Is there a way to programmatically wipe clean the connection pool used by Panche and/or Hibernate at runtime? Users are having issues with stale connections when using Lambda SnapStart.
--Bill BurkeRed Hat
--
You received this message because you are subscribed to the Google Groups "Quarkus Development mailing list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to quarkus-dev...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/quarkus-dev/CAL%3DE%3DjTcY40DEzsgVvvAxdq9UwAuXnQt6jENZBOHZaUJA3nSdg%40mail.gmail.com.
--
To view this discussion on the web visit https://groups.google.com/d/msgid/quarkus-dev/CALeTM-k6fxAw%2BEmiMhW2miqybXw0kXVFsj6NimkwN%2B-5eF0Cdw%40mail.gmail.com.
Better still would be to have a Quarkus API for this so that we don't have a dependency on these APIs.
Yes Georgios, was going to suggest that the user implement an observer for StartEvent and register a Crac callback to flush() the connection pool. Looks like they can just inject the AgroalDataSource and do this.We should automatically do this in the Agroal extension, IMO. I'll do a separate PR for this.
Yes Georgios, was going to suggest that the user implement an observer for StartEvent and register a Crac callback to flush() the connection pool. Looks like they can just inject the AgroalDataSource and do this.We should automatically do this in the Agroal extension, IMO. I'll do a separate PR for this.