[HDFS-6654] Setting Extended ACLs recursively for another user belonging to the same group is not working

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Not A Problem
• Affects Version/s: 2.4.1
• Assignee: Unassigned
• Reporter: J.Andreina
• Created: 10/Jul/14 06:48
• Updated: 11/Jul/14 02:07
• Resolved: 11/Jul/14 02:07

Description

1.Setting Extended ACL recursively for  a user belonging to the same group  is not working

Step 1: Created a Dir1 with User1
./hdfs dfs -rm -R /Dir1
Step 2: Changed the permission (600) for Dir1 recursively
./hdfs dfs -chmod -R 600 /Dir1
Step 3: setfacls is executed to give read and write permissions to User2 which belongs to the same group as User1
./hdfs dfs setfacl -R -m user:User2:rw /Dir1

./hdfs dfs -getfacl -R /Dir1
No GC_PROFILE is given. Defaults to medium.

1. file: /Dir1
2. owner: User1
3. group: supergroup
   user::rw-
   user:User2:rw-
   group::---
   mask::rw-
   other::---
   Step 4: Now unable to write a File to Dir1 from User2

./hdfs dfs -put hadoop /Dir1/1
No GC_PROFILE is given. Defaults to medium.
put: Permission denied: user=User2, access=EXECUTE, inode="/Dir1":User1:supergroup:drw------

   2. Fetching filesystem name , when one of the disk configured for NN dir becomes full returns a value "null".

2014-07-08 09:23:43,020 WARN org.apache.hadoop.hdfs.server.namenode.NameNodeResourceChecker: Space available on volume 'null' is 101060608, which is below the configured reserved amount 104857600
2014-07-08 09:23:43,020 WARN org.apache.hadoop.hdfs.server.namenode.FSNamesystem: NameNode low on available disk space. Already in safe mode.
2014-07-08 09:23:43,166 WARN org.apache.hadoop.hdfs.server.namenode.NameNodeResourceChecker: Space available on volume 'null' is 101060608, which is below the configured reserved amount 104857600