Right now it is impossible to take a heap snapshot of a running API server to debug memory leaks in production - since taking the snapshot (can take up to a few minutes) will make the node non-responsive and fail the health check, which causes Render to kill the server soon after the snapshotting is done.
Hey,
Disabling or reconfiguring health checks isn’t possible. If needed, you can temporarily upgrade the instance type associated with your service to increase memory.
Jérémy.
Render Support, UTC+3
That’s not a solution, heap snapshots will make servers unresponsive no matter what even if you have infinite memory
Hey,
It’s not possible to disable health checks on Render—there’s no option for that. Heap snapshots aren’t the only way to troubleshoot memory leaks; there are various tools available depending on the programming language you’re using that can help you profile memory usage. We can offer some help if you describe the issue you’re facing and let us know what you’ve already tried.
Jérémy.
Render Support, UTC+3
Whats your suggestion for finding the root cause of memory leak for NodeJS?
Hey,
The best solutions are already available online, so I’m not going to share anything groundbreaking here. Articles like https://betterstack.com/community/guides/scaling-nodejs/high-performance-nodejs/nodejs-memory-leaks/ should help, and there are several dependencies you can use.
Jérémy.
Render Support, UTC+3
So you pointed me to a resource that tells me to use Memory heapsnapshot? Anyway is there any way to surface this feedback back to the engineering team @ Render?
Hey,
There are plenty of tools that can help with investigating memory leaks in Node applications, but I won’t list them all here. The link I shared earlier provides a good overview of troubleshooting memory leaks in Node, but you’re free to use any tools you prefer.
Jérémy.
Render Support, UTC+3
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.