I was having some issues running ipband on a system that had some other monitoring utilities/servers on it, so I decided to just start fresh and reboot the server, after the reboot everything came back up (and my ipband issue was rectified) However, I have lost snmp communication with my firewall, cacti, mrtg and snmpwalk fail to return any data from it. It appears cacti is still collecting data from the snmp service on the server itself, however, it will not collect any from the firewall/router. mrtg has just been outputting the same number since the reboot (indicating it can't talk to the firewall) and when I try snmpwalk I get a timeout error. I have recreated the snmp group on the firewall, restarted the firewall, enabled/disabled/re-enable snmp access to firewall. Changed settings in cacti, then changed them back. Added a new source in cacti for the device and received the same error.

The device is a Juniper SSG140 running netscreen, the server is an HP Xeon quadcore with 4 GB of RAM running one VM in addition to the monitoring utilities, no issues with the VM or anything else on the server, just the snmp monitoring utilities.

Note: I have two NICS in this unit and one NIC is connected to the network for network access and the other NIC is connected running in promiscuous mode for the monitoring utilities.

Any idea what may be going on here or how to rectify it? I have googled all ready but didn't see anything that seemed applicable, although I may have overlooked something.

Thanks in advance for the assistance.

Alex

Dunno if this will help but you could try running tcpdump on port 161, running a snmpwalk and then analysing the pcap using wireshark to see what the server response actually is. Also tail -f /var/log/messages | grep -i snmp while you are running snmpwalk

Actually it was very odd, but shortly after I posted this snmp data started being collected again and I have not had a major issue since. Thank you for the reply though.