[
https://issues.apache.org/jira/browse/HDDS-837?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16686848#comment-16686848
]
Jitendra Nath Pandey edited comment on HDDS-837 at 11/14/18 4:47 PM:
---------------------------------------------------------------------
I think we should also add originPipelineId along with the originNodeId. Even
if it is not used actively, it will be useful for debugging or looking at the
history of a container. In case of multi-raft, with multiple pipelines being
served from the same DN, it will be helpful to be able to distinguish which
container belongs to which pipeline.
was (Author: jnp):
I think we should also add pipelineId along with the originNodeId. Even if it
is not used actively, it will be useful for debugging or looking at the history
of a container. In case of multi-raft, with multiple pipelines being served
from the same DN, it will be helpful to be able to distinguish which container
belongs to which pipeline.
> Persist originNodeId as part of .container file in datanode
> -----------------------------------------------------------
>
> Key: HDDS-837
> URL: https://issues.apache.org/jira/browse/HDDS-837
> Project: Hadoop Distributed Data Store
> Issue Type: Improvement
> Components: Ozone Datanode
> Reporter: Nanda kumar
> Assignee: Nanda kumar
> Priority: Major
>
> To differentiate the replica of QUASI_CLOSED containers we need
> {{originNodeId}} field. With this field, we can uniquely identify a
> QUASI_CLOSED container replica. This will be needed when we want to CLOSE a
> QUASI_CLOSED container.
> This field will be set by the node where the container is created and stored
> as part of {{.container}} file and will be sent as part of ContainerReport to
> SCM.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
