[
https://issues.apache.org/jira/browse/HDDS-5362?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Rakesh Radhakrishnan updated HDDS-5362:
---------------------------------------
Description:
We are introducing 3 kind of bucket layouts as of now in Ozone:
FSO (File System Optimized):- For this bucket, we need the configuration
“ozone.om.enable.filesystem.paths” to be set to true.
Behavior:
- Supports _atomic rename and delete_ operations.
- Files written to this bucket can be read via S3.
- Keys written via S3 with a delimiter “/” *will* create intermediate
directories.
- This bucket *will* allow interoperability between S3 and FS APIs.
OBS (Object Store):- For this bucket, we need the configuration
“ozone.om.enable.filesystem.paths” to be set to false.
Behavior:
- Keys written to this bucket will behave as a pure object-store.
- Keys written via S3 with a delimiter “/” *will not* create intermediate dirs.
- This bucket *will not* allow interoperability between S3 and FS APIs.
LEGACY:- This bucket layout has been introduced to handle the case of upgrading
an existing cluster with pre-created buckets. Users can’t create a LEGACY
bucket explicitly via the Ozone client.
was:
We are introducing 3 types of buckets as of now in Ozone.
FSO (File System Optimized):- For this bucket, we need the configuration
“ozone.om.enable.filesystem.paths” to be set to true.
Behavior:
- Supports _atomic rename and delete_ operations.
- Files written to this bucket can be read via S3.
- Keys written via S3 with a delimiter “/” *will* create intermediate
directories.
- This bucket *will* allow interoperability between S3 and FS APIs.
OBS (Object Store):- For this bucket, we need the configuration
“ozone.om.enable.filesystem.paths” to be set to false.
Behavior:
- Keys written to this bucket will behave as a pure object-store.
- Keys written via S3 with a delimiter “/” *will not* create intermediate dirs.
- This bucket *will not* allow interoperability between S3 and FS APIs.
LEGACY:- This bucket type has been introduced to handle the case of upgrading
an existing cluster with pre-created buckets. Users can’t create a LEGACY
bucket explicitly via the Ozone client.
> [FSO] Support bucket layout in OM
> ---------------------------------
>
> Key: HDDS-5362
> URL: https://issues.apache.org/jira/browse/HDDS-5362
> Project: Apache Ozone
> Issue Type: Sub-task
> Components: Ozone Manager
> Reporter: Aryan Gupta
> Assignee: Aryan Gupta
> Priority: Major
> Labels: pull-request-available
> Attachments: Ozone Bucket Types_23_June_2021.pdf
>
>
> We are introducing 3 kind of bucket layouts as of now in Ozone:
> FSO (File System Optimized):- For this bucket, we need the configuration
> “ozone.om.enable.filesystem.paths” to be set to true.
> Behavior:
> - Supports _atomic rename and delete_ operations.
> - Files written to this bucket can be read via S3.
> - Keys written via S3 with a delimiter “/” *will* create intermediate
> directories.
> - This bucket *will* allow interoperability between S3 and FS APIs.
> OBS (Object Store):- For this bucket, we need the configuration
> “ozone.om.enable.filesystem.paths” to be set to false.
> Behavior:
> - Keys written to this bucket will behave as a pure object-store.
> - Keys written via S3 with a delimiter “/” *will not* create intermediate
> dirs.
> - This bucket *will not* allow interoperability between S3 and FS APIs.
> LEGACY:- This bucket layout has been introduced to handle the case of
> upgrading an existing cluster with pre-created buckets. Users can’t create a
> LEGACY bucket explicitly via the Ozone client.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
