[ 
https://issues.apache.org/jira/browse/HDFS-7579?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Charles Lamb updated HDFS-7579:
-------------------------------
    Attachment: HDFS-7579.001.patch

[~cnauroth], [~arpitagarwal],

Thanks for the review. I've uploaded a new patch which addresses your comments.

An example message looks like this:

{quote}
2015-01-08 11:41:05,277 INFO  datanode.DataNode 
(BPServiceActor.java:blockReport(525)) - Successfully sent 2 of 2 blockreports 
for 5 total blocks using 2 RPCs. This took 1 msec to generate and 7 msecs for 
RPC and NN processing. Got back 2 commands: FinalizeCommand/5; 
FinalizeCommand/5.
{quote}

Rather than just display the number of reports I now display the number of RPCs 
and the number of reports (as well the number of actual reports sent). 
Regarding success/non-success, it's funny you should mention that. My initial 
version of the patch did exactly that, but then I took it out before posting it 
since I figured it could be derived from the fact that not all the reports 
would have been sent. But being explicit is probably better so I have added it 
back in for the .001 version. cmds can not be null, but I added a final to its 
declaration & initial assignment anyway just to make sure. I think FB might 
have warned about an unnecessary null check if I were to add a check for 
nullness when assigning nCmds.


> Improve log reporting during block report rpc failure
> -----------------------------------------------------
>
>                 Key: HDFS-7579
>                 URL: https://issues.apache.org/jira/browse/HDFS-7579
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: datanode
>    Affects Versions: 2.7.0
>            Reporter: Charles Lamb
>            Assignee: Charles Lamb
>            Priority: Minor
>              Labels: supportability
>         Attachments: HDFS-7579.000.patch, HDFS-7579.001.patch
>
>
> During block reporting, if the block report RPC fails, for example because it 
> exceeded the max rpc len, we should still produce some sort of LOG.info 
> output to help with debugging.



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

Reply via email to