Frequently asked questions about Postman Insights

The following are some frequently asked questions you might have about Postman Insights.

Why Insights if I already use New Relic and Sentry?

Just like Sentry makes it easier to debug certain errors even if you have New Relic, we're aiming to make it easier to debug API errors even if you have other tools installed.

Will the Agent slow down my application?

The short answer is no. The Insights Agent is not a proxy and doesn't sit in the path of traffic. Instead, the agent passively watches traffic using the Berkeley Packet Filter (BFP). For more information, see Footprint of the Postman Insights Agent.

How much memory does the agent use?

The Postman Insights Agent takes up about 500 MB of memory. The actual amount of memory depends on the volume of API calls the agent is watching. If you have concerns, please email the Postman Insights team to help you decide if you should increase your memory limits. Coming soon: memory limits on the agent.

Are there risks of running the Agent to uptime for my production service?

For Kubernetes, to the best of our knowledge we're completely isolated, so we shouldn't affect your service’s performance or uptime.

For Elastic Beanstalk, the agent shares the same memory pool as the rest of the application, so there's a chance of running out of memory.

• Current mitigation: increasing memory.
• Coming soon: memory limits on the agent.

What happens in Repro Mode to access tokens and other request values?

Insights scrubs a set of default fields. For more information, see Repro mode data redactions.

If there are additional values you'd like to scrub, you may specify additional fields, values, and regular expressions to redact in Settings.

What happens to the data when Repro Mode is off?

If Repro Mode is off, all payloads are obfuscated. For more information, see Repro mode data redactions.