Configuring an HA Hub for Failover
This section provides a high-level overview of the tasks required to set up a high-availability solution with the HA probe. All probe configuration tasks can be performed in either Admin Console or Infrastructure Manager.
Follow these steps:
- Set up a secondary hub. See the CA Nimsoft Monitor Server Installation Guide for instructions.
- If you have remote hubs that connect to the primary hub through tunnels:
- Set up tunnels from those hubs to the HA hub.
- Set up get queues on the HA hub. Set up the same set of queues that are configured on the primary hub.
- Deploy probes from your local archive to the secondary hub.
- For essential functions, deploy:
- HA, to provide failover
- data_engine, to send messages to the database
- NAS and alarm_enrichment, to collect alarm messages
- emailgtw, for the ability to email alarms
- service_host and admin_console, to provide access to Admin Console
- Deploy additional probes as needed. Consider the probes that run on the primary hub and decide if their operation is required during failover.
For example, if RSP is performing remote monitoring from the primary hub, you may want to enable it on the HA hub so monitoring is not interrupted. You may also want to deploy discovery_server to ensure discovery data continues to flow to the database.
Note: Because the probes running on the HA hub have their own configuration files and monitoring profiles, you can be selective about whether your want to enable full monitoring upon HA failover. You may choose to enable a subset of the probes, monitor fewer systems, or use less frequent polling during failover.
- Configure the HA probe on the HA hub:
- Specify the probes and queues to enable upon failover. In the HA configuration GUI, add items to the Probes to Enable and Queues to Enable folders by using the drop-down lists and Actions commands.
- Check the order in which probes will be enabled, and modify the order if necessary. In the HA Raw Configure utility, select the probes_up folder, edit the key values as needed, then click Apply. Note the following:
- Data_engine must start before service_host to avoid logging invalid errors.
- Alarm_enrichment must start before nas.
- Admin_console is a web app run by service_host, so it does not appear in the list of probes to enable.
If desired, you can change the order in which queues are enabled (queues_up), or specify probes and queues to disable upon failover (probes_down and queues_down).
- Ensure NAS Auto-Operator on Failover is enabled. Click the Advanced folder, ensure the feature is checked, and click Save.
This feature helps you manage alarms by allowing you to define profiles to close/acknowledge certain alarms, automatically assign alarms, or send a message when a specific rule is met.
- Configure the NAS probe on the primary hub and on the HA hub:
- On the primary hub, enable replication, auto-operator, and nis-bridge.
- On the HA hub, enable replication. Disable auto-operator and nis-bridge.
Note: Do not enable nis-bridge on the HA hub, as this will result in database constraint violation errors in the NAS log. These errors do not cause database issues, but result in error messages until the primary NAS successfully imports data into the NIS database.
- Install Infrastructure Manager on the HA hub if you normally access the application on the primary hub.
Note: This is not necessary if you have Infrastructure Manager on a computer other than the primary hub and manage your components from there.
Failover configuration is complete.
Copyright © 2014 CA Technologies.
All rights reserved.