The following table provides information on various types of polling groups.
|Polling Group Name||Polling Period||Agents Group||Snmp masks||Notes|
|G_CISCO_IPSLA_INV ENTORY||900 seconds. A 15 minute polling period allows quick probe discovery when they are set up.||IPSLA||CISCO-IPSLAINVENTORY||This polling group is used to poll inventory data on the probes.|
|PG_CISCO_IPSLA_HIS TORY||300 seconds.||IPSLA_HISTORY||CISCO-IPSLAHISTORY||This polling group is used to poll the IPSLA History table (rttMonHistoryColelctionTable ). This table contains dat about the latest “n” operations for a probe. “n” being the number of statistics buckets configured for the probe. This group must be enabled manually to be used.|
|PG_CISCO_IPSLA||300 seconds.||IPSLA||CISCO-IPSLASTATSTABLE||Poll all probes, except Jitter, HTTP and IcmpPathEcho, using the statistics table.|
|PG_CISCO_IPSLA_JITTER||300 seconds.||IPSLA||CISCO-IPSLAJITTERSTATSTABLE||Polls the Jitter probes, using the statistics table.|
|PG_CISCO_IPSLA_HTTP||300 seconds.||IPSLA||CISCO-IPSLAHTTPSTATSTABLE||Polls the HTTP probes, using the statistics table.|
60 seconds. The polling period should be set to the shortest probe frequency in order to gather all data. Data point will be missing if polling occurs more slowly.
|PG_CISCO_IPSLA_LAT EST_OPERATION||CISCO-IPSLAHTTP- LATEST, CISCO-IPSLAJITTER- LATEST||Use this polling group to poll the latest operation tables for HTTP and Jitter probes. This group is OPTIONAL as it generates a lot of network traffic and cpu load on the device and collector.|
NOTE about the POLLING PERIOD on the polling groups using the STATISTICS TABLES:
The polling period must be set to a value equal or higher than the slowest probe being queried. Otherwise, spurious values (with a zero value) may appear in the reports. This applies to the polling groups using the CISCO-IPSLA-XXX-STATSTABLE.
NOTE about the POLLING PERIOD on the polling groups using the LATEST-OPERATION tables:
The polling period must be set to a value smaller than the probe with the smallest frequency. This shall be done to ensure that no data points are lost.
NOTE about the POLLING PERIOD on the polling groups using the HISTORY tables:
The period may be set to 300 seconds when multiple history lives and buckets are used. For a typical application (a IcmpEcho probe, setup with a 60 seconds frequency, 15 buckets: 15 minutes retention time), a 300 seconds (10 minute) period could be used. However, it would also be possible to use 600 seconds (10 minutes) as it still falls within the probe's data retention time.
The period must be set to a value smaller than the most frequently occurring probe if the probe buckets and lives are set to 1.