1 Respuesta
- Más nuevo
- Más votos
- Más comentarios
0
I found the cause - It was a resource/thread starvation issue.
The lambda's runtime was being re-used for multiple requests.
The lambda created a new TimeStreamWriteClient for each request.
TimeStreamWriteClient implements Closable (or AutoClosable) and it needs to be closed as having multiple open instances around uses resources.
I wasn't closing the client. Closing the client after each request - in this case a Kotlin use statement (roughly equivalent to a Java try-with-resources) - appears to have solved this issue.
respondido hace 2 años
Contenido relevante
- OFICIAL DE AWSActualizada hace 3 años
I'm not sure if accepting my own answer violates the conventions, but AWS re:Post sent me an email telling me to accept the answer if I though it was right, so I did.