You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@phoenix.apache.org by "Kadir OZDEMIR (Jira)" <ji...@apache.org> on 2020/03/21 02:08:00 UTC

[jira] [Commented] (PHOENIX-5788) Handling deletes on non-existing rows

    [ https://issues.apache.org/jira/browse/PHOENIX-5788?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17063719#comment-17063719 ] 

Kadir OZDEMIR commented on PHOENIX-5788:
----------------------------------------

This actually does not cause an issue for inline verification. However, it can confuse the human who is comparing the result of raw scan of the data table with that of the index table. So, I will continue with fixing this issue. I have decided not to apply the delete mutation on the non-existent data row.

> Handling deletes on non-existing rows
> -------------------------------------
>
>                 Key: PHOENIX-5788
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-5788
>             Project: Phoenix
>          Issue Type: Sub-task
>            Reporter: Kadir OZDEMIR
>            Assignee: Kadir OZDEMIR
>            Priority: Major
>
> PHOENIX-5709 skips generating index delete mutations if the data table rows (and thus the index table rows) do not exist. However, it allows the delete mutations to be applied on the data table. This does not create a data integrity problem but creates an extra case to handle for the inline verification implementation to consider. PHOENIX-5709 currently does not handle this case. We should either apply these delete statements to both tables or apply none of them.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)