Dell Unity: How to Restart or Failover the Management Services (User Correctable)

Sometimes information is not updated correctly in Unisphere, or the UI may not be fully synchronized. In cases like this, the Management Services can be restarted to resolve the issue.
Sometimes it is necessary to have the Management Services failed over to the other SP for a full poll. This refreshes the data and the UI.

Steps to restart Management Services from the UI (Unisphere):

1. Go to Service > Service Tasks.
2. Click Restart Management Software.
3. Click Execute and OK. It can take up to 10 minutes for Management Services to restore completely.

Steps to restart Management Services from Unisphere CLI (UEMCLI):

1. Open a system where the UEMCLI tool is installed and run the command:

uemcli.exe -d <management IP> -u <user> -p <password> /service/system restart

Note: If 'admin' user does not work for the above command, restart can be attempted using the 'service' user credentials. 

Steps to restart Management Services from CLI (SSH):

1. Log in to the array over SSH (username: service)
2. Run the command:

svc_restart_service restart MGMT

Steps to failover Management Services from CLI (SSH):

1. Log in to the array over SSH (username: service)
2. Run the command:           

svc_restart_service failover MGMT

This is not disruptive to the array's data path (see notes section below). However, restarting the management software can temporarily disrupt management activities such as provisioning storage, configuration of networking, and other activities.
See the Additional info below for other details.

Restarting the management software when both Storage Processors are in Service Mode is not possible.

ATTENTION!
Before restarting the service, ensure that there are no Storage Processors space issues that may prevent the Storage Processors from starting the services again.

Although rare, if there is no space that is left on specific partitions, the Storage Processors (both) may go into rolling reboots ending in service mode. This may cause data unavailability. This is especially true in code levels earlier than 4.3.

Before any restart of the Management Services, run the following command on the primary Storage Processors to ensure that these partitions are not 100% full.

# df -h
Filesystem         Size  Used Avail Use% Mounted on
...
/dev/c4nasdba1       1013M   71M  891M   8% /nbsnas
/dev/c4nasdba2        3.5G  796M  2.6G  24% /EMC/backend/CEM
/dev/c4loga1          3.4G  1.5G  1.7G  48% /EMC/backend/log_shared

If the partitions are full or close to full, stop immediately and contact Dell Technical Support or your Authorized Service Representative. In the above example, there are no space issues.

NOTE: If a failover of Management Services is to be done, then the peer SP space should also be checked before running the failover command.