[馃挕 FEATURE REQUEST]: Better handling of failed redials [JOBS drivers] #1660
Labels
C-feature-accepted
Category: Feature discussed and accepted
help-heeded-medium
Call for participation: Experience needed to fix: Medium / intermediate
Plugin
JOBS
I have an idea!
Sometimes, for example, if the user has an unstable connection between RR and any JOBS driver, or the connection is lost, RR may exceed the number of redial attempts. In this case, the pipeline becomes unusable, but still sits in the JOBS pipeline list.
The user can check the
State
RPC endpoint to restart RR and fix network problems, but if there are many logs, the broken pipeline warning may just be buried in the log messages and thus missed.This ticket is about exploring a way to better handle these cases. Preferably - automatically from the RR side. As an option, the JOBS driver could send a command to the JOBS plugin to restart a specific pipeline. Commands are supported, but not currently used.
The text was updated successfully, but these errors were encountered: