[
https://issues.apache.org/jira/browse/PHOENIX-5025?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16719375#comment-16719375
]
Kadir OZDEMIR commented on PHOENIX-5025:
----------------------------------------
[~apurtell], No, it was not intentional. In my environment, the Java version is
set to 1.8 and I was under the impression that it is the right version for
Phoenix. I will reopen the Jira and fix the incompatibility issue. It is
interesting that Hadoop QA has not reported the problem. So, how should a
developer find/check that he/she is using the right Java version for a given
branch? Where is this info?
> Tool to clean up orphan views
> -----------------------------
>
> Key: PHOENIX-5025
> URL: https://issues.apache.org/jira/browse/PHOENIX-5025
> Project: Phoenix
> Issue Type: New Feature
> Reporter: Kadir OZDEMIR
> Assignee: Kadir OZDEMIR
> Priority: Major
> Fix For: 5.0.0
>
> Attachments: PHOENIX-5025.master.0001.patch, PHOENIX-5025.master.patch
>
>
> A view without its base table is an orphan view. Since views are virtual
> tables and their data is stored in their base tables, they are useless when
> they become orphan. A base table can have child views, grandchild views and
> so on. Due to some reasons/bugs, when a base table was dropped, its views
> were not not properly cleaned up in the past. For example, the drop table
> code did not support cleaning up grandchild views. This has been recently
> fixed by PHOENIX-4764. Although PHOENIX-4764 prevents new orphan views due to
> table drop operations, it does not clean up existing orphan views. It is also
> believed that when the system catalog table was split due to a bug in the
> past, it also contributed to creating orphan views as Phoenix did not support
> splittable system catalog. Therefore, Phoenix needs a tool to clean up orphan
> views.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
