asmcmd fails with CLSD: An error occurred while attempting to generate a full name

This a quick tip for a very simple, but confusing issue I faced a few days in an Oracle Clusterware. This occurred in a 6 node cluster, where the nodes 1-3 and node 6 was working perfectly, but nodes 4 and 5 was failing with this error CLSU-00101 when I tried to use asmcmd.

Problem

The error occurred while trying to open the asmcmd utility in specific nodes:

[grid@c03db04 ~]$ asmcmd
2026-06-19 15:15:35.967 
CLSD: An error occurred while attempting to generate a full name. Logging may not be active for this process
Additional diagnostics: CLSU-00107: operating system function: sclsdgcwd; failed with error data: 0; at location: sclsdgcwd13
CLSU-00101: operating system error message: Error 0
(:CLSD00183:)
ASMCMD>

When I was checking the log directory under the Grid Home, I found the folder with hostname was owned by oracle instead of grid user:

[grid@c03db04 ~]$ ls -l $ORACLE_HOME/log
total 0
drwxr----T 3 oracle oinstall 20 Mar 16 15:02 c03db04
drwxr-x--- 3 grid   oinstall 40 Mar 16 14:05 diag
[grid@c03db04 ~]$ 

Solution

So I just connected as root and changed the owner for this folder:

[grid@c03db04 ~]$ su - 
Password:
[root@c03db04 ~]# chown -R grid.oinstall /u01/app/23.0.0.0/grid_3/log/c03db04
[root@c03db04 ~]# 
[root@c03db04 ~]# exit
logout
[grid@c03db04 ~]$ 
[grid@c03db04 ~]$ 
[grid@c03db04 ~]$ asmcmd 
ASMCMD> 
ASMCMD> 

This is the kind of issue caused by human error, and it was probably caused by myself during some tests with the Grid Home in some Out-Of-Place testing.