Folks, we don't currently have a way to set a conn...
# general
r
Folks, we don't currently have a way to set a connection timeout on the Medplum client, right? The use case is doing some heavy batch upserts on the FHIR server using Lambdas and need to timeout before the lambda times out, so we can process it gracefully.
r
At present for the bots you are correct. Out of curiosity - what are you syncing?
r
We're getting the result of CCDA>FHIR conversion and batch inserting on a FHIR server. The code that does that runs on a Lambda, so we're looking to a bit more fine control over how we do things. 😄 Another thing it would be useful would be to optionally not to parse the response from the FHIR server into a JSON. In our case we don't use that response, and the conversion consumes additional cpu/memory that could otherwise be saved.
Another thing we need is to set a custom HTTP header while calling
executeBatch()
on the Medplum client.
I created a PR for the first and last item, both solved with passing
options
to `executeBatch`: https://github.com/medplum/medplum/pull/2137 I understand we didn't discuss this first, so no worries if you prefer a diff approach/route and we can close the PR.
r
Can you DM me your bot id? I can up your timeout
r
Hey Reshma! In this use case we're using the FHIR client on a lambda on our own infra, not a bot. Thanks, though! 🙂
r
ok! Great!
btw your PR was merged, and when we cut a new release the SDK will have it.
r
sounds good, thank you!
147 Views