[
https://issues.apache.org/jira/browse/NIFI-1464?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15150537#comment-15150537
]
ASF GitHub Bot commented on NIFI-1464:
--------------------------------------
Github user olegz commented on a diff in the pull request:
https://github.com/apache/nifi/pull/210#discussion_r53166575
--- Diff:
nifi-commons/nifi-properties/src/main/java/org/apache/nifi/util/NiFiProperties.java
---
@@ -71,6 +71,7 @@
public static final String ADMINISTRATIVE_YIELD_DURATION =
"nifi.administrative.yield.duration";
public static final String PERSISTENT_STATE_DIRECTORY =
"nifi.persistent.state.directory";
public static final String BORED_YIELD_DURATION =
"nifi.bored.yield.duration";
+ public static final String PROCESSOR_START_TIMEOUT =
"nifi.processor.start.timeout";
--- End diff --
Very true and unfortunately this is where I would go back to the
"convention-over-configuration" discussion where if we declare that
MILLISECONDS is a convention (the most widely used, hence natural and
intuitive) for expression of all things time related then consistency would
come.
I am just afraid that we are actually asking user to do more (think more),
rather then less. On top of that it creates more possibilities of spelling
errors to deal with in the event user describes it in a way we can't parse,
which may come with international users who I presume may naturally type things
in their own language variation ('seconde', 'segundo' etc.)
Anyway, I've changed it, so we're good here, but would like this discussion
to continue at some point may be on the WIKI or dev list.
> Refactor lifecycle code for Processors and other components
> -----------------------------------------------------------
>
> Key: NIFI-1464
> URL: https://issues.apache.org/jira/browse/NIFI-1464
> Project: Apache NiFi
> Issue Type: Improvement
> Reporter: Oleg Zhurakousky
> Assignee: Oleg Zhurakousky
> Fix For: 0.6.0
>
>
> Similar lifecycle handling improvements that went in as part of the NIFI-1164
> for ControllerServices, could/should be applied to other components (e..g,
> Processors).
> The improvements may also help to address NIFI-78 (may be without killing the
> thread).
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
