In my last screencast I have briefly (just over 6 minutes) explain what are and how to access clustershell, nodeshell, and systemshell.
If you rather read an article about those, some time ago I have also published an entry NetApp cDOT – Clustered Data ONTAP Command Line
What you have to remember about those 3 shells are:
-
- default shell is a clustershell – that’s the one you access when you log in via ssh to Your cluster (via cluster management IP address, or one of the nodes). This shell is used to work with Your cluster, this CLI is used to create and manage SVM (Storage Virtual Machines), this shell is intended for everyday operations.
- a nodeshell gives you a separate CLI to manage and monitor a single node within a cluster. If you want to access a nodeshell, you have to specify to which node you wish to connect to. There are different commands to work with a nodeshell, those are very similar to old-type Data ONTAP, called 7-mode. You would use a nodeshell to monitor Your node performance and utilization, to verify the connectivity of shelves and drives. But all provisioning tasks should be done from the clustershell. To access a nodeshell, you have to first access a clustershell.
- a systemshell is a low-level shell, which you shouldn’t access unless you have to. You can only access it from clustershell, and you have to unlock (and know the password) of build-in user diag, you access a systemshell only from the diagnostic privilege level. In my video I gave you example, that nodeshell can be used to review the audit logs which is true, however there are more secure ways to review the audit logs, which does not require you to enter a systemshell. I will go through this topic in a separate entry, but long story short You can access it via https protocol (provided that you have web logs enabled) :
https://cluster-mgmt-ip/spi/node-name/etc/log/
As always, please let me know if You have any comments or questions regarding the video.