Reference Information › CA AppLogic Support Knowledge Base › Overview of Support Knowledge Base › Steps for Proper Shutdown and Startup of Controller Before and After Maintenance

Steps for Proper Shutdown and Startup of Controller Before and After Maintenance

Follow these steps:

1. Check the non-primary nodes by the command :

   3t srv list
   

2. Go to the non primary nodes by ssh from the aldo or the bfc
3. "run "service heartbeat stop" on both of them one by one .
4. go to the primary node and run xm list to check the controller has gone down or not.
5. If the controller has gone down perform the maintainance tasks.
6. Come back to the aldo and run :

   ssh <non-primary-node1> 'cat /etc/applogic.d/ha_role'
   ssh <non-primary-node2> 'cat /etc/applogic.d/ha_role'
   

   both of them should return ' secondary '

7. Run ssh <primary-node> 'cat /etc/applogic.d/ha_role'. It should return 'primary'
8. run

   ssh <non-primary-node1> '3tsrv bd list --all'
   ssh <non-primary-node2> '3tsrv bd list --all'
   ssh <primary-node> '3tsrv bd list --all'
   

   There should be no stuck md or nbd or hoop devices.

9. if steps 7 and 8 are fine the run on the primary node

   '3tsrv set role=primary --recover' 
   

   The controller should come up.