Client blocking and request timeout

[xarg]

Seems your requests client doesn't have any timeout as well as it's running in a "blocking" operation which means that there's a chance to block some operation for a very long time your backends are slow for some reason.

Any chance to add a "timeout" param to the SDK. Optionally this identification runs in a thread similar to the Segment Python SDK?

The point is here is that I prefer to fail sending a notification from the app rather than cause a downtime due to your service slowdown. The "sync" mode still makes sense in a queue/consumer scenario where you want to have a synchronous failing situation to retry the sending.

[cjbell]

hey @xarg – great question here. we can absolutely look at adding some requests overrides so that you can set timeouts. I'm less convinced about building the non-blocking queue like segment right now as the majority of our customers do this
themselves. happy to take a pass at this though for the defaults!