Demystifying Hadoop concepts Series: Safe mode

 

What is is safe mode of hadoop, may time we come across this exception “ org.apache.hadoop.ipc.RemoteException: org.apache.hadoop.hdfs.server.namenode.SafeModeException”  or some other exceptions Which contains safe mode in it .

 

First let me tell what Safe mode is in context to Hadoop : as we all know Name node contains fsimage (metadata) of the data present on the cluster, which can be large or small based on the size of the cluster and the size of date present on the cluster, so when the name node starts it loads this fsimage and the edit logs from the disk in the Primary memory RAM for fast processing, and after loading it waits for data nodes to report about the present on those data nodes, so during this process that is loading the fsimage and edit logs and waiting for data nodes to report about the data block in safe mode, which is a read only mode for name node this is done to maintain the consistency of the data present, this is just like saying “ i will not receive any thing till i know what i already have”. And during this period no modification to the file blocks are allowed as to maintain the correctness of the data.

 

How long safemode exist :

Generally name node automatically comes out of safe mode in 30 seconds if all data present are consistent according the fsimage and the editlogs.

 

Related commands :

Put Namenode in Safemode: bin/hadoop dfsadmin –safemode

Leave Safemode : bin/hadoop dfsadmin -safemode leave

 

What to do if you encounter this exception :

 

First, wait a minute or two and then retry your command. If you just started your cluster, it's possible that it isn't fully initialized yet. If waiting a few minutes didn't help and you still get a "safe mode" error, check your logs to see if any of your data nodes didn't start correctly (either they have Java exceptions in their logs or they have messages stating that they are unable to contact some other node in your cluster). If this is the case you need to resolve the configuration issue (or possibly pick some new nodes) before you can continue.