Hi guys, Now I'm able to generate "HDFSAuditLogObject" based on MAPR's hdfs audit log message, but met a problem of the "userID" part. Like I mentioned before, mapR's audit log message contains ids( eg: "2050.36.131336" is the file id of "/tmp/file.w" , "5098" is the id of user "daniel") but not human readable names. Now by using class "MapRCliCommands", I'm able to convert between fileIDs and its names. As user id is more related to local machine, to get its username, I must get the access to that machine's "/etc/passwd". My questions is:
1. Is there a better way than this kind of time-consuming query? 2. Is it a requirement that the cluster nodes should have a consistent set of known users? Thanks and regards, :) Daniel
