SolarWinds NPM: Good, but not great
March 06, 2020
SolarWinds NPM: Good, but not great
Score 8 out of 10
Vetted Review
Verified User
Overall Satisfaction with SolarWinds Network Performance Monitor
Solarwinds NPM is being used in our department to monitor over 1000 network devices. They monitor all ports and routing statuses. The problems that it addresses is visibility into network device health and status. With other network add-ons such as Network Configuration Manager, configuration backup is also possible. What NPM has helped us with is identifying when devices become unresponsive or when certain routing adjacencies fail. This can help us identify issues in the network without relying on our customer to tell us they are having performance or reachability issues. We also rely on NPM to identify link errors in our environment. We were able to identify many links where CRC/input errors were occurring. Without NPM, it would have been a monumental task to identify that as a possible performance issue point.
- Identifying down nodes in network.
- Identifying neighbor adjacency changes.
- Graphing feature to give a quick look at interface bandwidth trends.
- "Network Top 10" to identify issues in the network.
- Support is almost non-existent. Almost always solved by Googling for a solution.
- Care and feeding sometimes requires direct database interaction. For example, when neighbor adjacencies are removed, they will show up on the alerts page. But deleting them from alerts won't work. Changing the auto delete timeout is not a workable solution as sometimes real links are down for extended periods of time. One needs to go into the database to delete the routing neighbors that are no longer valid to permanently remove them.
- Upgrades almost always have unforeseen issues. Whether a package isn't installed/ configured correctly to additional polling engines not able to be simultaneously upgraded, etc.
- No Linux version. Windows requires many patches that required reboots on a monthly basis.
- Proactive response to network issues. No longer relying on customers telling us there is a problem.
- Visibility into the network in general. No more having custom scripts to ping.
We have evaluated Cacti and still use an older version to monitor tunnel traffic. But support is worse and use-ability not as good as Solarwinds. Also, there is the question of scalability to 1000+ devices.