Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
Flume >> mail # user >> fileChannel always reached it's capacity


Copy link to this message
-
fileChannel always reached it's capacity
I use hdfsSink and fileChannel, it seems hdfsSink consume too slow.here is
my configuration:
...
agent1.channels.thrift_ch2.capacity = 200000000
agent1.channels.thrift_ch2.transactionCapacity = 100000
...
agent1.sinks.hdfsSink.hdfs.batchSize = 100000
agent1.sinks.hdfsSink.hdfs.threadsPoolSize = 40

very appreciate for guiding !!!