Hi,
Has anyone had any strange memory readins via SNMP using NPM and a Cisco VSS cluster?
I am seeing incorrect readings, for example i have 1gb memory and 370mb is used. But NPM says 104mb in use, 2mb free?
Does anyone have fix?
Thanks!
We just brought up a VSS pair at one of our campuses last night and noticed very similar behavior.
SolarWinds is showing the core at 96% utilization yet we're running around 365Mb of 1Gb. What's interesting, like you said, is that NPM is showing the max memory is 100Mb, not 1Gb. I'm wondering if it's not using the right MIB or is only looking at a certain process for its reporting.
Have you had any luck sorting the anomaly out?
We opened a ticket with Solarwinds. Here is what they said. I havent had a chance to verify yet.
When we poll cisco devices we use the same RFC standard mibs. I’ve included them below. We’ll want to test these OIDs using either the universal device poller or the toolset if you have it. This will show us what the device is reporting for those snmp requests. OID = CiscoMemoryPoolFree 1.3.6.1.4.1.9.9.48.1.1.1.6 OID = ciscoMemoryPoolUsed 1.3.6.1.4.1.9.9.48.1.1.1.5
We can only go by what the device tells us when we poll it via SNMP. NPM is showing the same thing as a basic SNMP request for memory. You are going to have to talk to Cisco and find out why the device responds incorrectly to the SNMP request."
How have you made out with this? I am seeing the same symptoms on my 6500 VSS and am wondering if/how you were able to resolve it.
We are observing a similar issue across various Cisco devices. Support advised to submit a feature request or create custom poller for deprecated OID to get correct reading. Support is ignoring straight forward questions.......
Update for Case #347847 - "Memory reporting on Cisco 1811 seems incorrect" Hi Stan, Okay, according to the email, it looks like the value for the displayed memory is from a deprecated OID that we are not using. If you want to read that value directly, you will have to create a Universal Device Poller to poll that specific OID. Then you would be able to run a report on that for planning your upgrades. It looks like the OID you want to use is 1.3.6.1.4.1.9.3.6.6. Please create an UnDP to poll that against a sample node and make sure the value returned is what you are expecting. Please see the Administrator's Guide for details on how to create a Universal Poller: www.solarwinds.com/NetPerfMon/SolarWinds/wwhelp/wwhimpl/js/html/wwhelp.htm#href=OrionAGMonitoringCustomMIBsCreate.htm<http://www.solarwinds.com/NetPerfMon/SolarWinds/wwhelp/wwhimpl/js/html/wwhelp.htm#href=OrionAGMonitoringCustomMIBsCreate.htm> Thanks,
