You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Andrés de la Peña (JIRA)" <ji...@apache.org> on 2017/04/06 12:17:41 UTC

[jira] [Commented] (CASSANDRA-13276) Regression on CASSANDRA-11416: can't load snapshots of tables with dropped columns

    [ https://issues.apache.org/jira/browse/CASSANDRA-13276?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15958813#comment-15958813 ] 

Andrés de la Peña commented on CASSANDRA-13276:
-----------------------------------------------

It seems that {{NativeSSTableLoaderClient.createTableMetadata}} creates table metadata without taking dropped columns into account. Working in a patch.

> Regression on CASSANDRA-11416: can't load snapshots of tables with dropped columns
> ----------------------------------------------------------------------------------
>
>                 Key: CASSANDRA-13276
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-13276
>             Project: Cassandra
>          Issue Type: Bug
>            Reporter: Matt Kopit
>            Assignee: Andrés de la Peña
>             Fix For: 3.0.13, 3.11.0, 4.0
>
>
> I'm running Cassandra 3.10 and running into the exact same issue described in CASSANDRA-11416: 
> 1. A table is created with columns 'a' and 'b'
> 2. Data is written to the table
> 3. Drop column 'b'
> 4. Take a snapshot
> 5. Drop the table
> 6. Run the snapshot schema.cql to recreate the table and the run the alter
> 7. Try to restore the snapshot data using sstableloader
> sstableloader yields the error:
> java.lang.RuntimeException: Unknown column b during deserialization



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)