You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flume.apache.org by "Jack Jiang (Jira)" <ji...@apache.org> on 2020/06/19 09:17:00 UTC
[jira] [Updated] (FLUME-2880) how to config a custom hbase sink?
[ https://issues.apache.org/jira/browse/FLUME-2880?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jack Jiang updated FLUME-2880:
------------------------------
Description:
my flume version is 1.6.0 and i'm ready to sink logs to hbase,before this action,i need to process logs.so i defined a class named AsyncHbaseLogEventSerializer,it implements the interface of AsyncHbaseEventSerializer and i overwrite the method inside.here is my sink part of configuration file:
server.sinks.hdfs_sink.type = asynchbase
server.sinks.hdfs_sink.table = access_log
server.sinks.hdfs_sink.columnFamily = cb
server.sinks.hdfs_sink.serializer = com.xxxx.flume.sink.hbase.AsyncHbaseLogEventSerializer
server.sinks.hdfs_sink.serializer.columns = host_name,remote_host,remote_user,event_ts,req,req_status,resp_bytes,ref,agent
my question is should i package my own class as a jar package,and put it to lib?
was:
my flume version is 1.6.0 and i'm ready to sink logs to hbase,before this action,i need to process logs.so i defined a class named AsyncHbaseLogEventSerializer,it implements the interface of AsyncHbaseEventSerializer and i overwrite the method inside.here is my sink part of configuration file:
server.sinks.hdfs_sink.type = asynchbase
server.sinks.hdfs_sink.table = access_log
server.sinks.hdfs_sink.columnFamily = cb
server.sinks.hdfs_sink.serializer = com.huateng.flume.sink.hbase.AsyncHbaseLogEventSerializer
server.sinks.hdfs_sink.serializer.columns = host_name,remote_host,remote_user,event_ts,req,req_status,resp_bytes,ref,agent
my question is should i package my own class as a jar package,and put it to lib?
> how to config a custom hbase sink?
> ----------------------------------
>
> Key: FLUME-2880
> URL: https://issues.apache.org/jira/browse/FLUME-2880
> Project: Flume
> Issue Type: Question
> Reporter: Jack Jiang
> Priority: Major
> Fix For: notrack
>
>
> my flume version is 1.6.0 and i'm ready to sink logs to hbase,before this action,i need to process logs.so i defined a class named AsyncHbaseLogEventSerializer,it implements the interface of AsyncHbaseEventSerializer and i overwrite the method inside.here is my sink part of configuration file:
> server.sinks.hdfs_sink.type = asynchbase
> server.sinks.hdfs_sink.table = access_log
> server.sinks.hdfs_sink.columnFamily = cb
> server.sinks.hdfs_sink.serializer = com.xxxx.flume.sink.hbase.AsyncHbaseLogEventSerializer
> server.sinks.hdfs_sink.serializer.columns = host_name,remote_host,remote_user,event_ts,req,req_status,resp_bytes,ref,agent
> my question is should i package my own class as a jar package,and put it to lib?
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@flume.apache.org
For additional commands, e-mail: issues-help@flume.apache.org