I am getting incorrect values for the Cisco IP SLA HTTP sensor, specifically the TCP RTT.
PRTG is reporting the DNS RTT as identical to the TCP Connection RTT value on all my devices. Is it possible there is an error in your OID for that sensor (IOS 12.4-10 through 15.1)?
Article Comments
I am monitoring a Cisco router, IOS 15.1M. (Sorry for the viewing format) Here is command-line snippet for the IP SLA 2 (HTTP probe):
IPSLA operation id: 1 Latest RTT: 78 milliseconds Latest operation start time: 12:28:49 EDT Wed Mar 16 2016 Latest operation return code: OK Latest DNS RTT: 1 ms Latest TCP Connection RTT: 39 ms Latest HTTP Transaction RTT: 38 ms Number of successes: 34 Number of failures: 0 Operation time to live: Forever IPSLA operation id: 1 Latest RTT: 92 milliseconds Latest operation start time: 12:32:49 EDT Wed Mar 16 2016 Latest operation return code: OK Latest DNS RTT: 1 ms Latest TCP Connection RTT: 42 ms Latest HTTP Transaction RTT: 49 ms Number of successes: 38 Number of failures: 0 Operation time to live: Forever Now, let's look at the PRTG Live-Data View: Date Time Avg. Round Trip Time (RTT) DNS RTT TCP RTT Transaction RTT Message Body Octets Downtime Coverage 1 to 50 of 121 Items: 50 3/16/2016 12:29:50 PM 78 msec 1 msec 1 msec 39 msec 1,785 # 0 % 100 % 3/16/2016 12:29:50 PM 78 msec 1 msec 1 msec 39 msec 1,785 # 0 % 100 % 3/16/2016 12:29:25 PM 78 msec 1 msec 1 msec 38 msec 1,785 # 0 % 100 % 3/16/2016 12:28:25 PM 94 msec 1 msec 1 msec 46 msec 1,785 # 0 % 100 %
Notice how reported DNS RTT and TCP RTT are identical (1ms). The CML output shows the TCP Connect RTT to be 42ms for the timeslot above.
Mar, 2016 - Permalink
Ok, so is there anyway to do a custom query that presents itself in the same way as the IP SLA module? It appears that your query is only setup to poll 4/5 OIDs (TCP Connect RTT is missing).
. iso (1) . org (3) . dod (6) . internet (1) . private (4) . enterprises (1) . cisco (9) . ciscoMgmt (9) . ciscoRttMonMIB (42) . ciscoRttMonObjects (1) | - -- rttMonLatestOper (5) | - -- rttMonLatestHTTPOperTable (1) | | | - -- rttMonLatestHTTPOperEntry (1) | | | | -- rttMonLatestHTTPOperRTT (1) | | | | -- rttMonLatestHTTPOperDNSRTT (2) | | | | -- rttMonLatestHTTPOperTCPConnectRTT (3) | | | | -- rttMonLatestHTTPOperTransactionRTT (4) | | | | -- rttMonLatestHTTPOperMessageBodyOctets (5) | | | | -- rttMonLatestHTTPOperSense (6) | | | | -- rttMonLatestHTTPErrorSenseDescription (7)
Mar, 2016 - Permalink
Hello fti_tywu, thank you for your reply.
As you've shared the MIB tree data, the require information is stored in an SNMP Table rttMonLatestOper. This table is part of the CISCO-RTTMON-MIB.
You should be able to deploy an SNMP Custom Table sensor for this table (The OID of the table is 1.3.6.1.4.1.9.9.42.1.5.1), and there's a guide for setting up SNMP Custom Table Sensors here.
The sensor should be able to query all values stored within that Table, that is:
- rttMonLatestHTTPOperRTT (1)
- rttMonLatestHTTPOperDNSRTT (2)
- rttMonLatestHTTPOperTCPConnectRTT (3)
- rttMonLatestHTTPOperTransactionRTT (4)
- rttMonLatestHTTPOperMessageBodyOctets (5)
- rttMonLatestHTTPOperSense (6)
- rttMonLatestHTTPErrorSenseDescription (7)
Additional details about the MIB can be found using the Cisco SNMP Object Navigator.
Best Regards,
Mar, 2016 - Permalink
Could you share some more details? Especially the IP SLA configuration? Just so we could possible recreate the issue within our labs?
Mar, 2016 - Permalink