Resource Monitor Implementation
Policy Implementation Monitor Routers
Service A monitors two routers: Rtr_1 and Rtr_2. Service A functions if Rtr_1 or Rtr_2 is up, but it cannot function if both routers are down. Service A uses the Contact Status Low Sensitivity policy to monitor the service.
This policy rule set stipulates the following information:
- If all resources are down, service is down.
- If any 1 resource(s) are down, service is degraded.
Service with Policy
Service A seems to adequately monitor Rtr_1 and Rtr_2, but, on second glance, not enough to determine Service A actual viability. What about the case where Service A is not only dependent on the routers but also on particular router ports? The Contact Status Policy only monitors whether the routers are up or down.
The contact status of the routers could be established, simultaneously the ports on which Service A depends could be unavailable. Because of the limited scope of monitoring provided by its policy, Service A appears viable when actually it is not. Service A requires the more precise method of monitoring its resources that resource monitors provide.
Resource Monitor Implementation Monitor Routers and Their Ports
Scrutiny of the policy implementation reveals that it does not monitor the status of router ports that support Service A. The ports must be monitored with contact status of the routers. In addition, ports 4 and 5 on each router must be available for Service A to function optimally and that at least two of the four ports must be available for Service A to function adequately.
The following two resource monitors can be created as resources of Service A:
- Router Contact Resource Monitor -- Monitors the contact status of Rtr_1 and Rtr_2 (same as the policy implementation).
- Port Status Resource Monitor -- Monitors the port status of ports 4 and 5 on each router. Because at least two of four ports must be available. This resource monitor uses a Port Status Percentage Policy that reports down when 75 percent of the ports are unavailable and degraded when 50 percent of the ports are unavailable.
The result is that Router Contact Resource Monitor and Port Status Resource Monitor become resources of Service A. Rtr_1 and Rtr_2 become resources of the Router Contact Resource Monitor. Ports 4 and 5 on each router become resources of the Port Status Resource Monitor.
Service A monitors the service health attribute of Router Contact Resource Monitor and Port Status Resource Monitor with the Service Health High Sensitivity Policy. This policy rule set stipulates the following information:
- When any 1 resource(s) are Down then the service is Down.
- When any 1 resource(s) are Degraded then the service is Degraded.
- When any 1 resource(s) are Slightly Degraded then the service is Slightly Degraded.
So, if either one of the resource monitors is Down, Service A is down.
Resource Monitors Monitor Routers and Ports
Refined Resource Monitor Implementation Monitor Routers, Ports, and Response Time Tests
Further scrutiny of service A reveals that database server responsiveness and FTP transfer time are critical to its functionality. Assume that two RTM_Test models in
DX NetOps Spectrumdefine critical, major, and minor thresholds for response time to the database server and the FTP server. These tests can be resources of another resource monitor (the Response Time resource monitor), which becomes another resource of Service A.
Response Time Resource Monitor monitors the response time tests with the Response Time High Sensitivity Policy. This Policy rule set stipulates the following information:
- When any 1 resource(s) are down then the service is down.
- When any 1 resource(s) are degraded then the service is degraded.
Therefore, if this resource monitor reports down (either test indicates unacceptable response time latency) then Service A is down as stipulated by the Service Health High Sensitivity Policy which monitors all Service A Resource Monitors.
Resource Monitors Monitor Routers, Ports, and Response Time
When you use a Resource Monitor for multiple Services that are in active and Maintenance Mode states, the following attribute allows you manage the resource monitor behavior for services in maintenance mode.
Allows you to manage the Resource Monitor status for Services in Maintenance Mode.
When the RollMaintenanceToResourceMonitor attribute value is set to ‘YES’ (default value)
If you add a resource monitor from an active service to a service that is in maintenance mode, the resource monitor turns into maintenance mode in all the services where it is being used.
When the RollMaintenanceToResourceMonitor attribute value is set to ‘NO’
If you add a resource monitor from an active service to a service that is in maintenance mode, the resource monitor does not turn into maintenance mode in all the services. The resource monitor continue to be in its state in other active services.