[
https://issues.apache.org/jira/browse/NIFI-13992?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17896729#comment-17896729
]
David Handermann commented on NIFI-13992:
-----------------------------------------
To provide some background context, as commented in the Apache NiFI Slack:
For containerized deployments, the bootstrap process is really unnecessary, so
the {{nifi.sh run}} command starts NiFi and does not continue running the
bootstrap monitor.
When launching with {{nifi.sh start}} the bootstrap process continues running.
The status logging raises the question of "who watches the watchers?" as it
relates to the health of the bootstrap process itself. Defaulting the status
message to the info level, and logging every five seconds gives a positive
indication of the bootstrap status, and a historical track if the main process
dies. However, I appreciate that this could be considered excessive in some
scenarios. For such situations, simply changing the Logback configuration for
that logger to WARN is an option.
The current default is a bit a of middle-ground approach, although at a high
level, using nifi.sh run with some other robust monitoring strategy is the
better option.
> Excessive bootstrap logging
> ---------------------------
>
> Key: NIFI-13992
> URL: https://issues.apache.org/jira/browse/NIFI-13992
> Project: Apache NiFi
> Issue Type: Improvement
> Reporter: Mark Bean
> Assignee: Mark Bean
> Priority: Major
>
> The bootstrap application logger is writing a status message to
> `nifi-bootstrap.log` every 5 seconds. This is an excessive amount of logging.
> Recommend reducing the frequency and/or changing the logging to DEBUG level.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
