Hey, thanks for this library. I was able to reduce our DO code by like 90%
We are using it in production and now that we are getting some significant request volume there are some transient DO network errors popping up.
For example, we have several Cannot resolve Durable Object due to transient issue on remote node. and Network connection lost. in Sentry. These are on routes that do nothing but connect to DO and return some data so even though there is no stack trace I am pretty confident both of these issues are DO related (well obviously the first one is, but no clue where else that second one would be coming from).
It was suggested in the CF Discord to do retrying but that will be quite messy to add in the user code.
Would it be possible for this library to automatically do retrying with backoff on these types of errors?
Hey, thanks for this library. I was able to reduce our DO code by like 90%
We are using it in production and now that we are getting some significant request volume there are some transient DO network errors popping up.
For example, we have several
Cannot resolve Durable Object due to transient issue on remote node.
andNetwork connection lost.
in Sentry. These are on routes that do nothing but connect to DO and return some data so even though there is no stack trace I am pretty confident both of these issues are DO related (well obviously the first one is, but no clue where else that second one would be coming from).It was suggested in the CF Discord to do retrying but that will be quite messy to add in the user code.
Would it be possible for this library to automatically do retrying with backoff on these types of errors?
Thanks a lot!