Our network engineers recently implemented some ICMP IPSLA tests on some Cisco XR routers and I'm having issues with the views being generated for them. The graphs being drawn for the various metrics are mostly only populating every other poll cycle.
Other metrics for these devices (cpu, memory, interface stats, etc.) are just fine. It's just the IPSLA test metrics that are not looking right.
A sample report shows the problem clearly:
These requests are seeking out the current bucket ID/Index value to use when gathering test results. They happen with every poll cycle which shows a failure from the device per these GEN ERROR messages.
Aug 19 12:55:28.234: CiscoIPSLADynamicDiscovery processResponse called: baseOIDWithTestIndex=188.8.131.52.184.108.40.206.220.127.116.11.1.10.2, responseEvent=[ResponseEvent [,
request=GETNEXT[requestID=1729999254, errorStatus=Success(0), errorIndex=50, VBS[,
18.104.22.168.22.214.171.124.126.96.36.199.188.8.131.5221007184.108.40.206 = Null,
220.127.116.11.18.104.22.168 = Null,
response=RESPONSE[requestID=1729999254, errorStatus=General variable binding error(5), errorIndex=1, VBS[,
22.214.171.124.126.96.36.199.188.8.131.52.184.108.40.20621007220.127.116.11 = 36,
18.104.22.168.22.214.171.124.0 = 77 days, 19:49:53.66,
userObject=ItemBasedRequestState[responseReceivedTimestamp=1629395728227(Thu Aug 19 12:55:28 CDT 2021), itemList=],
This is something the device admins will need to provide a fix for. If they're not sure of the cause and fix, might need to get help from the Vendor.
All supported DX NetOps Performance Management releases
Engage the device administrators for assistance. If they are unsure of the cause and solution, engage the device vendor for assistance.
Detailed Poll Logging was used to determine the root cause. Directions for its use can be found in Knowledge Base article Enable and download logging from DCDebug in Performance Management