topic Re: pvcreate getting i/o errors in Operating System - HP-UX

pvcreate getting i/o errors

KPS — Thu, 14 Aug 2008 18:58:57 GMT

I'm having an issue on a BL870C where when we run pvcreate on some newly presented SAN Disk from an EMC DMX-3, we're getting i/o errors. see below the error:

pvcreate /dev/rdsk/c7t8d3
pvcreate: IO error on Physical Volume device "/dev/rdsk/c7t8d3".

We're running HP-UX 11.23 on this BL870C. I tried running dd against the disk in question and got equal records and out. I also ran diskinfo -v and it doesn't even recognize the size of the disk: (See Below)

diskinfo -v /dev/rdsk/c7t8d3
SCSI describe of /dev/rdsk/c7t8d3:
vendor: EMC
product id: SYMMETRIX
type: direct access
size: 0 Kbytes
bytes per sector: 0
rev level: 5772
blocks per disk: 0
ISO version: 0
ECMA version: 0
ANSI version: 2
removable media: no
response format: 2
(Additional inquiry bytes: (32)39 (33)39 (34)33 (35)31 (36)30 (37)30 (38)30 (39)61 (40)0 (41)0 (42)8 (43)c1 (44)0 (45)0 (46)0 (47)0 (48)0 (49)0 (50)0 (51)57 (52)72 (53)0 (54)85 (55)0 (56)0 (57)0 (58)0 (59)0 (60)12 (61)21 (62)7 (63)0 (64)0 (65)0 (66)0 (67)0 (68)0 (69)0 (70)0 (71)0 (72)0 (73)0 (74)0 (75)0 (76)0 (77)0 (78)0 (79)0 (80)0 (81)0 (82)0 (83)57 (84)72 (85)0 (86)85 (87)0 (88)12 (89)21 (90)7 (91)0 (92)0 (93)0 (94)0 (95)0 (96)0 (97)0 (98)0 (99)12 (100)0 (101)0 (102)0 (103)54 (104)31 (105)0 (106)8 (107)bf (108)80 (109)0 (110)0 (111)0 (112)0 (113)0 (114)0 (115)61 (116)0 (117)13 (118)0 (119)1 (120)90 (121)10 (122)11 (123)0 (124)0 (125)0 (126)0 (127)0 (128)0 (129)0 (130)0 (131)0 (132)0 (133)0 (134)0 (135)0 (136)0 (137)0 (138)0 (139)0 (140)0 (141)0 (142)0 (143)0 )

Would anyone have any ideas to what could be causing this issue. Could this be the way the devices are presented to our HP-UX host?

Thanks in advance,
/KPS

Re: pvcreate getting i/o errors

Tim Nelson — Thu, 14 Aug 2008 20:10:54 GMT

My first guess would be the DMX controller flags or some flag on the presentation.

Do you have other working devices presented to this server ?

Discussion about controller flags can be found in the EMC Host connectivity guides found on the EMC Powerlink website.

If you supplied some more info on your environment I might be able to guide you more.

e.g.

Is this the first or only HPUX server in this fabric ?
Are there others working?
If so use the "others" to compare the configurations of presentation, masking and zoning of DMX controllers.

Re: pvcreate getting i/o errors

Deepak Kr — Thu, 14 Aug 2008 22:42:16 GMT

Hi,
what is the version of powerpath software on your hp box?

also is this the first time you are using disk from emc san to this box

Re: pvcreate getting i/o errors

Bill Hassell — Thu, 14 Aug 2008 22:56:48 GMT

The I/O error is due to pvcreate trying to write beyond the end of the disk. Since your disk (LUN) is replying with a size of 0 bytes, the I/O error is quite normal. Talk to your SAN administrator -- the LUN has not been mapped or presented correctly.

Re: pvcreate getting i/o errors

KPS — Fri, 15 Aug 2008 10:49:11 GMT

Good question about any other working devices. They gave me 7 devices in this mapping and masking presentation to our BL870. Only one of them out of the 7 LUN's is not having that i/o error when I run pvcreate against it. The other 6 are all having the i/o error. Not sure if that helps diagnose the problem a little better.

THanks,
/KPS

Re: pvcreate getting i/o errors

Bill Hassell — Fri, 15 Aug 2008 22:40:31 GMT

> Only one of them out of the 7 LUN's is not having that i/o error when I run pvcreate against it. The other 6 are all having the i/o error. Not sure if that helps diagnose the problem a little better.

Yes, this eliminates any communication problems such as the fibre or HBA. There's nothing for you to do except get the storage administrator to fix the maps and zones, etc. Remind the admin that this is not a PC since Windows often requires a different setting.

Re: pvcreate getting i/o errors

KPS — Sat, 16 Aug 2008 01:29:12 GMT

This ended up being an issue with the SAN side of things as many of you expected. It turned out only one of the devices were flagged in a "READY" state. The other 6 were not. This was achieved by running the following from the HP-UX side if you have EMC's Solution Enabler installed:

symdev -sid ready

Many thanks to everyone for offering your recommendations and tips..

Thanks,
KPS