Fiber Channel Devices and Alternative Pathing

From: Lonnie Randall Webb (lrw_at_tel.bnsf.com)
Date: 02/25/04

Next message: Chris Hoogendyk: "recovering space in /var for doing recommended patches"

Previous message: Lauri Bettencourt: "metadb Issue"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

To: sunmanagers@sunmanagers.org
Date: Wed, 25 Feb 2004 13:48:30 -0600

I have connected a Sun v480 with dual x6767 HBA's to an IBM ESS F-20
Shark via Inrange FC/9000 Directors and loaded AP 2.3.1 along with any
patches for it I could find. I can see the disk partitions on the san
with a format command:

topslnbap001:~> format
Searching for disks...done

AVAILABLE DISK SELECTIONS:
       0. c1t0d0 <SUN36G cyl 24620 alt 2 hd 27 sec 107>
          /pci@9,600000/SUNW,qlc@2/fp@0,0/ssd@w21000004cfc952c3,0
       1. c1t1d0 <SUN36G cyl 24620 alt 2 hd 27 sec 107>
          /pci@9,600000/SUNW,qlc@2/fp@0,0/ssd@w21000004cfc956b8,0
       2. c2t5005076300C1928Bd0 <IBM-2105F20-.487 cyl 9255 alt 2 hd 30
sec 64>
          /pci@8,700000/SUNW,qlc@2/fp@0,0/ssd@w5005076300c1928b,0
       3. c2t5005076300C1928Bd1 <IBM-2105F20-.487 cyl 9255 alt 2 hd 30
sec 64>
          /pci@8,700000/SUNW,qlc@2/fp@0,0/ssd@w5005076300c1928b,1
       4. c2t5005076300C1928Bd2 <IBM-2105F20-.487 cyl 9255 alt 2 hd 30
sec 64>
          /pci@8,700000/SUNW,qlc@2/fp@0,0/ssd@w5005076300c1928b,2

...

      20. c3t5005076300C5928Bd8 <IBM-2105F20-.487 cyl 9255 alt 2 hd 30
sec 64>
          /pci@8,600000/SUNW,qlc@2/fp@0,0/ssd@w5005076300c5928b,8
      21. c3t5005076300C5928Bd9 <IBM-2105F20-.487 cyl 9255 alt 2 hd 30
sec 64>
          /pci@8,600000/SUNW,qlc@2/fp@0,0/ssd@w5005076300c5928b,9

I have partitioned each as follows:

partition> p
Current partition table (original):
Total disk cylinders available: 9255 + 2 (reserved cylinders)

Part Tag Flag Cylinders Size Blocks
  0 unassigned wm 2 - 3 1.88MB (2/0/0) 3840
  1 unassigned wu 0 0 (0/0/0) 0
  2 backup wu 0 - 9254 8.47GB (9255/0/0) 17769600
  3 unassigned wu 0 0 (0/0/0) 0
  4 unassigned wm 4 - 4 0.94MB (1/0/0) 1920
  5 unassigned wu 0 0 (0/0/0) 0
  6 usr wm 5 - 9254 8.47GB (9250/0/0) 17760000
  7 unassigned wu 0 0 (0/0/0) 0

Intending to use the 4th slice for the ap database. I built the ap
databases as follows:

topslnbap001:~> apdb -c /dev/rdsk/c2t5005076300C1928Bd0s4
topslnbap001:~> apdb -c /dev/rdsk/c2t5005076300C1928Bd1s4
topslnbap001:~> apdb -c /dev/rdsk/c2t5005076300C1928Bd2s4
topslnbap001:~> apdb -c /dev/rdsk/c2t5005076300C1928Bd3s4
topslnbap001:~> apdb -c /dev/rdsk/c2t5005076300C1928Bd4s4
topslnbap001:~> apdb -c /dev/rdsk/c2t5005076300C1928Bd5s4

topslnbap001:~> apconfig -D

path: /dev/rdsk/c2t5005076300C1928Bd0s4
major: 118
minor: 92
timestamp: Tue Feb 24 16:03:12 CST 2004
checksum: 2348860931
corrupt: No
inaccessible: No

path: /dev/rdsk/c2t5005076300C1928Bd1s4
major: 118
minor: 84
timestamp: Tue Feb 24 16:03:12 CST 2004
checksum: 1801799390
corrupt: No
inaccessible: No

path: /dev/rdsk/c2t5005076300C1928Bd2s4
major: 118
minor: 76
timestamp: Tue Feb 24 16:03:12 CST 2004
checksum: 2086193348
corrupt: No
inaccessible: No

path: /dev/rdsk/c2t5005076300C1928Bd3s4
major: 118
minor: 68
timestamp: Tue Feb 24 16:03:12 CST 2004
checksum: 776671888
corrupt: No
inaccessible: No

path: /dev/rdsk/c2t5005076300C1928Bd4s4
major: 118
minor: 60
timestamp: Tue Feb 24 16:03:12 CST 2004
checksum: 842496248
corrupt: No
inaccessible: No

path: /dev/rdsk/c2t5005076300C1928Bd5s4
major: 118
minor: 52
timestamp: Tue Feb 24 16:03:12 CST 2004
checksum: 4041822888
corrupt: No
inaccessible: No

I can view the following alternate path ports :

topslnbap001:~> apinst
fp:0
        /dev/dsk/c1t0d0
        /dev/dsk/c1t1d0
fp:1
        /dev/dsk/c2t5005076300C1928Bd0
        /dev/dsk/c2t5005076300C1928Bd1
        /dev/dsk/c2t5005076300C1928Bd2
        /dev/dsk/c2t5005076300C1928Bd3
        /dev/dsk/c2t5005076300C1928Bd4
        /dev/dsk/c2t5005076300C1928Bd5
        /dev/dsk/c2t5005076300C1928Bd6
        /dev/dsk/c2t5005076300C1928Bd7
        /dev/dsk/c2t5005076300C1928Bd8
        /dev/dsk/c2t5005076300C1928Bd9
fp:2
        /dev/dsk/c3t5005076300C5928Bd0
        /dev/dsk/c3t5005076300C5928Bd1
        /dev/dsk/c3t5005076300C5928Bd2
        /dev/dsk/c3t5005076300C5928Bd3
        /dev/dsk/c3t5005076300C5928Bd4
        /dev/dsk/c3t5005076300C5928Bd5
        /dev/dsk/c3t5005076300C5928Bd6
        /dev/dsk/c3t5005076300C5928Bd7
        /dev/dsk/c3t5005076300C5928Bd8
        /dev/dsk/c3t5005076300C5928Bd9

So I set up the disk path group:

topslnbap001:~> apdisk -c -p fp:1 -a fp:2
topslnbap001:~> apconfig -S
topslnbap001:~> apconfig -S -u

c3 fp:2
c2 fp:1 P A
        metadiskname(s):
                mc2t5005076300C1B U
                mc2t5005076300C1B U
                mc2t5005076300C1B U
                mc2t5005076300C1B U
                mc2t5005076300C1B U
                mc2t5005076300C1B U
                mc2t5005076300C1B U
                mc2t5005076300C1B U
                mc2t5005076300C1B U
                mc2t5005076300C1B U

These appear a bit strange...but I go ahead and commit them:
topslnbap001:~> apdb -C
topslnbap001:~> ls -l /dev/ap/dsk
total 32
lrwxrwxrwx 1 root other 39 Feb 24 16:05 mc2t5005076300C1Bs0
-> ../../../devices/pseudo/ap_dmd@0:16,blk
lrwxrwxrwx 1 root other 39 Feb 24 16:05 mc2t5005076300C1Bs1
-> ../../../devices/pseudo/ap_dmd@0:17,blk
lrwxrwxrwx 1 root other 39 Feb 24 16:05 mc2t5005076300C1Bs2
-> ../../../devices/pseudo/ap_dmd@0:18,blk
lrwxrwxrwx 1 root other 39 Feb 24 16:05 mc2t5005076300C1Bs3
-> ../../../devices/pseudo/ap_dmd@0:19,blk
lrwxrwxrwx 1 root other 39 Feb 24 16:05 mc2t5005076300C1Bs4
-> ../../../devices/pseudo/ap_dmd@0:20,blk
lrwxrwxrwx 1 root other 39 Feb 24 16:05 mc2t5005076300C1Bs5
-> ../../../devices/pseudo/ap_dmd@0:21,blk
lrwxrwxrwx 1 root other 39 Feb 24 16:05 mc2t5005076300C1Bs6
-> ../../../devices/pseudo/ap_dmd@0:22,blk
lrwxrwxrwx 1 root other 39 Feb 24 16:05 mc2t5005076300C1Bs7
-> ../../../devices/pseudo/ap_dmd@0:23,blk
lrwxrwxrwx 1 root other 39 Feb 24 14:51 mc3t5005076300C5Bs0
-> ../../../devices/pseudo/ap_dmd@0:96,blk
lrwxrwxrwx 1 root other 39 Feb 24 14:51 mc3t5005076300C5Bs1
-> ../../../devices/pseudo/ap_dmd@0:97,blk
lrwxrwxrwx 1 root other 39 Feb 24 14:51 mc3t5005076300C5Bs2
-> ../../../devices/pseudo/ap_dmd@0:98,blk
lrwxrwxrwx 1 root other 39 Feb 24 14:51 mc3t5005076300C5Bs3
-> ../../../devices/pseudo/ap_dmd@0:99,blk
lrwxrwxrwx 1 root other 40 Feb 24 14:51 mc3t5005076300C5Bs4
-> ../../../devices/pseudo/ap_dmd@0:100,blk
lrwxrwxrwx 1 root other 40 Feb 24 14:51 mc3t5005076300C5Bs5
-> ../../../devices/pseudo/ap_dmd@0:101,blk
lrwxrwxrwx 1 root other 40 Feb 24 14:51 mc3t5005076300C5Bs6
-> ../../../devices/pseudo/ap_dmd@0:102,blk
lrwxrwxrwx 1 root other 40 Feb 24 14:51 mc3t5005076300C5Bs7
-> ../../../devices/pseudo/ap_dmd@0:103,blk
topslnbap001:~> ls -l /dev/ap/rdsk
total 32
lrwxrwxrwx 1 root other 39 Feb 24 16:05 mc2t5005076300C1Bs0
-> ../../../devices/pseudo/ap_dmd@0:16,raw
lrwxrwxrwx 1 root other 39 Feb 24 16:05 mc2t5005076300C1Bs1
-> ../../../devices/pseudo/ap_dmd@0:17,raw
lrwxrwxrwx 1 root other 39 Feb 24 16:05 mc2t5005076300C1Bs2
-> ../../../devices/pseudo/ap_dmd@0:18,raw
lrwxrwxrwx 1 root other 39 Feb 24 16:05 mc2t5005076300C1Bs3
-> ../../../devices/pseudo/ap_dmd@0:19,raw
lrwxrwxrwx 1 root other 39 Feb 24 16:05 mc2t5005076300C1Bs4
-> ../../../devices/pseudo/ap_dmd@0:20,raw
lrwxrwxrwx 1 root other 39 Feb 24 16:05 mc2t5005076300C1Bs5
-> ../../../devices/pseudo/ap_dmd@0:21,raw
lrwxrwxrwx 1 root other 39 Feb 24 16:05 mc2t5005076300C1Bs6
-> ../../../devices/pseudo/ap_dmd@0:22,raw
lrwxrwxrwx 1 root other 39 Feb 24 16:05 mc2t5005076300C1Bs7
-> ../../../devices/pseudo/ap_dmd@0:23,raw
lrwxrwxrwx 1 root other 39 Feb 24 14:51 mc3t5005076300C5Bs0
-> ../../../devices/pseudo/ap_dmd@0:96,raw
lrwxrwxrwx 1 root other 39 Feb 24 14:51 mc3t5005076300C5Bs1
-> ../../../devices/pseudo/ap_dmd@0:97,raw
lrwxrwxrwx 1 root other 39 Feb 24 14:51 mc3t5005076300C5Bs2
-> ../../../devices/pseudo/ap_dmd@0:98,raw
lrwxrwxrwx 1 root other 39 Feb 24 14:51 mc3t5005076300C5Bs3
-> ../../../devices/pseudo/ap_dmd@0:99,raw
lrwxrwxrwx 1 root other 40 Feb 24 14:51 mc3t5005076300C5Bs4
-> ../../../devices/pseudo/ap_dmd@0:100,raw
lrwxrwxrwx 1 root other 40 Feb 24 14:51 mc3t5005076300C5Bs5
-> ../../../devices/pseudo/ap_dmd@0:101,raw
lrwxrwxrwx 1 root other 40 Feb 24 14:51 mc3t5005076300C5Bs6
-> ../../../devices/pseudo/ap_dmd@0:102,raw
lrwxrwxrwx 1 root other 40 Feb 24 14:51 mc3t5005076300C5Bs7
-> ../../../devices/pseudo/ap_dmd@0:103,raw

These are definitely strange and appear wrong. Any attempt to access
them leads to a file not found because of the stange embedded characters
as well as the truncation. Does anybody have any idea what's wrong?
I will summarize.

-- 
Lonnie Randall Webb <lrw@tel.bnsf.com>
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers

Next message: Chris Hoogendyk: "recovering space in /var for doing recommended patches"

Previous message: Lauri Bettencourt: "metadb Issue"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Relevant Pages

Re: Corrupt partition or virus?
... It sounds to me like drive and/or partition problems. ... viruses out there that are capable of rendering entire directory structures ... what would cause this if it was a corrupt ...
(microsoft.public.windowsxp.general)
Re: Guidance Software response to iSEC report on EnCase (fwd)
... crashing EnCase is on par with kicking retarded children out of their ... MBR partition table really doesn't break anything, it just requires that the boot code in the first ... corrupt the integrity of the image you made, and thats even I can't grab ...
(Bugtraq)
Mandrak 9.1 off the deep end on ocassion
... Why does the recovery process loose information or corrupt files? ... Then recover damaged files as needed? ... It corrupts the /root partition last of all. ...
(comp.os.linux.setup)
Re: Recovery of data from corrupt Windows Vista install
... which may be able to read the corrupt ... partition as simply a data drive (esp. ... You could also install another HD, ... The master HDD has a corrupted Windows Vista ...
(comp.sys.ibm.pc.hardware.storage)
Re: detecting corrupt/stuffed files (how to?)
... (have the same checksum). ... > If the file system got damaged, the OS has to care about this (once it ... > If just the data in the file is corrupt, and not the file system, it ... but is a good protection for accidents. ...
(comp.lang.java.programmer)