[
https://issues.apache.org/jira/browse/ASTERIXDB-1535?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15437624#comment-15437624
]
Ian Maxon commented on ASTERIXDB-1535:
--------------------------------------
Likely this is related to ASTERIXDB-1534. If a dataset were to be evicted,
weird stuff can start to happen akin to what happens in that bug on a restart,
due to the logic for locating the root page being run.
> CC stop answering query from 19002 RESTAPI port
> -----------------------------------------------
>
> Key: ASTERIXDB-1535
> URL: https://issues.apache.org/jira/browse/ASTERIXDB-1535
> Project: Apache AsterixDB
> Issue Type: Bug
> Components: HTTP API
> Environment: master
> commit a89fae64ac21fb8eefde79f79d2dbe1a0e54c364
> Date: Wed Jul 6 07:58:55 2016 -0700
> Reporter: Jianfeng Jia
> Assignee: Ian Maxon
> Attachments: cc.jstack
>
>
> The 8888/adminconsole showed that there are many pending jobs while the
> ingestion and the query works fine in nc.
> If this situation lasts longer enough, say 2 days, the 19002 API will stop
> response any queries, while the web interface from 19001 port can still
> answer the query.
> I need to restart the cluster to recover the service. Before that I record
> the jstack log of the cc as attached.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
