Removing failed storage ISCSI connections from the cache engines.

Overview:

The following article describes the procedure for disconnecting a Faulty Storage Device.

Step by step guide:

1. As soon as the storage enclosure failure is detected, please disconnect the iSCSI from the faulty enclosure to restore the system to nominal operation.
   • Stop the pang service using the CLI. :
     • SSH to the management IP of the server and enter padmin credentials.
     • Run su - and enter the root password to get access.
     • Type su admin to get access to the CLI console.
     • Type console> show systemid to get the management service tag.
     • Type console> enable and enter the first service tag displayed in the previous step, as for enabling password.
     • Type console# oper service stop to stop the PANG server.
       
   • If the system was under load when the storage enclosure fails, then it is probable that all the cache engine will reboot.
   • If by some chance, any of the cache engines did not reboot manually, you have to do it with the following commands:
     ssh root@ce-x
echo b > /proc/sysrq-trigger
2. After all the cache engines have been rebooted, on every cache engine, run yast with the command: 
   yast
3. From the yast screen, please select the Network Services
   • Select iSCSI 
     
4. Select Discovered Targets 
   
5. Select a node that has a false connection status
   • Choose Delete for each false connection node
     
     Note: Repeat step 5 until there are no more nodes with a connected false status 
     
     
6. Choose Finish and Quit

Confirmation: Inside the Discovered Targets screen, you shouldn't see any false status anymore 

Content Author: Erick Tovar