Hi, 
When we use bulkload to load hfiles to hbase(not mapreduce), regionserver will 
call 
org.apache.hadoop.hbase.regionserver.SecureBulkLoadManager.secureBulkLoadHFiles 
-> org.apache.hadoop.hbase.regionserver.bulkLoadHFiles, secureBulkLoadHFiles 
calls bulkLoadHFiles with 
```
return region.bulkLoadHFiles(familyPaths, true,
new SecureBulkLoadListener(fs, bulkToken, conf), request.getCopyFile(),
clusterIds, request.getReplicate());
```
secureBulkLoadHFiles passes true to assignSeqId:boolean in bulkLoadHFiles , 
this will cause a flush before bulkload. If we have many bulkload, it will 
generate many small hfiles. Can we get the assignSeqId from config?
| |
xicm000
|
|
[email protected]
|
签名由网易邮箱大师定制

Reply via email to