(BSM) Support Tip: RUM 9.40 Performance issues on Probes
Verify if traffic processed by each probe exceeds the limit which is for up to 700 Mbps (HTTP with a loaded configuration). The following are the most significant parameters that affect the number of RUM Sniffer Probe machines needed for a deployment:
- Amount of traffic in megabits/sec of monitored (not total) traffic
- Application type (HTTP, TCP, DB, and so forth)
- Traffic type (HTTP/HTTPS)
- Number of text searches
- Snapshot on Error configuration
- Operating system
Tip: You can significantly reduce the performance overhead of text pattern events by defining them on specific actions, rather than globally.
It is also a good idea to get the values for Megabits/Sec, Pages/Sec and Number of Sessions; check the Appendix A: How to Obtain Required Sizing Information from the RUM Sizing Guide.
Check the logs for any entries on WARN or ERROR status, especially the capture.log. Further logging can be enabled for this log on a specific logger on the ~\etc\rum_probe\log4cplus.properties file, e.g:
2018-02-20 13:23:16  ERROR common.ProbeSystemInfoReader <collector/common/ProbeSystemInfo.cpp:266> - unable to retrieve interface via - /usr/bin/read_net.sh 'eth8' 2> /dev/null
Add or change to: log4cplus.logger.common=DEBUG, STDOUT, R, Q
If you find that this or any post resolves your issue, please be sure to mark it as an accepted solution.