A1000 Configuration

I have migrated an A1000 to a new server. While doing so, I reconfigured it from a RAID5 volume to a RAID1. Upon doing a probe-scsi-all, I am now seeing 15 device entries for the A1000. Below is a sniplet from the output of the format command. On controller 2, I have targets 0 thru 15. In RAID Manager, everything seems fine in that there is reference to a single device.

I have now installed Veritas Volume Manager. The multiple references to the A1000 are causing Veritas to generate error messages during boot.

What is the proper way to fix this problem?

Thanks, Greg

6. c2t0d0 <Symbios-StorEDGEA1000-0003 cyl 65533 alt 2 hd 64 sec 84>

/sbus@2,0/QLGC,isp@1,10000/sd@0,0

7. c2t1d0 <Symbios-StorEDGEA1000-0003 cyl 65533 alt 2 hd 64 sec 84>

/sbus@2,0/QLGC,isp@1,10000/sd@1,0

8. c2t2d0 <Symbios-StorEDGEA1000-0003 cyl 65533 alt 2 hd 64 sec 84>

/sbus@2,0/QLGC,isp@1,10000/sd@2,0

9. c2t3d0 <Symbios-StorEDGEA1000-0003 cyl 65533 alt 2 hd 64 sec 84>

/sbus@2,0/QLGC,isp@1,10000/sd@3,0

10. c2t4d0 <Symbios-StorEDGEA1000-0003 cyl 65533 alt 2 hd 64 sec 84>

/sbus@2,0/QLGC,isp@1,10000/sd@4,0

11. c2t5d0 <Symbios-StorEDGEA1000-0003 cyl 65533 alt 2 hd 64 sec 84>

/sbus@2,0/QLGC,isp@1,10000/sd@5,0

12. c2t6d0 <Symbios-StorEDGEA1000-0003 cyl 65533 alt 2 hd 64 sec 84>

/sbus@2,0/QLGC,isp@1,10000/sd@6,0

13. c2t8d0 <Symbios-StorEDGEA1000-0003 cyl 65533 alt 2 hd 64 sec 84>

/sbus@2,0/QLGC,isp@1,10000/sd@8,0

14. c2t9d0 <Symbios-StorEDGEA1000-0003 cyl 65533 alt 2 hd 64 sec 84>

/sbus@2,0/QLGC,isp@1,10000/sd@9,0

15. c2t10d0 <Symbios-StorEDGEA1000-0003 cyl 65533 alt 2 hd 64 sec 84>

/sbus@2,0/QLGC,isp@1,10000/sd@a,0

16. c2t11d0 <Symbios-StorEDGEA1000-0003 cyl 65533 alt 2 hd 64 sec 84>

/sbus@2,0/QLGC,isp@1,10000/sd@b,0

17. c2t12d0 <Symbios-StorEDGEA1000-0003 cyl 65533 alt 2 hd 64 sec 84>

/sbus@2,0/QLGC,isp@1,10000/sd@c,0

18. c2t13d0 <Symbios-StorEDGEA1000-0003 cyl 65533 alt 2 hd 64 sec 84>

/sbus@2,0/QLGC,isp@1,10000/sd@d,0

19. c2t14d0 <Symbios-StorEDGEA1000-0003 cyl 65533 alt 2 hd 64 sec 84>

/sbus@2,0/QLGC,isp@1,10000/sd@e,0

20. c2t15d0 <Symbios-StorEDGEA1000-0003 cyl 65533 alt 2 hd 64 sec 84>

/sbus@2,0/QLGC,isp@1,10000/sd@f,0

[2421 byte] By [tulgkb] at [2007-11-25 23:12:13]
# 1
Sorry Greg I lost the plot with this one. I am not sure how many disks you have or which ones that have been duplicated, but from vxdisklist you should be able to tell the problems and then do a manual removal using the "vxdisk rm <disk access name>" command.
bmacdo at 2007-7-5 18:02:05 > top of Java-index,Storage Forums,Storage General Discussion...
# 2
Sorry again, that is "vxdiskadm" and then option list!Brian :-)
bmacdo at 2007-7-5 18:02:05 > top of Java-index,Storage Forums,Storage General Discussion...
# 3

Thanks for your reply Brian.

I ran vxdiskadm as you suggested. The "list" command only shows the disk that are put under VXVM control. The invalid /dev/dsk and /dev/rdsk references have not been put under VXVM control. I believe the problem to be at the OS device level. It is my understanding that the /kernel/drv/sd.conf file is controlling the creation of the /dev/dsk and /dev/rdsk entries.

Back to reference to the number of disk. On controller 2, there is one A1000 that contains 12 drives. However, the drives are not visable to the OS. As you can see from the format output in my original posting, the OS is listing the A1000 15 times. One for each target on that SCSI channel.

Any ideas?

Thanks, Greg

tulgkb at 2007-7-5 18:02:05 > top of Java-index,Storage Forums,Storage General Discussion...
# 4
I'm assuming you are using Differential Ultra Wide SCSI controller pn 370-2443. Since probe-scsi-all and format are yielding same results you could have a SCSI problem. Do you have a differential terminator on back of the A1000, pn 150-1890?
jds2n at 2007-7-5 18:02:05 > top of Java-index,Storage Forums,Storage General Discussion...
# 5
Correct on both accounts. And I'm using a 530-2384 cable.
tulgkb at 2007-7-5 18:02:05 > top of Java-index,Storage Forums,Storage General Discussion...
# 6

I've tested the HBA, cable and terminator. They all seem fine. I found a doc on SunSolve that indicates this is normal behavior. See: ( http://sunsolve.sun.com/pub-cgi/retrieve.pl?doc=finfodoc%2F22747&zone_32=A1 000%20Boot )

Does anyone know if it is possible to get it out of the OS device trees?

Thanks

tulgkb at 2007-7-5 18:02:05 > top of Java-index,Storage Forums,Storage General Discussion...