Configuring parameters for health checks

Edit online

If required, you can configure preferred conditions under which health check events are generated, by specifying identical OQL inserts to the Virtual Domain process schema file (VirtualDomainSchema.cfg) on both the primary and backup servers.

About this task

The Virtual Domain component uses two databases (config and state) to support Network Manager failover. The health check status records and filters are stored in these tables, which can be updated using the VirtualDomainSchema.cfg file.

To change the default settings for the health check parameters:

Procedure

On the primary Network Manager server, edit the $NCHOME/etc/precision/VirtualDomainSchema.cfg file by specifying the following OQL inserts:
- Update the column values in the config.defaults table to specify different time periods for the failover health checks.
  For example, you can use the m_HealthCheckPeriod column to change the time interval between each health check. Or you can use the m_FailoverTime column to change the interval after which failover is triggered by the backup domain, when the primary domain is deemed to be in poor health. The default settings are as follows:
```
insert into config.defaults
(
    m_SocketKeepAlivePeriod,
    m_HealthCheckPeriod,
    m_FailoverTime,
    m_DirectoryScanPeriod,
    m_AutoTopologyDownload
)
values
(
    60,
    60,
    300,
    100,
    1
);
```
- If required, update the state.filters table to define individual filters for each poller configured in the $NCHOME/etc/precision/CtrlServices.cfg file. For example, for an additionally configured poller, PingPoller:
```
insert into state.filters
(
    m_ServiceName,
    m_Filter,
    m_Description
)
values
(
    "PingPoller",
    "m_ChangeTime > eval(time,'$TIME - 300') and m_CtrlState <> 7",
    "The Poller has been running within the last 300 seconds"
);
```
Save and close the file.
Make identical changes to the $NCHOME/etc/precision/VirtualDomainSchema.cfg file on the backup server.