CDAP-21091: Increasing timeout for task workers to recover when app fabric restarts#15855
CDAP-21091: Increasing timeout for task workers to recover when app fabric restarts#15855anshumanks wants to merge 1 commit intodevelopfrom
Conversation
| private static final int NUMBER_OF_RETRIES = 20; | ||
| private static final int MIN_WAIT_TIME_MILLISECOND = 2000; | ||
| private static final int MAX_WAIT_TIME_MILLISECOND = 60000; |
There was a problem hiding this comment.
can we please move these to cconf?
There was a problem hiding this comment.
Let's move only the NUMBER_OF_RETRIES to cconf.
That's the value we would need to modify mainly to configure the retry behavior.
There was a problem hiding this comment.
I think we should keep min / max wait time the same and make NUMBER_OF_RETRIES configurable in cconf.
If later we figure out that the number of retries are insufficient, we won't have to make data plane changes for it.
itsankit-google
left a comment
itsankit-google
left a comment
There was a problem hiding this comment.
please add JIRA title in PR description
itsankit-google
left a comment
itsankit-google
left a comment
There was a problem hiding this comment.
PR description should be of the form: data-integrations/google-cloud#1473 (comment)
|
anshumanks
changed the title
4 participants
Increasing timeout for task workers to recover when app fabric restarts
Jira: CDAP-21091
Description
This change lets the task workers restart gracefully when app fabric pod is restarted.
Code change
ComputeEngineCredentials.java