Sun ZFS Storage 7320
 
Oracle ZFS Storage ZS3-BA
 
Oracle ZFS Storage ZS3-2
 
Oracle ZFS Storage ZS3-4
 
Sun ZFS Storage 7420
 
Oracle ZFS Storage ZS4-4
 

PLA-Support>Sun Systems>Sun_Other>Sun Collections>SN-OTH: DISK-CAP VCAP
 

In this Document

Goal

Solution

References

Oracle Confidential PARTNER - Available to partners (SUN).
Reason: FRU CAP

Applies to:

Goal

 Describe the procedure needed to replace a DE2-24C or DE2-24P chassis, including the possible need to re-program the serial-number of the chassis.

Solution

DISPATCH INSTRUCTIONS

- WHAT SKILLS DOES THE FIELD ENGINEER/ADMINISTRATOR NEED?

                Training and experience with Oracle ZFS Storage ZS3 systems or Oracle ZFS Storage 7x20 systems

   TIME ESTIMATE:  60 minutes

   TASK COMPLEXITY:  2

FIELD ENGINEER/ADMINISTRATOR INSTRUCTIONS:
   PROBLEM OVERVIEW: The disk centerplane in the DS2-24P/C tray has been determined faulty and needs to be replaced. This require the whole tray chassis to be replaced as the centerplance is not considered to be field-replaceable.

   WHAT STATE SHOULD THE SYSTEM BE IN TO BE READY TO PERFORM THE RESOLUTION ACTIVITY?:

   The system needs to be in such a state that the affected drive tray can be powered off, usually this means completely shut down of the NAS head(s). In some cases where the pools have been configured as "No Single Point of Failure", aka NSPF, can the   tray be powered off with services still running. If the tray redundancy provided with the NSPF setting has not been tested and confirmed please shut NAS head(s) down..

   WHAT ACTION DOES THE FIELD ENGINEER/ADMINISTRATOR NEED TO TAKE?:

If customer has not powered the NAS head(s) down please start from Step 1 if clustered and step 2 for a Stand Alone system. If NAS head(s) are down start from step 3:

1. In clustered environments, power off one head first (skip this step in a stand-alone configuration)
   1. This keeps everything simple and means cluster failback can be tested as the final step.
   2. Power off using power-button in BUI, or "maintenance system poweroff" in the CLI
2. On the Head still up verify all resources have been imported after the peer node was shut down.
   1. Check that all pools and interfaces are now owned by this head.
3. Power off remaining components of the appliance (ie, running head and JBODs) 
   1. Power off the head using power-button in BUI, or "maintenance system poweroff" in the CLI
   2. Once the head is off, turn off the affected JBOD that is identified for replacement using the power switches on the back of the array. Let the other JBODs remain powered on.
4. Swap JBOD chassis and re-cable, taking care to check all wiring and making sure all components go in to the same positions in the new chassis.
   
   1. Label cables and disks before removal --- location of hardware and cables is important when powered back on
5. Sequence to Power on the replaced JBOD and the head that was powered off last.
   1. Power up the replaced JBOD first.
   2. Power on head by pressing the small power-button the front or by using "start /SYS" from the SP.
6. Verify pools are visible, and that the new JBOD is properly visible in the BUI/CLI (Maintenance -> hardware
   
   1. If the new JBOD shows a long and strange serial number like a WWN "50xxxxxxxxxxxxxx" e.g. , please attempt step 6.2 first  before planning to replace the chassis. It is not expected that a new chassis from factory would have this issue.
   2. If the new JBOD shows a long and strange serial number "AAxxxxxxxxxxxxxx" e.g "SHM0367301R80PH" (normal format is YYWWFMXXXX), please refer to KM Doc ID 1920586.1 for a reprogramming procedure.
7. For clustered systems, power up the second node.
   1. Power on head by pressing the small power-button on the front or by using "start /SYS" from SP.
   2. Once the second node has rejoined the cluster and you can log in, issue a failback from BUI or CLI on the node that was powered up first. The failback can be issued from the Configuration -> Cluster tab in BUI or by running "configuration cluster failback" in CLI.
   3. Verify that each node now owns the desired resources.
8. If the new JBOD is showing a proper serial-number that differs from the one of the original JBOD, and the system is running on a 2011.1.x version:
   1. You must run the command akfixfru on each head.
   2. Run CLI confirm shell akfixfru  (/usr/lib/ak/tools/akfixfru) to update the JBOD disk FRU with the new chassis serial no.
   3. It is important to perform this after chassis replacement otherwise disk replacements will not work. Refer to bug 15789419 for more information.
      
   4. If there also is a disk failure & needs replacement after chassis replacement then follow the procedure outlined in bug 15789419.

OBTAIN CUSTOMER ACCEPTANCE
   WHAT ACTION DOES THE FIELD ENGINEER/ADMINISTRATOR NEED TO TAKE TO RETURN THE SYSTEM TO AN OPERATIONAL STATE?:

  Steps 5 through 8 above brings system back to an operational state including verification of status.

PARTS NOTE:

DE2-24P chassis part-number: 7043630

DE2-24-C chassis part-number: 7044319

DE2-24P NEBS compliant chassis part-number: 7078863

REFERENCE INFORMATION:

References

This solution has no attachment