[
https://issues.apache.org/jira/browse/HDFS-7864?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14347700#comment-14347700
]
Jing Zhao commented on HDFS-7864:
---------------------------------
bq. I think "striped blocks" in this Jira's description actually denotes to
"BlockGroup" in "HDFSErasureCodingDesign-20150206.pdf" , right?
Yes, you're right. We should update the doc later for this.
bq. so a "BlockGroup" should be counted as 9 received blocks, right?
A block group (or a striped block) consists of 9 blocks in this case. But for
safemode calculation, I guess we can still count it as 1 block. This is because
for a single striped block (suppose it's m+k), we need a different logic to
declare it's "safe" (as increase the safe amount by 1): the NN has received m
blocks belonging to it from block reports (since we need m blocks to recover
all the data/parity blocks).
> Erasure Coding: Update safemode calculation for striped blocks
> --------------------------------------------------------------
>
> Key: HDFS-7864
> URL: https://issues.apache.org/jira/browse/HDFS-7864
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Jing Zhao
> Assignee: GAO Rui
>
> We need to update the safemode calculation for striped blocks. Specifically,
> each striped block now consists of multiple data/parity blocks stored in
> corresponding DataNodes. The current code's calculation is thus inconsistent:
> each striped block is only counted as 1 expected block, while each of its
> member block may increase the number of received blocks by 1.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
