feat: implement prefix-level versioning exclusion (#14828)

Spark/Hadoop workloads which use Hadoop MR 
Committer v1/v2 algorithm upload objects to a 
temporary prefix in a bucket. These objects are 
'renamed' to a different prefix on Job commit. 
Object storage admins are forced to configure 
separate ILM policies to expire these objects 
and their versions to reclaim space.

Our solution:

This can be avoided by simply marking objects 
under these prefixes to be excluded from versioning, 
as shown below. Consequently, these objects are 
excluded from replication, and don't require ILM 
policies to prune unnecessary versions.

-  MinIO Extension to Bucket Version Configuration
```xml
<VersioningConfiguration xmlns="http://s3.amazonaws.com/doc/2006-03-01/"> 
        <Status>Enabled</Status>
        <ExcludeFolders>true</ExcludeFolders>
        <ExcludedPrefixes>
          <Prefix>app1-jobs/*/_temporary/</Prefix>
        </ExcludedPrefixes>
        <ExcludedPrefixes>
          <Prefix>app2-jobs/*/__magic/</Prefix>
        </ExcludedPrefixes>

        <!-- .. up to 10 prefixes in all -->     
</VersioningConfiguration>
```
Note: `ExcludeFolders` excludes all folders in a bucket 
from versioning. This is required to prevent the parent 
folders from accumulating delete markers, especially
those which are shared across spark workloads 
spanning projects/teams.

- To enable version exclusion on a list of prefixes

```
mc version enable --excluded-prefixes "app1-jobs/*/_temporary/,app2-jobs/*/_magic," --exclude-prefix-marker myminio/test
```
This commit is contained in:
Krishnan Parthasarathi
2022-05-06 19:05:28 -07:00
committed by GitHub
parent 3ec1844e4a
commit ad8e611098
20 changed files with 478 additions and 59 deletions

View File

@@ -243,6 +243,26 @@ Replication from a source bucket to multiple destination buckets is supported. F
Note that on the source side, the `X-Amz-Replication-Status` changes from `PENDING` to `COMPLETED` after replication succeeds to each of the targets. On the destination side, a `X-Amz-Replication-Status` status of `REPLICA` indicates that the object was replicated successfully. Any replication failures are automatically re-attempted during a periodic disk scanner cycle.
### Interaction with extended Bucket Versioning configuration
When Bucket Versioning with excluded prefixes are configured objects matching these prefixes are excluded from being versioned and replicated.
```
<VersioningConfiguration xmlns="http://s3.amazonaws.com/doc/2006-03-01/">
<Status>Enabled</Status>
<ExcludeFolders>true</ExcludeFolders>
<ExcludedPrefixes>
<Prefix>app1-jobs/*/_temporary/</Prefix>
</ExcludedPrefixes>
<ExcludedPrefixes>
<Prefix>app2-jobs/*/_magic/</Prefix>
</ExcludedPrefixes>
<!-- .. up to 10 prefixes in all -->
</VersioningConfiguration>
```
In the above sample config, objects under prefixes matching any of the `ExcludedPrefixes` glob patterns will neither be versioned nor replicated.
## Explore Further
- [MinIO Bucket Replication Design](https://github.com/minio/minio/blob/master/docs/bucket/replication/DESIGN.md)