[
https://issues.apache.org/jira/browse/HBASE-3040?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12914914#action_12914914
]
jinglong.liujl commented on HBASE-3040:
---------------------------------------
We can reproduce this issue in our environment.
Our environment is HBase 0.20.6 + Hadoop (CDH3b2). 5 region server and 5
datanode with 2193 region and 48.95 TB data in HDFS. And our client is put 5
column family and 5 qualifier in one row, and commit to HBase every 500 rows.
We start 8 clients in one machine (have 6 machines, totally we have 48 clients,
). block size is 64K, region size is 256 M, one cell has 100B random string.
each machine has 8 core CPU, 48G memory, 12 1T disks.
In region split , HMaster cosider the region has been load. But region
server has not been completed for loading.
From client, it can not see the daughter region . After several retrys,
exception will be raised. exception like below
org.apache.hadoop.hbase.client.NoServerForRegionException: No server
address listed in .META. for region
> BlockIndex readIndex too slowly in heavy write scenario
> -------------------------------------------------------
>
> Key: HBASE-3040
> URL: https://issues.apache.org/jira/browse/HBASE-3040
> Project: HBase
> Issue Type: Improvement
> Components: regionserver
> Affects Versions: 0.20.6
> Environment: 1master, 7 region servers, 4 * 7 clients(all clients run
> on region server host), sequential put
> Reporter: andychen
>
> region size is configured with 128M, block size is 64K, the table has 5
> column families
> at the beginning, when region split, master assigns daughters to new region
> servers, new region server open region, readIndex of this region's
> storefile(about 1000 blocks) spent 30~50ms, with the data import region
> server spent more and more time (sometimes up to several seconds) to load
> 1000 block indices
> at right now, we resolve this issue by getting all indices of one hfile
> within one DFS read instead of 1000 reads.
> is there any other better resolution?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
