[
https://issues.apache.org/jira/browse/HBASE-19561?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ted Yu updated HBASE-19561:
---------------------------
Resolution: Fixed
Hadoop Flags: Reviewed
Fix Version/s: 2.0.0-beta-1
Status: Resolved (was: Patch Available)
Thanks for the patch, Guangxu.
Thanks for the review, Zach.
> maxCacheSize in CacheEvictionStats can't be accumulated repeatedly When
> dealing with each region
> ------------------------------------------------------------------------------------------------
>
> Key: HBASE-19561
> URL: https://issues.apache.org/jira/browse/HBASE-19561
> Project: HBase
> Issue Type: Bug
> Affects Versions: 2.0.0
> Reporter: Guangxu Cheng
> Assignee: Guangxu Cheng
> Fix For: 2.0.0-beta-1
>
> Attachments: HBASE-19561.master.001.patch
>
>
> {code:title=HRegionServer.java|borderStyle=solid}
> public CacheEvictionStats clearRegionBlockCache(Region region) {
> BlockCache blockCache = this.getCacheConfig().getBlockCache();
> long evictedBlocks = 0;
> for(Store store : region.getStores()) {
> for(StoreFile hFile : store.getStorefiles()) {
> evictedBlocks +=
> blockCache.evictBlocksByHfileName(hFile.getPath().getName());
> }
> }
> return CacheEvictionStats.builder()
> .withEvictedBlocks(evictedBlocks)
> .withMaxCacheSize(blockCache.getMaxSize())
> .build();
> }
> {code}
> BlockCache is the regionserver level object.If we accumulate this value for
> each region, we will get a wrong result.
> The value of maxCacheSize should be set to BlockCache.getMaxSize() after
> processing all the regions on this regionserver.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
