[ 
https://issues.apache.org/jira/browse/HDFS-10826?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Takanobu Asanuma updated HDFS-10826:
------------------------------------
    Attachment: HDFS-10826.2.patch

I uploaded a new patch.

There are two types of problems, corrupted ec files and missing ec files. This 
patch fixes BlockManager to handle corrupted ec files and NamenodeFsck to 
handle missing ec files. It also includes the unit tests.

> The result of fsck should be CRITICAL when there are unrecoverable ec block 
> groups.
> -----------------------------------------------------------------------------------
>
>                 Key: HDFS-10826
>                 URL: https://issues.apache.org/jira/browse/HDFS-10826
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>          Components: erasure-coding
>            Reporter: Takanobu Asanuma
>            Assignee: Takanobu Asanuma
>         Attachments: HDFS-10826.2.patch, HDFS-10826.WIP.1.patch
>
>
> For RS-6-3, when there is one ec block group and
> 1) 0~3 out of 9 internal blocks are missing, the result of fsck is HEALTY.
> 2) 4~8 out of 9 internal blocks are missing, the result of fsck is HEALTY.
> {noformat}
> Erasure Coded Block Groups:
>  Total size:    536870912 B
>  Total files:   1
>  Total block groups (validated):        1 (avg. block group size 536870912 B)
>   ********************************
>   UNRECOVERABLE BLOCK GROUPS:   1 (100.0 %)
>   ********************************
>  Minimally erasure-coded block groups:  0 (0.0 %)
>  Over-erasure-coded block groups:       0 (0.0 %)
>  Under-erasure-coded block groups:      1 (100.0 %)
>  Unsatisfactory placement block groups: 0 (0.0 %)
>  Default ecPolicy:              RS-DEFAULT-6-3-64k
>  Average block group size:      5.0
>  Missing block groups:          0
>  Corrupt block groups:          0
>  Missing internal blocks:       4 (44.444443 %)
> FSCK ended at Wed Aug 31 13:42:05 JST 2016 in 4 milliseconds
> The filesystem under path '/' is HEALTHY
> {noformat}
> 3) 9 out of 9 internal blocks are missing, the result of fsck is CRITICAL. 
> (Because it is regarded as a missing block group.)
> In case 2), the result should be CRITICAL since the ec block group is 
> unrecoverable.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
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