[
https://issues.apache.org/jira/browse/HDFS-12979?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16839531#comment-16839531
]
Erik Krogen commented on HDFS-12979:
------------------------------------
Thanks [~vagarychen]! The new {{isPrimary}} logic seems good to me.
I think the {{IOException}} vs. {{InterruptedException}} handling seems fine.
However, do we even need the block starting with the comment:
{code}
// cancel the rest of the tasks, and close the pool
{code}
I think this was previously necessary since we broke out of the loop upon the
first exception, but now we cycle through all of the futures regardless. So I
think the logic in this block is already taken care of above.
{{CheckportReceiverEntry}} looks to have a typo, I assume you meant
{{CheckpointReceiverEntry}}. I am also wondering if it makes sense for this
class to also store the {{Future<TransferResult>}} instead of having a separate
map for storing these? I am not sure if it will make things cleaner or messier.
> StandbyNode should upload FsImage to ObserverNode after checkpointing.
> ----------------------------------------------------------------------
>
> Key: HDFS-12979
> URL: https://issues.apache.org/jira/browse/HDFS-12979
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: hdfs
> Reporter: Konstantin Shvachko
> Assignee: Chen Liang
> Priority: Major
> Attachments: HDFS-12979.001.patch, HDFS-12979.002.patch,
> HDFS-12979.003.patch, HDFS-12979.004.patch, HDFS-12979.005.patch,
> HDFS-12979.006.patch
>
>
> ObserverNode does not create checkpoints. So it's fsimage file can get very
> old making bootstrap of ObserverNode too long. A StandbyNode should copy
> latest fsimage to ObserverNode(s) along with ANN.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
