Nagios Network Analyzer CPU 100% Gateway Timeout when extracting Reports

[smallboy679]

hi Nagios Community

we are currently testing out Nagios Network Analyzer on our Gateways..

The Network Analyzer Application does not give pdf reports and Apache is showing under htop 100% utilization..

my Core Count is 18 and memory 32g..

my Nic is 10Gbps..

can please direct me to documentation that will optimize the reporting as well as Application..

as we are trialing out this application..

Thanks

[sgardil]

hi Nagios Community

we are currently testing out Nagios Network Analyzer on our Gateways..

The Network Analyzer Application does not give pdf reports and Apache is showing under htop 100% utilization..

my Core Count is 18 and memory 32g..

my Nic is 10Gbps..

can please direct me to documentation that will optimize the reporting as well as Application..

as we are trialing out this application..

Thanks

Hey @smallboy679

Glad to hear your giving network analyzer a shot. I wasn't able to find any documentation or tips for optimizing your NNA server. For the PDF reports is there certain reports or pages you are looking for PDF extracts of? I see an export as PDF option on most pages that have information but im unsure if we are missing it in a few locations. Here is some documentation on our reports and if you have any suggestions on things that NNA is lacking feel free to let us know!
https://assets.nagios.com/downloads/nag ... eports.pdf

[smallboy679]

@sgardil

Thank you for the link,

just one question

it just came to me, are the interface for the netflow ingress and the Nagios Network Analyzer the same..

currently both are using the 10gbps NIC, maybe that is why the Gateway is timing out..

is it best pratice to keep the Nagios Network Analyzer interface separate from the Ingress Netflow interfaces...

Appreciate your advise

Thanks

[sgardil]

Hey @smallboy679

I will preface my knowledge in this area is limited. From what i've researched it seems that if your having issues with 100% utilization that it may be best to have them be separate to avoid possible resource contention between nna commands and the netflow data coming in. That being said I didnt find any posts about doing so with NNA, there has been some things with XI though, and I don't know what the process would look like. If I find anymore data on it I'll make sure to let you know.

[jack_002]

The issue, where Nagios Network Analyzer hits 100% CPU usage and triggers gateway timeouts during report generation, is typically caused by the heavy processing load involved in compiling large or complex reports. PDF generation in particular is resource-intensive, especially if a wide time range or multiple data sources are selected.

In order to improve the performance even better, you will have to adjust your report configurations. This adjustment will help with covering the shorter time ranges or limit the number of included data sources. In addition to this, tuning your Apache and PHP configurations can help.you can also try increasing the parameters such as memory_limit, max_execution_time, andMaxRequestWorkers. This will definitely allow your server to handle larger operations more efficiently.

It’s also important to verify the number of background worker threads Nagios Network Analyzer uses. With an 18-core CPU, increasing worker threads can help distribute the workload more effectively. Reviewing your flow data retention settings is another good step, retaining excessive data can lead to high disk I/O and memory consumption during reporting.

Lastly, Nagios provides official documentation on system requirements and performance tuning. Referring to these resources can guide you through recommended optimizations tailored to your environment.

Nagios Network Analyzer System Requirements & Performance Tuning
Nagios Core Performance Tuning Guide


You should monitor the system behavior after applying these adjustments. This will help you identify whether further scaling or architectural changes are needed to support your reporting requirements smoothly.

For reference, this behavior is conceptually similar to what happens in Java applications under memory and processing strain. If you’re curious how performance bottlenecks can lead to application freezes or spikes in other environments, here’s a helpful read: Solving Intermittent Application Freezes with a Java Memory Analyzer