[ 
https://issues.apache.org/jira/browse/HDFS-15624?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17222273#comment-17222273
 ] 

Ayush Saxena commented on HDFS-15624:
-------------------------------------

HDFS-15660 is handling a different issue, things getting fixed here are quite 
specific to NVDMIM, like the broken FsImage Compatibility, due to change in 
ordinal of Storage Types. Rolling Upgrade issue.

Regarding HDFS-15660 : That tends to handle the exception due to missing 
storage type at Client, and so the protobuf response while decoding the new 
Storage Type, fetches an exception. This would happen with NVDMIM also, in case 
of 3.3.0 client and and 3.4.0 server, but that is something not being chased 
here.

No point holding this IMO, I didn't check at what stage the code is now, 
considering Vinay was following. I think that is almost at conclusion, and we 
should get this in and sort the mess from HDFS-15025. 

But I am ok holding it as well, but in that case we should revert HDFS-15025, 
So, as if this doesn't get concluded for any reasons, later getting rid of this 
shouldn't be a problem, and even prevent someone backporting the original in 
there internal versions.

[~vinayakumarb]/ [~liuml07] let me know if you folks too want to hold it for 
HDFS-15660(This won't be too quick and is quite different as well), Will revert 
the original by tomorrow EOD and we can track there in that case. 

>  Fix the SetQuotaByStorageTypeOp problem after updating hadoop 
> ---------------------------------------------------------------
>
>                 Key: HDFS-15624
>                 URL: https://issues.apache.org/jira/browse/HDFS-15624
>             Project: Hadoop HDFS
>          Issue Type: Bug
>          Components: hdfs
>            Reporter: YaYun Wang
>            Priority: Major
>              Labels: pull-request-available
>          Time Spent: 4h 10m
>  Remaining Estimate: 0h
>
> HDFS-15025 adds a new storage Type NVDIMM, changes the ordinal() of the enum 
> of StorageType. And, setting the quota by storageType depends on the 
> ordinal(), therefore, it may cause the setting of quota to be invalid after 
> upgrade.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org

Reply via email to