Typically, the spare part runs a certain version of S2600 R1 such as 1.04.01.212.T08 (the latest version at the delivery), which may differ from the onsite software version. Do as follows to recover this error.
If there is another S2600 that runs properly in the field and has the same version as the faulty device, perform the following steps:
1. Remove one controller from the functional S2600 without interrupting services. Insert a faulty controller. The system will automatically recover the faulty controller.
2. The recovery process lasts 10 to 20 minutes. You can connect the faulty controller to a serial port to monitor the recovery process. If the storage array finally recovers to the normal dual-controller state, the faulty controller is recovered successfully.
3. The other faulty controller can be recovered by the same method. After both controllers are recovered, put them back to the original device and power it on. If the problem persists, contact technical support engineers.
If there is no other S2600 that has the same version as the faulty one, or there is only one S2600 in the field, perform the following steps:
Step 1 Apply for a new S2600 controller and two S2600 disks.
Step 2 Obtain the software version of the faulty device. On the CLI, run the showctrlinfo command to query the version, whose general format is 1.04.0x.xxx.Txx such as 1.04.01.212.T08. You may also obtain the software version of the device by viewing historical processing data, delivery documentation, and ISM logs or consulting with maintenance engineers. Obtaining the software version is mandatory. Otherwise, you cannot carry on with the following steps.
Step 3 After the device software version is obtained. Download the correct upgrade package corresponding to the version from the support website.
Step 4 After both the spare part and upgrade package are delivered to the field, you can start to recover the faulty device.
Step 5 Power off the faulty device and remove the following components:
Cascading cables of the controller enclosure (if there is any)
All disks of the controller enclosure. Do not pull disks out but only make them disconnected from the backplane.
Controller A and controller B (one controller for the single-controller configuration)
Step 6 Remove the coffer disks from slots (0,0) and (0,2). Record the mapping relationship between the slots and disk IDs for putting back the disks in the future.
Step 7 Insert the two spare disks into slots (0,0) and (0,2) and the spare controller into the slot of controller A. Keep the device in the single-controller state and power it on.
Step 8 Log in to the system from the serial port. On the CLI, run the showctrlinfo command to view the current system version. If it is inconsistent with the original system version, use the upgrade package obtained in step 2 to upgrade the controller to the version the same as the original device.
Step 9 Insert a faulty controller into another controller slot to automatically recover it. The recovery process lasts 10 to 20 minutes. Then, remove the recovered controller and insert the other faulty controller.
Step 10 After both controllers are recovered, power off the storage array. Put back the two coffer disks removed in step 6. Push back disks removed in step 5. Then, insert the two recovered controllers.
To use the spare controller, you need to install the BBU and optical module of the original controller to the spare controller.
Step 11 Power on the system. After the system runs normal, check whether the configurations such as RAID, LUN, and mapping are correct and start to recover services.