[
https://issues.apache.org/jira/browse/FALCON-240?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13859664#comment-13859664
]
Venkatesh Seetharam commented on FALCON-240:
--------------------------------------------
AFAIK, this was by design since retention is not for a specific instance.
Eviction is done in bunk on a periodic basis with out regard to feed
periodicity. There is a correlation between the schedule of the eviction and
the feed's periodicity but thats it. Each invocation can evict/purge multiple
instances/partitions of a given feed and there is no way to tie the instance to
a particular run even though they are logged.
IMHO, We should not solve this since there could be a run every 6 hours and not
know which instances were evicted. I'd rather look at a pressing issue in
FALCON-166.
> Instance status from CLI on a feed doesn't give the retention details
> ---------------------------------------------------------------------
>
> Key: FALCON-240
> URL: https://issues.apache.org/jira/browse/FALCON-240
> Project: Falcon
> Issue Type: Bug
> Reporter: Suhas Vasu
> Assignee: Suhas Vasu
> Attachments: FALCON-240.patch
>
>
> Doing a instance status for a feed doesn't show details for retention.
> Ex:
> Running the following command -
> falcon instance -type feed -name X -status -start 2013-12-28T00:00Z -end
> 2013-12-30T12:00Z
> gives -
> Consolidated Status: SUCCEEDED
> Instances:
> Instance Cluster SourceCluster Status
> Start End Details Log
> -----------------------------------------------------------------------------------------------
> Additional Information:
> Response: STATUSRequest Id: f2f48373-44d7-4b9b-afc8-b2c6e8c68827
> It would be helpful to know the retention details of a feed.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
