[
https://issues.apache.org/jira/browse/HDFS-9079?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14905562#comment-14905562
]
Jing Zhao commented on HDFS-9079:
---------------------------------
Thanks for working on this, [~zhz]!
To have an event queue for failure handling is a very good direction and
actually I also planned to do it before. I have not gone through the whole
patch yet, but one quick suggestion is that maybe you can consider doing this
work on top of HDFS-9040. In HDFS-9040 we've made the logic more clear and
changes like syncing the streamers before failure handling should be necessary.
Also we consolidate {{ErrorState}} and {{failed}}, and simplify the block group
length calculation logic a lot. Hopefully these changes can make your work much
easier here :)
> Erasure coding: preallocate multiple generation stamps and serialize updates
> from data streamers
> ------------------------------------------------------------------------------------------------
>
> Key: HDFS-9079
> URL: https://issues.apache.org/jira/browse/HDFS-9079
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Affects Versions: HDFS-7285
> Reporter: Zhe Zhang
> Assignee: Zhe Zhang
> Attachments: HDFS-9079.00.patch
>
>
> A non-striped DataStreamer goes through the following steps in error handling:
> {code}
> 1) Finds error => 2) Asks NN for new GS => 3) Gets new GS from NN => 4)
> Applies new GS to DN (createBlockOutputStream) => 5) Ack from DN => 6)
> Updates block on NN
> {code}
> To simplify the above we can preallocate GS when NN creates a new striped
> block group ({{FSN#createNewBlock}}). For each new striped block group we can
> reserve {{NUM_PARITY_BLOCKS}} GS's. Then steps 1~3 in the above sequence can
> be saved. If more than {{NUM_PARITY_BLOCKS}} errors have happened we
> shouldn't try to further recover anyway.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
