the Disk Sequence in the SUN Cluster is Different from the Disk Sequence before the Cutover

Publication Date:  2012-07-18 Views:  279 Downloads:  0
Issue Description
1.       Before the VIS takes over the disk array, check the hard disk in Solaris by running the format command.
# format
Searching for disks...done
AVAILABLE DISK SELECTIONS:
       0. c1t0d0
          /pci@1f,700000/scsi@2/sd@0,0
       1. c1t1d0
          /pci@1f,700000/scsi@2/sd@1,0
       2. c1t2d0
          /pci@1f,700000/scsi@2/sd@2,0
       3. c1t3d0
          /pci@1f,700000/scsi@2/sd@3,0
       4. c3t40d0 (SUN-StorEdge3510-415F cyl 65533 alt 2 hd 64 sec 292)
          /pci@1c,600000/SUNW,qlc@1/fp@0,0/ssd@w216000c0ff8a7498,0
       5. c3t40d1 (SUN-StorEdge3510-415F cyl 47341 alt 2 hd 64 sec 255)
          /pci@1c,600000/SUNW,qlc@1/fp@0,0/ssd@w216000c0ff8a7498,1
       6. c3t40d2 (SUN-StorEdge3510-415F cyl 998 alt 2 hd 64 sec 32)
          /pci@1c,600000/SUNW,qlc@1/fp@0,0/ssd@w216000c0ff8a7498,2

The capacity of disk 4, disk 5, and disk 6 is 583 GB, 325 GB, and 1 GB respectively.
Check DID in the SUN cluster.
# scdidadm  -l
1        DB-1:/dev/rdsk/c0t0d0     /dev/did/rdsk/d1
2        DB-1:/dev/rdsk/c1t0d0     /dev/did/rdsk/d2
3        DB-1:/dev/rdsk/c1t1d0     /dev/did/rdsk/d3
4        DB-1:/dev/rdsk/c1t2d0     /dev/did/rdsk/d4
5        DB-1:/dev/rdsk/c1t3d0     /dev/did/rdsk/d5
6        DB-1:/dev/rdsk/c3t40d2    /dev/did/rdsk/d6
7        DB-1:/dev/rdsk/c3t40d1    /dev/did/rdsk/d7
8        DB-1:/dev/rdsk/c3t40d0    /dev/did/rdsk/d8
1.        After the VIS takes over the disk array, check the hard disk by running the format command.
# format
Searching for disks...done
AVAILABLE DISK SELECTIONS:
       0. c1t0d0
          /pci@1f,700000/scsi@2/sd@0,0
       1. c1t1d0
          /pci@1f,700000/scsi@2/sd@1,0
       2. c1t2d0
          /pci@1f,700000/scsi@2/sd@2,0
       3. c1t3d0
          /pci@1f,700000/scsi@2/sd@3,0
       4. c3t22040022A100D40Ad0 (SUN-StorEdge3510-415F cyl 998 alt 2 hd 64 sec 32)
          /pci@1c,600000/SUNW,qlc@1/fp@0,0/ssd@w22040022a100d40a,2      
       5. c3t22040022A100D40Ad1 (SUN-StorEdge3510-415F cyl 47341 alt 2 hd 64 sec 255)
          /pci@1c,600000/SUNW,qlc@1/fp@0,0/ssd@w22040022a100d40a,1
       6. c3t22040022A100D40Ad2 (SUN-StorEdge3510-415F cyl 65533 alt 2 hd 64 sec 292)
          /pci@1c,600000/SUNW,qlc@1/fp@0,0/ssd@w22040022a100d40a,0

The sequence of disk 4, disk 5, and disk 6 is different from that before the cut over. The capacity of disk 4, disk 5, and disk 6 is 1 GB, 325 GB, and 583 GB respectively.
Check DID in the SUN cluster.
scdidadm -l
1        DB-1:/dev/rdsk/c0t0d0     /dev/did/rdsk/d1    
2        DB-1:/dev/rdsk/c1t0d0     /dev/did/rdsk/d2    
3        DB-1:/dev/rdsk/c1t1d0     /dev/did/rdsk/d3    
4        DB-1:/dev/rdsk/c1t2d0     /dev/did/rdsk/d4    
5        DB-1:/dev/rdsk/c1t3d0     /dev/did/rdsk/d5    
6        DB-1:/dev/rdsk/c4t22010022A100D40Ad0 /dev/did/rdsk/d6    
6        DB-1:/dev/rdsk/c4t22110022A100D40Ad0 /dev/did/rdsk/d6    
7        DB-1:/dev/rdsk/c4t22010022A100D40Ad1 /dev/did/rdsk/d7    
7        DB-1:/dev/rdsk/c4t22110022A100D40Ad1 /dev/did/rdsk/d7    
8        DB-1:/dev/rdsk/c4t22010022A100D40Ad2 /dev/did/rdsk/d8    
8        DB-1:/dev/rdsk/c4t22110022A100D40Ad2 /dev/did/rdsk/d8    
The corresponding hard disks of D6, D7, and D8 are different from those corresponding hard disks before the cutover. The SUN host accesses DID. You need to modify other configuration files of the SUN host.

Alarm Information
None
Handling Process
Re-map LUNs to the SUN host on the VIS. The sequence of mapping volume ID remains the same as the sequence of volume mapped by the disk array to the host before the cutover.
The mapping volume of the 583 GB disk is ID 0, the mapping volume of the 325 GB disk is ID 1, and the mapping volume of the 1 GB disk is ID 2.
After the remapping, run the format command on the SUN host to check the hard disk.
# format
Searching for disks...done
AVAILABLE DISK SELECTIONS:
       0. c1t0d0
          /pci@1f,700000/scsi@2/sd@0,0
       1. c1t1d0
          /pci@1f,700000/scsi@2/sd@1,0
       2. c1t2d0
          /pci@1f,700000/scsi@2/sd@2,0
       3. c1t3d0
          /pci@1f,700000/scsi@2/sd@3,0
       4. c3t22040022A100D40Ad0 (SUN-StorEdge3510-415F cyl 65533 alt 2 hd 64 sec 292)
          /pci@1c,600000/SUNW,qlc@1/fp@0,0/ssd@w22040022a100d40a,0
       5. c3t22040022A100D40Ad1 (SUN-StorEdge3510-415F cyl 47341 alt 2 hd 64 sec 255)
          /pci@1c,600000/SUNW,qlc@1/fp@0,0/ssd@w22040022a100d40a,1
       6. c3t22040022A100D40Ad2 (SUN-StorEdge3510-415F cyl 998 alt 2 hd 64 sec 32)
          /pci@1c,600000/SUNW,qlc@1/fp@0,0/ssd@w22040022a100d40a,2

The sequence of disk 4, disk 5, and disk 6 is the same as the sequence before the cutover.
Check DID in the SUN cluster.
#scdidadm -l
1        DB-1:/dev/rdsk/c0t0d0     /dev/did/rdsk/d1    
2        DB-1:/dev/rdsk/c1t0d0     /dev/did/rdsk/d2    
3        DB-1:/dev/rdsk/c1t1d0     /dev/did/rdsk/d3    
4        DB-1:/dev/rdsk/c1t2d0     /dev/did/rdsk/d4    
5        DB-1:/dev/rdsk/c1t3d0     /dev/did/rdsk/d5    
6        DB-1:/dev/rdsk/c4t22010022A100D40Ad2 /dev/did/rdsk/d6    
6        DB-1:/dev/rdsk/c4t22110022A100D40Ad2 /dev/did/rdsk/d6    
7        DB-1:/dev/rdsk/c4t22010022A100D40Ad1 /dev/did/rdsk/d7    
7        DB-1:/dev/rdsk/c4t22110022A100D40Ad1 /dev/did/rdsk/d7    
8        DB-1:/dev/rdsk/c4t22010022A100D40Ad0 /dev/did/rdsk/d8    
8        DB-1:/dev/rdsk/c4t22110022A100D40Ad0 /dev/did/rdsk/d8    
The corresponding hard disks of D6, D7, and D8 do not change. No follow-up operation is required on the SUN host.

Root Cause
After the VIS takes over the disk array, the sequence of the host mapping volume ID must be the same as the sequence of the volume mapped by the disk array to the host before the VIS takeover.
Suggestions
Before the VIS takes over the disk array, check the sequence of the LUN mapped by the disk array to the host.
After the VIS takes over the disk array, the sequence of the mapping volume ID on the host is same as the sequence of mapping volume ID mapped by the disk array to the host before the VIS takeover.

END