[
https://issues.apache.org/jira/browse/KAFKA-667?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jay Kreps updated KAFKA-667:
----------------------------
Attachment: KAFKA-667-v1.patch
Trivial patch to change the name of the highwatermark file. Ready for review.
> Rename .highwatermark file
> --------------------------
>
> Key: KAFKA-667
> URL: https://issues.apache.org/jira/browse/KAFKA-667
> Project: Kafka
> Issue Type: Improvement
> Affects Versions: 0.8
> Reporter: Jay Kreps
> Assignee: Jay Kreps
> Priority: Minor
> Attachments: KAFKA-667-v1.patch
>
>
> The 0.8 branch currently has a file in each log directory called
> .highwatermark
> Soon we hope to add two more files in the same format. One will hold the
> cleaner position for log deduplication, and the other will hold the flusher
> position for log flush. Each of these is sort of a "highwater mark". It would
> be good to rename .highwatermark to be a little bit more intuitive when we
> add these other files. I propose:
> replication-offset-checkpoint
> flusher-offset-checkpoint
> cleaner-offset-checkpoint
> replication-offset-checkpoint would replace the .highwatermark file. I am not
> making them dot files since they represent an important part of the
> persistent state and so the user should see it. Also shell * doesn't match
> hidden files, so if you did something like cp my_log/* to my_backup_log/* you
> would not get corresponding .highwatermark file.
> I am filing this bug now because it might be nice to just make this trivial
> change now and avoid having to handle backwards compatibility later.
--
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
