I'd check the ZK logs, because it looks to me that you may have lost data in ZK related to coordinating log recovery, due to the full disk. Was the server that filled up a ZK node?
-- Christopher L Tubbs II http://gravatar.com/ctubbsii On Tue, Apr 30, 2013 at 10:28 AM, Chris Sigman <[email protected]> wrote: > Sorry, Accumulo 1.4.2, Hadoop 0.20.2, zoo 3.3.5 > > > -- > Chris > > > On Tue, Apr 30, 2013 at 10:19 AM, Josh Elser <[email protected]> wrote: >> >> Versions of Accumulo, ZK, and Hadoop? >> >> >> On Tuesday, April 30, 2013, Chris Sigman wrote: >>> >>> I've got a small cluster with 2 tablet/logger servers, and I was running >>> a job that filled up the free space on the root partition of one of the >>> servers with a bunch of junk logs. I fixed the issue creating all the junk, >>> but when I tried to restart everything, the other server is reporting the >>> following error message when starting the logger: >>> >>> Error checking on job >>> java.lang.IllegalArgumentException: Path cannot be null >>> at org.apache.zookeeper.common.PathUtils.validatePath(PathUtils.java:45) >>> at org.apache.zookeeper.ZooKeeper.exists(ZooKeeper.java:794) >>> at org.apache.accumulo.core.zookeeper.ZooCache$2.run(ZooCache.java:208) >>> at org.apache.accumulo.core.zookeeper.ZooCache.retry(ZooCache.java:130) >>> at org.apache.accumulo.core.zookeeper.ZooCache.get(ZooCache.java:233) >>> at org.apache.accumulo.core.zookeeper.ZooCache.get(ZooCache.java:188) >>> at >>> org.apache.accumulo.server.master.CoordinateRecoveryTask$RecoveryJob.isComplete(CoordinateRecoveryTask.java:172) >>> at >>> org.apache.accumulo.server.master.CoordinateRecoveryTask.run(CoordinateRecoveryTask.java:369) >>> at >>> org.apache.accumulo.core.util.LoggingRunnable.run(LoggingRunnable.java:34) >>> at java.lang.Thread.run(Thread.java:722) >>> >>> -- >>> Chris > >
