You hitting an instance of https://issues.apache.org/jira/browse/AMBARI-15482
I don't know of a way around this aside from: - Finalizing the upgrade - Starting NameNode manually from the command prompt It's probably best to just finalize the upgrade and start NameNode from the web client after finalization. On May 18, 2016, at 10:02 PM, Anandha L Ranganathan <analog.s...@gmail.com<mailto:analog.s...@gmail.com>> wrote: I am running rolling upgrade in dev cluster . It is completed with 100% but not yet finalized. I was testing int he dev cluster and validating is everything working fine. I was able to run the Hive Query using HS2 server. I don't remember what is the reason but I restarted all namenode services through Ambari UI and I started getting this error. It says run with --upgrade option. I thorugh rolling upgrade would take care of it. Please help me how do I handle this ? What are the steps should I do ? 2016-05-19 01:42:38,561 INFO util.GSet (LightWeightGSet.java:computeCapacity(356)) - 0.029999999329447746% max memory 1011.3 MB = 310.7 KB 2016-05-19 01:42:38,561 INFO util.GSet (LightWeightGSet.java:computeCapacity(361)) - capacity = 2^15 = 32768 entries 2016-05-19 01:42:38,579 INFO common.Storage (Storage.java:tryLock(715)) - Lock on /mnt/data/hadoop/hdfs/namenode/in_use.lock acquired by nodename 13159@usw2dxdpma01.glassdoor.local<mailto:13159@usw2dxdpma01.glassdoor.local> 2016-05-19 01:42:38,651 WARN namenode.FSNamesystem (FSNamesystem.java:loadFromDisk(690)) - Encountered exception loading fsimage java.io.IOException: File system image contains an old layout version -60. An upgrade to version -63 is required. Please restart NameNode with the "-rollingUpgrade started" option if a rolling upgrade is already started; or restart NameNode with the "-upgrade" option to start a new upgrade. at org.apache.hadoop.hdfs.server.namenode.FSImage.recoverTransitionRead(FSImage.java:245) at org.apache.hadoop.hdfs.server.namenode.FSNamesystem.loadFSImage(FSNamesystem.java:983) at org.apache.hadoop.hdfs.server.namenode.FSNamesystem.loadFromDisk(FSNamesystem.java:688) at org.apache.hadoop.hdfs.server.namenode.NameNode.loadNamesystem(NameNode.java:662) at org.apache.hadoop.hdfs.server.namenode.NameNode.initialize(NameNode.java:722) at org.apache.hadoop.hdfs.server.namenode.NameNode.<init>(NameNode.java:951) at org.apache.hadoop.hdfs.server.namenode.NameNode.<init>(NameNode.java:935) at org.apache.hadoop.hdfs.server.namenode.NameNode.createNameNode(NameNode.java:1641) at org.apache.hadoop.hdfs.server.namenode.NameNode.main(NameNode.java:1707) 2016-05-19 01:42:38,661 INFO mortbay.log (Slf4jLog.java:info(67)) - Stopped HttpServer2$SelectChannelConnectorWithSafeStartup@usw2dxdpma01.glassdoor.local<mailto:HttpServer2$SelectChannelConnectorWithSafeStartup@usw2dxdpma01.glassdoor.local>:50070 2016-05-19 01:42:38,663 INFO impl.MetricsSystemImpl (MetricsSystemImpl.java:stop(211)) - Stopping NameNode metrics system... 2016-05-19 01:42:38,664 INFO impl.MetricsSinkAdapter (MetricsSinkAdapter.java:publishMetricsFromQueue(141)) - timeline thread interrupted. 2016-05-19 01:42:38,664 INFO impl.MetricsSystemImpl (MetricsSystemImpl.java:stop(217)) - NameNode metrics system stopped. 2016-05-19 01:42:38,664 INFO impl.MetricsSystemImpl (MetricsSystemImpl.java:shutdown(607)) - NameNode metrics system shutdown complete. 2016-05-19 01:42:38,665 ERROR namenode.NameNode (NameNode.java:main(1712)) - Failed to start namenode.
