B/0B)pkts(0pps/0pps)OK u:0.00 s:0.03 r:50.09
</paste>

My first impression is that it's the MIB incorrectly polling the device, Can anybody confirm?

Cheers,

JG

Adam Armstrong

1 February 2013 2:06 PM

Further up the debug output you can see the data returned from the switch.
You're looking for ifHCInOctets && ifHCOutOctets.

If the SRX is fucking those up, which I can totally believe for a cluster,
you need to talk to Pooniper about it.

adam.

On Fri, 01 Feb 2013 14:09:32 +1100, Jacob Gardiner
<jacob@jacobgardiner.com> wrote:

Hi all,

Firstly - Cheers Adam et all for the development.

I'm looking to see if anybody has had any luck properly monitoring a 
Juniper SRX cluster. We've got a couple up and running at the moment, in

addition to some J-series units that we'll probably replace with another

pair of SRXs in the future but we've been running into some issues.

Basically when adding the host to observium, everything works a treat as

usual. Devices get detected properly, node0 & node1's routing engines 
get detected properly and we can see two different graphs for each in 
relation to processor, memory, temperature etc. However when polling the

interfaces, we get a zero response for bps.

This behavior happens on all interfaces except the lo0.X interfaces and 
fxpX interfaces. (They show graphs and results).

I've run a poll via cmdline with the -d flag and can see that the poller

is definitely seeing a 0 response from all of the other interfaces, 
output looks something like:

<paste>
SQL[INSERT INTO `ports-state` (`port_id`)  VALUES ('2151')] state

inserted

SQL[UPDATE `ports-state` set `poll_time` ='1359075465',`poll_period` 
='153',`ifInOctets` ='',`ifInOctets_rate` ='0',`ifInOctets_delta` 
='0',`ifOutOctets` ='',`ifOutOctets_rate` ='0',`ifOutOctets_delta` 
='0',`ifInErrors` ='',`ifInErrors_rate` ='0',`ifInErrors_delta` 
='0',`ifOutErrors` ='',`ifOutErrors_rate` ='0',`ifOutErrors_delta` 
='0',`ifInUcastPkts` ='',`ifInUcastPkts_rate` ='0',`ifInUcastPkts_delta`

='0',`ifOutUcastPkts` ='',`ifOutUcastPkts_rate` 
='0',`ifOutUcastPkts_delta` ='0',`ifInOctets_perc` 
='0',`ifOutOctets_perc` ='0',`ifOctets_rate` ='0',`ifUcastPkts_rate` 
='0',`ifErrors_rate` ='0' WHERE `port_id` = '2151'] 1 updated
Port reth0(568) VLAN ==
  ifInOctets (0 B) 0 Bps 153 secs

  ifOutOctets (0 B) 0 Bps 153 secs

  ifInErrors (0 B) 0 Bps 153 secs

  ifOutErrors (0 B) 0 Bps 153 secs

  ifInUcastPkts (0 B) 0 Bps 153 secs

  ifOutUcastPkts (0 B) 0 Bps 153 secs
bps(0bps/0bps)bytes(0B/0B)pkts(0pps/0pps)OK u:0.00 s:0.03 r:50.09
</paste>

My first impression is that it's the MIB incorrectly polling the device,

Can anybody confirm?

Cheers,

JG


_______________________________________________
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