-
Notifications
You must be signed in to change notification settings - Fork 3.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
HBASE-28502 Cleanup old backup manifest logic #5871
HBASE-28502 Cleanup old backup manifest logic #5871
Conversation
In older versions of HBase's backup mechanism, a manifest was written per table being backed up. This was since refactored to one manifest per backup, but the manifest code was not updated. A concrete issue with the old code was that the manifest for full backups did not correctly list the tables included in the backup.
🎊 +1 overall
This message was automatically generated. |
🎊 +1 overall
This message was automatically generated. |
🎊 +1 overall
This message was automatically generated. |
🎊 +1 overall
This message was automatically generated. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This looks good to me.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good to me. It would be good to solicit comments from whomever made the change that deprecated the old manifest logic in the first place. @DieterDP-ng do you know whom that might be? Maybe check the relevant commit logs. Please @-mention them on this ticket.
return new Path(getLogBackupDir(backupRootDir, backupId)); | ||
} | ||
|
||
// TODO we do not keep WAL files anymore |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Removal of this TODO is accurate?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
My interpretation of this TODO (introduced in HBASE-14135), is that it was already implemented in that same commit. If that weren't the case, the TODO message is too vague for me to understand what needs to be done, rendering it useless anyway.
Looks like both modified test methods were executed and passed. |
I think this was done in HBASE-14135, committed by @joshelser, with @VladRodionov mentioned. |
👋🏼 sorry, I don't recall exactly what was being put into the incremental backups anymore. I remember holding on to all of the WAL files quickly got out of control and we talked about consolidating multiple incremental backups (lists of WALs) into hfiles, but I don't remember exactly what was implemented. |
Merged to master. I'll start backporting and holler if I have any issues. Thanks for the contribution @DieterDP-ng and for the review @rmdmattingly ! |
In older versions of HBase's backup mechanism, a manifest was written per table being backed up. This was since refactored to one manifest per backup, but the manifest code was not updated. A concrete issue with the old code was that the manifest for full backups did not correctly list the tables included in the backup. Signed-off-by: Nick Dimiduk <ndimiduk@apache.org> Reviewed-by: Ray Mattingly <rmdmattingly@gmail.com>
In older versions of HBase's backup mechanism, a manifest was written per table being backed up. This was since refactored to one manifest per backup, but the manifest code was not updated. A concrete issue with the old code was that the manifest for full backups did not correctly list the tables included in the backup. Signed-off-by: Nick Dimiduk <ndimiduk@apache.org> Reviewed-by: Ray Mattingly <rmdmattingly@gmail.com>
In older versions of HBase's backup mechanism, a manifest was written per table being backed up. This was since refactored to one manifest per backup, but the manifest code was not updated. A concrete issue with the old code was that the manifest for full backups did not correctly list the tables included in the backup. Signed-off-by: Nick Dimiduk <ndimiduk@apache.org> Reviewed-by: Ray Mattingly <rmdmattingly@gmail.com>
In older versions of HBase's backup mechanism, a manifest was written per table being backed up. This was since refactored to one manifest per backup, but the manifest code was not updated. A concrete issue with the old code was that the manifest for full backups did not correctly list the tables included in the backup. Signed-off-by: Nick Dimiduk <ndimiduk@apache.org> Reviewed-by: Ray Mattingly <rmdmattingly@gmail.com>
In older versions of HBase's backup mechanism, a manifest was written per table being backed up. This was since refactored to one manifest per backup, but the manifest code was not updated. A concrete issue with the old code was that the manifest for full backups did not correctly list the tables included in the backup. Signed-off-by: Nick Dimiduk <ndimiduk@apache.org> Reviewed-by: Ray Mattingly <rmdmattingly@gmail.com>
In older versions of HBase's backup mechanism, a manifest was written per table being backed up. This was since refactored to one manifest per backup, but the manifest code was not updated. A concrete issue with the old code was that the manifest for full backups did not correctly list the tables included in the backup. Signed-off-by: Nick Dimiduk <ndimiduk@apache.org> Reviewed-by: Ray Mattingly <rmdmattingly@gmail.com>
In older versions of HBase's backup mechanism, a manifest was written per table being backed up. This was since refactored to one manifest per backup, but the manifest code was not updated. A concrete issue with the old code was that the manifest for full backups did not correctly list the tables included in the backup. Signed-off-by: Nick Dimiduk <ndimiduk@apache.org> Reviewed-by: Ray Mattingly <rmdmattingly@gmail.com>
In older versions of HBase's backup mechanism, a manifest was written per table being backed up. This was since refactored to one manifest per backup, but the manifest code was not updated. A concrete issue with the old code was that the manifest for full backups did not correctly list the tables included in the backup. Signed-off-by: Nick Dimiduk <ndimiduk@apache.org> Reviewed-by: Ray Mattingly <rmdmattingly@gmail.com>
…he#5871) In older versions of HBase's backup mechanism, a manifest was written per table being backed up. This was since refactored to one manifest per backup, but the manifest code was not updated. A concrete issue with the old code was that the manifest for full backups did not correctly list the tables included in the backup. Signed-off-by: Nick Dimiduk <ndimiduk@apache.org> Reviewed-by: Ray Mattingly <rmdmattingly@gmail.com>
…he#5871) In older versions of HBase's backup mechanism, a manifest was written per table being backed up. This was since refactored to one manifest per backup, but the manifest code was not updated. A concrete issue with the old code was that the manifest for full backups did not correctly list the tables included in the backup. Signed-off-by: Nick Dimiduk <ndimiduk@apache.org> Reviewed-by: Ray Mattingly <rmdmattingly@gmail.com>
…he#5871) (#94) In older versions of HBase's backup mechanism, a manifest was written per table being backed up. This was since refactored to one manifest per backup, but the manifest code was not updated. A concrete issue with the old code was that the manifest for full backups did not correctly list the tables included in the backup. Reviewed-by: Ray Mattingly <rmdmattingly@gmail.com> Signed-off-by: Nick Dimiduk <ndimiduk@apache.org> Co-authored-by: DieterDP <90392398+DieterDP-ng@users.noreply.github.com>
…he#5871) (#95) In older versions of HBase's backup mechanism, a manifest was written per table being backed up. This was since refactored to one manifest per backup, but the manifest code was not updated. A concrete issue with the old code was that the manifest for full backups did not correctly list the tables included in the backup. Reviewed-by: Ray Mattingly <rmdmattingly@gmail.com> Signed-off-by: Nick Dimiduk <ndimiduk@apache.org> Co-authored-by: DieterDP <90392398+DieterDP-ng@users.noreply.github.com>
In older versions of HBase's backup mechanism, a manifest was written per table being backed up. This was since refactored to one manifest per backup, but the manifest code was not updated. A concrete issue with the old code was that the manifest for full backups did not correctly list the tables included in the backup. Signed-off-by: Nick Dimiduk <ndimiduk@apache.org> Reviewed-by: Ray Mattingly <rmdmattingly@gmail.com>
…he#5871) (#94) In older versions of HBase's backup mechanism, a manifest was written per table being backed up. This was since refactored to one manifest per backup, but the manifest code was not updated. A concrete issue with the old code was that the manifest for full backups did not correctly list the tables included in the backup. Reviewed-by: Ray Mattingly <rmdmattingly@gmail.com> Signed-off-by: Nick Dimiduk <ndimiduk@apache.org> Co-authored-by: DieterDP <90392398+DieterDP-ng@users.noreply.github.com>
In older versions of HBase's backup mechanism, a manifest was written per table being backed up. This was since refactored to one manifest per backup, but the manifest code was not updated. A concrete issue with the old code was that the manifest for full backups did not correctly list the tables included in the backup.