[
https://issues.apache.org/jira/browse/HBASE-16570?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15473358#comment-15473358
]
Heng Chen commented on HBASE-16570:
-----------------------------------
{quote}
+ if (future != null) {
+ blockDistbn = future.get();
{quote}
The future could be null? It seems that the asyncGetBlockDistribution always
return one Object due to you catch all exceptions
{quote}
+ public ListenableFuture<HDFSBlocksDistribution> asyncGetBlockDistribution(
+ HRegionInfo hri) {
+ try {
+ return loader.reload(hri, EMPTY_BLOCK_DISTRIBUTION);
+ } catch (Exception e) {
+ return Futures.immediateFuture(new HDFSBlocksDistribution());
+ }
+ }
{quote}
By the way, {{new HDFSBlocksDistribution() }} could be replaced by
EMPTY_BLOCK_DISTRIBUTION ?
{quote}
getTopBlockLocations(HRegionInfo region)
{quote}
It could be removed, it seems no reference to the method.
Besides that, the patch looks good, could it accelerate the startup? Do you
have some numbers about it?
> Compute region locality in parallel at startup
> ----------------------------------------------
>
> Key: HBASE-16570
> URL: https://issues.apache.org/jira/browse/HBASE-16570
> Project: HBase
> Issue Type: Sub-task
> Reporter: binlijin
> Assignee: binlijin
> Attachments: HBASE-16570-master_V1.patch, HBASE-16570-master_V2.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
