[HDFS-2026] 1073: 2NN needs to handle case of reformatted NN better

• Type: Sub-task
• Status: Resolved
• Priority: Critical
• Resolution: Fixed
• Affects Version/s: Edit log branch (HDFS-1073)
• Fix Version/s: Edit log branch (HDFS-1073)
• Component/s: namenode
• Hadoop Flags: Reviewed
• Assignee: Todd Lipcon
• Reporter: Todd Lipcon
• Created: 02/Jun/11 22:46
• Updated: 23/Jun/11 23:10
• Resolved: 23/Jun/11 23:10

Description

Currently in the 1073 branch, the following steps ends up with a very confused 2NN:

• format NN, run NN
• start 2NN, perform some checkpoints
• reformat NN, start NN on new namespace
• restart same 2NN

The 2NN currently saves the new VERSION info into its local storage directory but doesn't clear out the old checkpoint or edits files. This is obviously wrong and might lead to a corrupt checkpoint getting uploaded.

If the 2NN has storage directories with VERSION info, and connects to an NN with different VERSION info, there are two alternatives:
a) refuse to perform any checkpoints until the operator issues a "secondarynamenode -format" command (this is similar to how the backupnode/checkpointnode works)
b) clear the current contents of the storage directory and save the new NN's VERSION info.

Issue Links

• relates to: HDFS-2104 1073: Add a flag to 2NN to format its checkpoint dirs on startup