You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Michael Krumpholz (JIRA)" <ji...@apache.org> on 2012/09/21 09:01:08 UTC
[jira] [Created] (CASSANDRA-4696) improve ALTER TABLE
Michael Krumpholz created CASSANDRA-4696:
--------------------------------------------
Summary: improve ALTER TABLE
Key: CASSANDRA-4696
URL: https://issues.apache.org/jira/browse/CASSANDRA-4696
Project: Cassandra
Issue Type: Improvement
Affects Versions: 1.1.5
Reporter: Michael Krumpholz
Please improve the ALTER TABLE statement so that it automatically checks the data in the table against the new data type to change to. I know that it's in the docs that this is the current state and not checked. It may be the easiest and fastest way for implementation for now but it should be improved in the future.
docs on ALTER:
{noformat}
The column must already have a type in the column family metadata. The column may or may not already exist in current rows (no validation of existing data occurs). The bytes stored in values for that column remain unchanged, and if existing data is not deserializable according to the new type, your CQL driver or interface might report errors.
{noformat}
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (CASSANDRA-4696) improve ALTER TABLE
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-4696?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis resolved CASSANDRA-4696.
---------------------------------------
Resolution: Won't Fix
doing a seq scan across potentially TB of data is not an option. if you change it to something incompatible by mistake, you can always change it back.
> improve ALTER TABLE
> -------------------
>
> Key: CASSANDRA-4696
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4696
> Project: Cassandra
> Issue Type: Improvement
> Affects Versions: 1.1.5
> Reporter: Michael Krumpholz
> Labels: alter, cql, datatype
>
> Please improve the ALTER TABLE statement so that it automatically checks the data in the table against the new data type to change to. I know that it's in the docs that this is the current state and not checked. It may be the easiest and fastest way for implementation for now but it should be improved in the future.
> docs on ALTER:
> {noformat}
> The column must already have a type in the column family metadata. The column may or may not already exist
> in current rows (no validation of existing data occurs). The bytes stored in values for that column remain
> unchanged, and if existing data is not deserializable according to the new type, your CQL driver or
> interface might report errors.
> {noformat}
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (CASSANDRA-4696) improve ALTER TABLE
Posted by "Michael Krumpholz (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-4696?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13460527#comment-13460527 ]
Michael Krumpholz commented on CASSANDRA-4696:
----------------------------------------------
that kind of locking would be there too if the schema change (or migration to a new structure) would require a longer downtime with code external to cassandra, the result for the users is the same: they can't use the application that accesses the data or am i wrong there?
> improve ALTER TABLE
> -------------------
>
> Key: CASSANDRA-4696
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4696
> Project: Cassandra
> Issue Type: Improvement
> Affects Versions: 1.1.5
> Reporter: Michael Krumpholz
> Labels: alter, cql, datatype
>
> Please improve the ALTER TABLE statement so that it automatically checks the data in the table against the new data type to change to. I know that it's in the docs that this is the current state and not checked. It may be the easiest and fastest way for implementation for now but it should be improved in the future.
> docs on ALTER:
> {noformat}
> The column must already have a type in the column family metadata. The column may or may not already exist
> in current rows (no validation of existing data occurs). The bytes stored in values for that column remain
> unchanged, and if existing data is not deserializable according to the new type, your CQL driver or
> interface might report errors.
> {noformat}
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (CASSANDRA-4696) improve ALTER TABLE
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-4696?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13460516#comment-13460516 ]
Jonathan Ellis commented on CASSANDRA-4696:
-------------------------------------------
It's unfortunate if you don't know what data types you've inserted, but inflicting the kind of locking + scanning validation that other databases do for schema changes on everyone, because of this possibility, is optimizing for the wrong scenario.
> improve ALTER TABLE
> -------------------
>
> Key: CASSANDRA-4696
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4696
> Project: Cassandra
> Issue Type: Improvement
> Affects Versions: 1.1.5
> Reporter: Michael Krumpholz
> Labels: alter, cql, datatype
>
> Please improve the ALTER TABLE statement so that it automatically checks the data in the table against the new data type to change to. I know that it's in the docs that this is the current state and not checked. It may be the easiest and fastest way for implementation for now but it should be improved in the future.
> docs on ALTER:
> {noformat}
> The column must already have a type in the column family metadata. The column may or may not already exist
> in current rows (no validation of existing data occurs). The bytes stored in values for that column remain
> unchanged, and if existing data is not deserializable according to the new type, your CQL driver or
> interface might report errors.
> {noformat}
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (CASSANDRA-4696) improve ALTER TABLE
Posted by "Michael Krumpholz (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-4696?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13460502#comment-13460502 ]
Michael Krumpholz commented on CASSANDRA-4696:
----------------------------------------------
but is scanning data manually and having to implement custom code every time any different or better than having done that more in a generic way in cassandra?
> improve ALTER TABLE
> -------------------
>
> Key: CASSANDRA-4696
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4696
> Project: Cassandra
> Issue Type: Improvement
> Affects Versions: 1.1.5
> Reporter: Michael Krumpholz
> Labels: alter, cql, datatype
>
> Please improve the ALTER TABLE statement so that it automatically checks the data in the table against the new data type to change to. I know that it's in the docs that this is the current state and not checked. It may be the easiest and fastest way for implementation for now but it should be improved in the future.
> docs on ALTER:
> {noformat}
> The column must already have a type in the column family metadata. The column may or may not already exist
> in current rows (no validation of existing data occurs). The bytes stored in values for that column remain
> unchanged, and if existing data is not deserializable according to the new type, your CQL driver or
> interface might report errors.
> {noformat}
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (CASSANDRA-4696) improve ALTER TABLE
Posted by "Michael Krumpholz (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-4696?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Michael Krumpholz updated CASSANDRA-4696:
-----------------------------------------
Description:
Please improve the ALTER TABLE statement so that it automatically checks the data in the table against the new data type to change to. I know that it's in the docs that this is the current state and not checked. It may be the easiest and fastest way for implementation for now but it should be improved in the future.
docs on ALTER:
{noformat}
The column must already have a type in the column family metadata. The column may or may not already exist
in current rows (no validation of existing data occurs). The bytes stored in values for that column remain
unchanged, and if existing data is not deserializable according to the new type, your CQL driver or
interface might report errors.
{noformat}
was:
Please improve the ALTER TABLE statement so that it automatically checks the data in the table against the new data type to change to. I know that it's in the docs that this is the current state and not checked. It may be the easiest and fastest way for implementation for now but it should be improved in the future.
docs on ALTER:
{noformat}
The column must already have a type in the column family metadata. The column may or may not already exist in current rows (no validation of existing data occurs). The bytes stored in values for that column remain unchanged, and if existing data is not deserializable according to the new type, your CQL driver or interface might report errors.
{noformat}
> improve ALTER TABLE
> -------------------
>
> Key: CASSANDRA-4696
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4696
> Project: Cassandra
> Issue Type: Improvement
> Affects Versions: 1.1.5
> Reporter: Michael Krumpholz
> Labels: alter, cql, datatype
>
> Please improve the ALTER TABLE statement so that it automatically checks the data in the table against the new data type to change to. I know that it's in the docs that this is the current state and not checked. It may be the easiest and fastest way for implementation for now but it should be improved in the future.
> docs on ALTER:
> {noformat}
> The column must already have a type in the column family metadata. The column may or may not already exist
> in current rows (no validation of existing data occurs). The bytes stored in values for that column remain
> unchanged, and if existing data is not deserializable according to the new type, your CQL driver or
> interface might report errors.
> {noformat}
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira