Cisco WRV-210

[str8edgedave]

After reading http://forums.cacti.net/post-164174.html, I am able to say that I am experiencing the same issue.

Unlike the other poster, I am comfortable with Linux and Cacti, and I have had J.P./Linegod look at this device and my cacti configuration closely. That said, I'm still working at learning SNMP. For the record, I'm using Fedora 11, with Cacti 0.8.7e, and using spine as the poller. SELinux is enabled, but configured properly, and not causing any issues.

After tweaking Cacti's configuration (e.g. changing the SNMP Timeout, increasing the Maximum OID's Per Get Request) I am sure the problem relates to the device.

When using the standard net-snmp command line tools to query the device, there are 2 distinct behaviors that are happening, the first is that the queries take a long time to come back. I'm talking seconds. The second is that with each group of values returned, the delay between the groups of OIDs grows. If you request a specific OID it seems to come back fairly quickly. SNMPwalks, SNMPbulkwalks and other queries that grab as much information as possible consistently time out. I have compared the results from the WRV210 to my other SNMP network devices: a D-link WAP, a GTA Robox Firewall, a Cisco 2925 switch and my Linux systems. They all work as expected, and Cacti generates pretty graphs. The graphs for the WRV-210 all seem to be low bandwidth chatter (arp, smb, etc).

Since my gaming systems (consoles and a computer) all route out through the WRV-210, I know that the graphs should be showing significant usage. I have pulled down a couple of Linux ISOs through the device, at close to full utilization, and the graphs did not register any activity.

I opened a case with Cisco's technical support back in April, but I have not heard anything since then.

Other than roundaboutrc and me, has anyone else got one of these devices, and if so, anyone got one working properly in Cacti?

David
[edit] - fixed cacti version number.