You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Using rover dev and getting regular error messages like error[E004]: error sending request for url (http://localhost:4001/): connection error: Connection reset by peer (os error 54)
and error[E004]: error sending request for url (http://localhost:4002/): connection closed before message completed
This causes the router instance to stop and not restart until the next introspection query.
Steps to reproduce
Run rover dev --supergraph-config supergraph.yaml --polling-interval=5
where supergraph.yaml looks like
Rover dev should handle the errors (or at least retry) before removing the subgraph from the session. Investigationg have shown that the subgraph is sending a connection reset message to rover but this causes rover to fail the subgraph straight away even though the next request would succeed.
Environment
Version: 0.22.0
Install Location: /Users//.rover/bin/rover
OS: Mac OS 14.2.1 [64-bit]
Shell: /bin/zsh
The text was updated successfully, but these errors were encountered:
Hi @dandersonhall, apologies for the length of time taken to reply. We have a PR open that I believe would fix this issue, is there any chance you could have a go building Rover and trying it out in your specific situation? No guarantees obviously but it would be good to know if it helps :)
Description
Using
rover dev
and getting regular error messages likeerror[E004]: error sending request for url (http://localhost:4001/): connection error: Connection reset by peer (os error 54)
and
error[E004]: error sending request for url (http://localhost:4002/): connection closed before message completed
This causes the router instance to stop and not restart until the next introspection query.
Steps to reproduce
Run
rover dev --supergraph-config supergraph.yaml --polling-interval=5
where supergraph.yaml looks like
Expected result
Rover dev should handle the errors (or at least retry) before removing the subgraph from the session. Investigationg have shown that the subgraph is sending a connection reset message to rover but this causes rover to fail the subgraph straight away even though the next request would succeed.
Environment
Version: 0.22.0
Install Location: /Users//.rover/bin/rover
OS: Mac OS 14.2.1 [64-bit]
Shell: /bin/zsh
The text was updated successfully, but these errors were encountered: