Troubleshooting polling timeouts

If polling times out because some devices cannot respond in time, adjust how those devices are polled.

To adjust polling for slow devices, complete the following steps.

Identify all of the devices that are generating SNMP timeout errors.
Create one or more default pollers. If there are a large number of devices that are not responding in time, create more than one poller. For information on how to create a poller, see Setting up an additional poller.
Create a new poll policy and assign it to each of the newly created pollers. For information on how to create a poll policy, see Creating and editing poll policies.
Edit the poll policy and scope it to only the slow devices. Select the devices based on class, or select individual devices from the network view in the Device Filter tab in the Poll Policy Editor.
Increase the poll interval in the General tab of the Poll Policy Editor. Save and start the poll policy.
Optional: Verify that the poll policy is applied successfully to the poller by running the following command:
```
ncp_oql -domain domain_name -service SnmpPoller -poller poller_name -query "select * from profiling.policy;"
```
If the Target Count is greater than zero, then the poll policy is running on that poller.
Optional: Verify the list of network entities that are currently monitored by the poller by running the following command:
```
ncp_oql -domain domain_name -service SnmpPoller -poller poller_name -query "select * from monitors.current;"
```
If the devices still do not respond in time to SNMP polls, configure the aggregation limit:
1. Stop the Network Manager core processes.
2. Back up and edit the etc/precision/NcPollerSchema.cfg file.
3. Find the insert into the config.properties database and set the value for the UseGetBulk property to 1.
4. Add the following line at the end of the file:
```
Update config.properties set aggregationLimit = 10
```
  Lowering the aggregation limit from the default of 30 configures the poller to retrieve fewer varbinds at a time when interface filtering is enabled.
5. Save and close the file.
6. Start the Network Manager core processes.
7. Edit the poll definitions that aply to these devices and enable interface filtering.
If the devices still do not respond in time, increase the SNMP timeout by running the following command and increasing the value from the default of 3000:
```
ncp_oql -domain domain_name -service SnmpPoller -query "update snmpStack.verSecurityTable set m_TimeOut=< >3000 >;"
```