4 answers
- 10-1
Hello Rodney
Can you run that command in your pick hour? This is due we need to see the real amount of data to deal with,
After that we can give some recommendations in HW,
BR
Add your comment... - 10-1
/usr/local/opmantek/bin/opflow_stats.pl
........................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................
Average Flows Per Second: 3625
Average Flows per Cycle: 102371
Flows Inserted per Cycle: 0
Conversations per Cycle: 33462
Conversations Compression: N/A%
Average LoadFlow: 29.4491554054054
Average FlowReader: 2.57547297297298
Average MvKill: 0.00988738738738728Add your comment... - 10-1
Hello Rodney,
It sounds like so much flow data is arriving at the server that it is having difficulty completing all the calculations needed to draw the webpage before the server times out (the 504 error you're seeing).
As I'm sure you know, server sizing and performance for NetFlow analysis is a complex subject due to the number of variables involved. I suggest we start by checking your Flow Volume following these instructions - Tracking Flow FlowVolume - Received by opFlow
Please post the output from the opflow_stats.pl command here so we can provide some further direction.
Also, it sounds like you may have built this VM from scratch? If so, please make sure that you've completed everything in the opFLow Setup: opFlowInstallationGuide - opFlowSetup
Regards,
Mark H
Add your comment... - 10-1
Hello Rodney.
We recommend to you run opFlow alone in one VM.
VM1 NMIS8, opCharts, OAE
VM2 opFlow
How many flows are received in opFlow?
How many nodes are you monitoring in NMIS?
Regards
Add your comment...
I am currently running NMIS8, opFlow, opCharts and Open-Audit on a VMware ESX/ESXi 4.0 (VM version 7) host with 4 CPU's, 8G of memory and and two 80G hard drives running CentOS 6.6. I've seen my memory get sucked up and proc running at conistently high levels. OpFlow is terribly slow and I often get 'Unknown error: 504 Gateway Time-out' errors. Are my hardware specs and software specs appropriate and are there any kinds of tweaking I can do to relieve my issues?