[
https://issues.apache.org/jira/browse/HDDS-6654?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17529540#comment-17529540
]
Attila Doroszlai commented on HDDS-6654:
----------------------------------------
Please consider the following:
* Multiple links can point to the same bucket, it's not a 1:1 mapping.
* Link can point to non-existent bucket (dangling link).
* Source bucket to which previously dangling link points may be created,
"fixing" the bucket.
* Source bucket may be deleted even if there are existing links to it.
* (Note: bucket links are not specific to s3v volume.)
Users (or admins, too) can already filter {{ozone sh bucket list}} output for
source volume and bucket:
{code}
$ ozone sh volume create /vol1
$ ozone sh bucket create /vol1/bucket1
$ ozone sh bucket link /vol1/bucket1 /vol1/link-to-bucket1
$ ozone sh bucket link /vol1/bucket2 /vol1/link-to-nonexistent-bucket
$ ozone sh bucket link /vol1/bucket2 /vol1/other-link-to-nonexistent-bucket
$ ozone sh bucket list /vol1 | jq -r '.[] | select(.sourceVolume == "vol1") |
select(.sourceBucket == "bucket2")'
{
"volumeName": "vol1",
"name": "link-to-nonexistent-bucket",
...
"sourceVolume": "vol1",
"sourceBucket": "bucket2",
...
"link": true
}
{
"volumeName": "vol1",
"name": "other-link-to-nonexistent-bucket",
...
"sourceVolume": "vol1",
"sourceBucket": "bucket2",
...
"link": true
}
{code}
> Source bucket does not have linked bucket details
> --------------------------------------------------
>
> Key: HDDS-6654
> URL: https://issues.apache.org/jira/browse/HDDS-6654
> Project: Apache Ozone
> Issue Type: Bug
> Components: OM
> Reporter: Lokesh Jain
> Priority: Major
>
> Ozone provides functionality to link any bucket with a placeholder s3 bucket
> in s3v volume. Using this feature user can access the ozone bucket via the s3
> gateway as well.
> Problem is the link info (source bucket and source volume) only exists in the
> s3 bucket. The Jira aims to provide the link details (s3 bucket name) in the
> source bucket as well.
> Lack of link details in source bucket would mean that admin would have to do
> bookkeeping to get info on which s3 bucket is linked to a particular source
> bucket. Or the admin would have to look through all buckets in s3v in order
> to get that info today.
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
