Troubleshooting NetCrunch database issues - tips and tricks
Part of NetCrunch housekeeping advice - learn about the best practices and solutions to database-related issues that may arise if you decide to store data in the long term.
Best practices to optimize the database resource usage
Please refer to this article to learn more about NetCrunch databases. It is recommended to follow these tips after creating your NetCrunch atlas. Follow them later on when configuring or tuning your NetCrunch.
- Create and manage alerts through monitoring packs instead of creating events for every node separately.
- In the case of Windows Event Log monitoring, do not set NetCrunch to monitor all Windows Event Log events to prevent database overload. You should collect data about specific events that you are interested in.
- If needed, set Extended monitoring only on critical nodes.
- Set the device type on every node in the atlas.
- Avoid running antivirus software on the same server: Please read more about it in the documentation or the knowledge base.
- Exclude from monitoring or limit monitoring of less important nodes - you may increase monitoring time or set these nodes to simplified monitoring mode.
- Configure alerting scripts first to assign them later to desired events.
- Add email/text message notifications only to important events to prevent alert floods and missing important information.
- Log off from the NetCrunch server machine when you are not actively using the NetCrunch console. Avoid running the NetCrunch console on NetCrunch Server, also via RDP, if possible. It eats up NetCrunch Server resources.
Troubleshooting 'database size is close to its limit' messages
Insufficient database resources might cause performance issues, so it's recommended to lower the size of the event database.
Removing some older events
To decrease the database size you should remove some older events. You can also enter a lower "Keep Events for" value. To do it, in the top menu go to NetCrunchNetCrunch SystemMaintenance. In the window, select NetCrunch Server tabEvent Database and by clicking a 3 dot menu run Verify database integrity... and Optimize options. Also, restart the NetCrunch Server service when done.
Removing all events
Another option would be to wipe the events from the database completely. There are two ways to do this:
- Safer but longer: export full backup and import it again as a new atlas with Events DB and Inventory unselected under Options. After importing and checking if everything is OK, delete the old atlas from Atlas Manager.
- Less safe but much faster: stop NetCrunch Server and NC Advanced SQL Server services, go to the atlas data folder (default: c:\ProgramData\AdRem\NetCrunch\data<atlas id>) and delete/move ElevateDB folder and ncevents.bpk file. Next, start NC Advanced SQL Server and NC Server (in this order!).
Verifying flow monitoring configuration
Flow monitoring is notorious for overflowing database if not configured properly. One of the symptoms may be the fact that the flow dashboard is not showing data:
- Please stop the NC Flow Collector service, go to the NC data folder, and remove the NetFlow database (by default it is in C:\ProgramData\AdRem\NetCrunch\data\NetFlowMonitor\2, where "2" is your atlas id).
- Start NC Flow Collector service and check if data can be seen on the Flow dashboard.
- Check what the current flow sampling rate is set on your flow devices and decrease it accordingly.
NetCrunch has been designed for the long-term storage of your network performance data. Please manage NetCrunch settings and plan to provide sufficient resources to NetCrunch Server so that you can keep the data for as long as you need.
