I am investigating if sentry fits our needs.
Events get created in a remote data center.
Unfortunately the network connection to our central sentry server in the local data center is slow.
How does the sentry client handle this?
What happens if the network connection between the client and the server is down for about ten minutes?
Do events get buffered/queued on the client side?
It depends on the SDK, but generally things are transmitted asynchronously and won’t be affected by high latency of downtime.
Hi matt, thank you for your reply. Great if the client does submit things in async. If I use sentry, then I will use the python client. Where can I find more about the implementation if this async and buffered transmission?
Unsure if we document it well, but our python SDK will by default transmit using a thread. It does not buffer or retry on failures though.
I want to use sentry. But my constraint is, that a downtime of 30minutes must not result in message loss.
Is there a way to solve this?
In theory, this would be relatively easy to do this logic in your own in a custom Transport. But we don’t have support for this natively.