Execute your plan, test, and validate the deployment.
Implementing and testing your deployment will go quickly and effortlessly if you have spent adequate time in the Planning stage. The more detail you Plan has, the easier Implemenation is.
Your plan should be executed as documented, following the steps you identified and tested, and then running the appropriate tests at each stage to ensure the implementation meets your requirements. Any step which fails to meet your test criteria should result in an implementation stoppage until you can positively identify the cause. If the issue requires a deviation from the plan then you should rollback that step and return to the Planning stage.
TESTING
Polling Server Performance -
- Is the NMS server operating within acceptable performance metrics (CPU, Memory, polling, etc)
- Are all devices being collected on, as desired? i.e. are appropriate device models being auto-assigned to each device, are devices responding to ICMP and SNMP polling
- Are polling and update cycles being completed within acceptable windows (i.e. before the next collect cycle)
- Are log files rotating properly
- Is disk usage within the expected range(s)
Master Server Performance (requires opHA)
- Are device performance and events appearing on the Master server from each Poller?
- If you are using opEvents, and events are forwarded by the opEvents' API, is opEvents on the Master being populated
Next Up
Observe - Monitor the system for performance, operation, and results.