[
https://issues.apache.org/jira/browse/HDDS-3077?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17070956#comment-17070956
]
Istvan Fajth commented on HDDS-3077:
------------------------------------
Hi [~bharat],
the scenario we were trying out with Beata, is to set up an HBase cluster that
uses Ozone via S3 Gateway as an S3 object store behind the HBase data files,
while HBase was using HDFS still for the WAL files.
We ran into some problems during the trials, and we were unsure whether the
access key or the secret key format was causing a problem, and we realized that
we are not allowed to change the access key if we do not want to loose the
reachability of our data in the S3 buckets we already created.
I suggested to report this behaviour because, in S3 at least, user is allowed
to re-generate the access key, but if we bound the volume which contains the
buckets to the access key, we effectively make the user unable to re-generate
it, and this is something which might be unexpected from the user's
perspective, who wants to use us as he uses S3.
> Lost volume after changing access key
> -------------------------------------
>
> Key: HDDS-3077
> URL: https://issues.apache.org/jira/browse/HDDS-3077
> Project: Hadoop Distributed Data Store
> Issue Type: Bug
> Reporter: Beata Sudi
> Priority: Critical
>
> When using the S3 API, Ozone generates the volume depending on the user's
> access key. When the access key is changed, it becomes no longer reachable
> by the user.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
