You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@kudu.apache.org by "Dan Burkert (JIRA)" <ji...@apache.org> on 2016/09/13 17:55:22 UTC

[jira] [Commented] (KUDU-1607) Unable to delete FAILED tablets as part of drop table

    [ https://issues.apache.org/jira/browse/KUDU-1607?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15487910#comment-15487910 ] 

Dan Burkert commented on KUDU-1607:
-----------------------------------

{{kudu remote_tablet delete}} is not able to delete the failed tablet.  The tablet goes into {{TABLET_DATA_TOMBSTONED}} state, but it's still listed on the /tablets endpoint, and on tserver restart it comes back with state {{FAILED}}.

> Unable to delete FAILED tablets as part of drop table
> -----------------------------------------------------
>
>                 Key: KUDU-1607
>                 URL: https://issues.apache.org/jira/browse/KUDU-1607
>             Project: Kudu
>          Issue Type: Bug
>    Affects Versions: 0.10.0
>            Reporter: Dan Burkert
>
> When dropping a table with a failed tablet, the tablet will fail to be deleted. The /tablets endpoint shows its Last Status as: {{Invalid argument: Unable to delete on-disk data from tablet 2e3b95dba81044748a9a52eb138dd59c: The metadata for tablet 2e3b95dba81044748a9a52eb138dd59c still references orphaned blocks. Call DeleteTabletData() first}}
> Steps to repro:
> 1. Create a table with a single tablet, single replica.
> 2. Load enough data to have the WAL roll a segment
> 3. Corrupt the magic bytes of a non-current segment
> 4. Restart the TS
> 5. Drop the table



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)