Hi all,
While deploying flows to MiNiFi I noticed that a Remote Process Group whose
target URL is unreachable produces a WARNING bulletin and log entry
("Unable to connect to … due to …") every 60 seconds, even when
transmission is disabled.
Looking at StandardRemoteProcessGroup, initialize() schedules
InitializationTask every 60 seconds whatever the transmission state, and
every failure is reported through the event reporter as a WARNING bulletin.
In practice we have flows where some RPGs are intentionally left stopped
(e.g. a common flow deployed to several MiNiFi instances, where not every
instance has a reachable target). In that case the repeated bulletins are
noise and can hide real problems, and on MiNiFi there is no UI to show that
the RPG is simply not in use.
Would there be interest in an improvement along these lines?
- when the RPG is not transmitting, log the failure at DEBUG/INFO and do
not create a bulletin;
- when it is transmitting, keep the current WARNING and bulletin, possibly
only on state changes (reachable → unreachable and back) instead of every
60 seconds.
The periodic check itself would stay as it is, since the UI uses it to show
remote ports and authorization issues before transmission is enabled.
I would be happy to open a Jira and submit a PR if this sounds reasonable.
Thanks,
Vincenzo