Since compression algorithms such as LZO need to be installed and configured on each cluster this is a frequent source of startup error. If you see messages like this...
11/02/20 01:32:15 ERROR lzo.GPLNativeCodeLoader: Could not load native gpl library java.lang.UnsatisfiedLinkError: no gplcompression in java.library.path at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1734) at java.lang.Runtime.loadLibrary0(Runtime.java:823) at java.lang.System.loadLibrary(System.java:1028)
.. then there is a path issue with the compression libraries. See the Configuration section on LZO compression configuration.
See the Getting Started section on ulimit and nproc configuration.
This typically shows up in the DataNode logs.
See the Getting Started section on xceivers configuration.
See the Getting Started section on ulimit and nproc configuration.
If you see warning messages like this...
2009-02-24 10:01:33,516 WARN org.apache.hadoop.hbase.util.Sleeper: We slept xxx ms, ten times longer than scheduled: 10000 2009-02-24 10:01:33,516 WARN org.apache.hadoop.hbase.util.Sleeper: We slept xxx ms, ten times longer than scheduled: 15000 2009-02-24 10:01:36,472 WARN org.apache.hadoop.hbase.regionserver.HRegionServer: unable to report to master for xxx milliseconds - retrying
... or see full GC compactions then you may be experiencing full GC's.
These errors can happen either when running out of OS file handles or in periods of severe network problems where the nodes are unreachable.
See the Getting Started section on ulimit and nproc configuration and check your network.