Contributor
Contributor

vmkernel log shows errors Peripheral qualifier 0x1 not supported for MD3000i paths

Jump to solution

I have a vSphere cluster connected to MD3000i and its configured by the book on two different subnets for each port on each controler. Round robin is enabled for multi pahing on VC.

this is what i am seeing in vmkernel log

Sep 21 16:19:29 db-esx2 vmkernel: 6:04:15:02.613 cpu4:6882)ScsiScan: 839: Path 'vmhba33:C0:T0:L0': Vendor: 'DELL ' Model: 'MD3000i ' Rev: '0735'

Sep 21 16:19:29 db-esx2 vmkernel: 6:04:15:02.613 cpu4:6882)ScsiScan: 842: Path 'vmhba33:C0:T0:L0': Type: 0x0, ANSI rev: 5, TPGS: 0 (none)

Sep 21 16:19:29 db-esx2 vmkernel: 6:04:15:02.613 cpu4:6882)ScsiScan: 105: Path 'vmhba33:C0:T0:L0': Peripheral qualifier 0x1 not supported

Sep 21 16:19:29 db-esx2 vmkernel: 6:04:15:02.618 cpu4:6882)ScsiScan: 839: Path 'vmhba33:C1:T0:L0': Vendor: 'DELL ' Model: 'MD3000i ' Rev: '0735'

Sep 21 16:19:29 db-esx2 vmkernel: 6:04:15:02.618 cpu4:6882)ScsiScan: 842: Path 'vmhba33:C1:T0:L0': Type: 0x0, ANSI rev: 5, TPGS: 0 (none)

Sep 21 16:19:29 db-esx2 vmkernel: 6:04:15:02.618 cpu4:6882)ScsiScan: 105: Path 'vmhba33:C1:T0:L0': Peripheral qualifier 0x1 not supported

Sep 21 16:19:29 db-esx2 vmkernel: 6:04:15:02.623 cpu4:6882)ScsiScan: 839: Path 'vmhba33:C2:T0:L0': Vendor: 'DELL ' Model: 'MD3000i ' Rev: '0735'

Sep 21 16:19:29 db-esx2 vmkernel: 6:04:15:02.623 cpu4:6882)ScsiScan: 842: Path 'vmhba33:C2:T0:L0': Type: 0x0, ANSI rev: 5, TPGS: 0 (none)

Sep 21 16:19:29 db-esx2 vmkernel: 6:04:15:02.623 cpu4:6882)ScsiScan: 105: Path 'vmhba33:C2:T0:L0': Peripheral qualifier 0x1 not supported

Sep 21 16:19:29 db-esx2 vmkernel: 6:04:15:02.628 cpu4:6882)ScsiScan: 839: Path 'vmhba33:C3:T0:L0': Vendor: 'DELL ' Model: 'MD3000i ' Rev: '0735'

Sep 21 16:19:29 db-esx2 vmkernel: 6:04:15:02.628 cpu4:6882)ScsiScan: 842: Path 'vmhba33:C3:T0:L0': Type: 0x0, ANSI rev: 5, TPGS: 0 (none)

Sep 21 16:19:29 db-esx2 vmkernel: 6:04:15:02.628 cpu4:6882)ScsiScan: 105: Path 'vmhba33:C3:T0:L0': Peripheral qualifier 0x1 not supported

I see this on both hosts vmkernel log...

should i be concerned. Its not necesserely showing error but :" Peripheral qualifier 0x1 not supported " is getting my attention.

0 Kudos
1 Solution

Accepted Solutions
Hot Shot
Hot Shot

If you do NOT have any luns assigned as Lun0 then you will get this error. So always start presenting luns with Lun0 aka: Lun0 - First Disk you assign to it, Lun1 - Second Disk etc. You don't neccessarly have to reassign your currently lun, you could create a new lun and present it as Lun 0. So try that if you are able to create a new Lun. Otherwise if you have existing VM's running on the current Lun5 you would have to stop them, unregister them from the inventory, change the lunID from 5 to 0 and then reregister them I believe. Also, make sure also you dont skip any lun numbers ... aka go Lun0,Lun1,Lun2,Lun3 etc.

View solution in original post

0 Kudos
16 Replies
Immortal
Immortal

What do you have on MD3000i side?

IMHO I suggest to use the default multipath policy and not round robin.

If you have more that one virtual disk you can use different path (on different VD) also without round robin.

Andre

Andre | http://about.me/amauro | http://vinfrastructure.it/ | @Andrea_Mauro
0 Kudos
Contributor
Contributor

Based on limited experience/evidence I would speculate that the MD3000i controller management LUN - which appears on LUN 0 - is saying "I'm not a real disk" (peripheral qualifier 0x1) and ESX is logging that. But I could be wrong!

0 Kudos
Hot Shot
Hot Shot

Did you figure this out? having same issue with different san vender.

0 Kudos
VMware Employee
VMware Employee

Hi,

did you mapped lun 0 on MD3000i? please have a recheck, thanks

I also hit similar vmkernel messages when no lun 0 on MD3000 mapped to my hosts,

3:22:40:04.829 cpu6:4193)ScsiScan: SCSIVerifyStandardInquiryData:110: Path 'vmhba2:C0:T1:L0': Peripheral qualifier 0x1 not supported

3:22:45:04.771 cpu4:4194)ScsiScan: SCSILogPathInquiry:1056: Path 'vmhba2:C0:T0:L0': Vendor: 'DELL ' Model: 'MD3000 ' Rev: '0735'

3:22:45:04.783 cpu4:4194)ScsiScan: SCSILogPathInquiry:1059: Path 'vmhba2:C0:T0:L0': Type: 0x0, ANSI rev: 5, TPGS: 0 (none)

3:22:45:04.794 cpu4:4194)ScsiScan: SCSIVerifyStandardInquiryData:110: Path 'vmhba2:C0:T0:L0': Peripheral qualifier 0x1 not supported

3:22:45:04.810 cpu6:4194)ScsiScan: SCSILogPathInquiry:1056: Path 'vmhba2:C0:T1:L0': Vendor: 'DELL ' Model: 'MD3000 ' Rev: '0735'

3:22:45:04.823 cpu6:4194)ScsiScan: SCSILogPathInquiry:1059: Path 'vmhba2:C0:T1:L0': Type: 0x0, ANSI rev: 5, TPGS: 0 (none)

3:22:45:04.833 cpu6:4194)ScsiScan: SCSIVerifyStandardInquiryData:110: Path 'vmhba2:C0:T1:L0': Peripheral qualifier 0x1 not supported

0 Kudos
VMware Employee
VMware Employee

Hi Guys,

if it's still relevant, this might be of interest.

Cheers

Valentin

Contributor
Contributor

No the LUN is number 5 on the MD3000i if thats what you meant.

0 Kudos
Hot Shot
Hot Shot

Yes, this is a LUN 0 Issue, you must have a Lun presented as Lun 0 otherwise you may see this in /var/log/vmkernel

0 Kudos
VMware Employee
VMware Employee

Hi armaniseal,

LUN 0 is just used as an examlple in the KB, it could be any LUN over any path*. Please contact your storage vendor to check if everything is set up correct.

HTH

Valentin

*edit: theoretically, I don't know for sure if your current issue is not due to some LUN (0) presentation issues, as I said, please contact your storage vendor Smiley Happy

0 Kudos
Contributor
Contributor

jcwuerfl are you saying the LUN that i have which is LUN5 within MD3000i needs to be LUN 0?

what if you have more then one lun ?? Doesn't make sense to me... I mean i can change the LUN number to 0 as its not taken but what if you have more then one LUN presented to the esx host they cant all be LUN 0.

0 Kudos
Contributor
Contributor

vbondzio the design is based of the MD3000i and ESX guide. I built it based on vendor recommendations from both sides. There is nothing more they can tell me.. funny thing is.. look at the log

Oct 27 10:20:19 db-esx2 vmkernel: 21:18:35:02.614 cpu8:23097)ScsiScan: 105: Path 'vmhba33:C0:T0:L0': Peripheral qualifier 0x1 not supported

Oct 27 10:20:19 db-esx2 vmkernel: 21:18:35:02.619 cpu8:23097)ScsiScan: 839: Path 'vmhba33:C1:T0:L0': Vendor: 'DELL ' Model: 'MD3000i ' Rev: '0735'

Oct 27 10:20:19 db-esx2 vmkernel: 21:18:35:02.619 cpu8:23097)ScsiScan: 842: Path 'vmhba33:C1:T0:L0': Type: 0x0, ANSI rev: 5, TPGS: 0 (none)

Oct 27 10:20:19 db-esx2 vmkernel: 21:18:35:02.619 cpu8:23097)ScsiScan: 105: Path 'vmhba33:C1:T0:L0': Peripheral qualifier 0x1 not supported

Oct 27 10:20:19 db-esx2 vmkernel: 21:18:35:02.624 cpu8:23097)ScsiScan: 839: Path 'vmhba33:C2:T0:L0': Vendor: 'DELL ' Model: 'MD3000i ' Rev: '0735'

Oct 27 10:20:19 db-esx2 vmkernel: 21:18:35:02.624 cpu8:23097)ScsiScan: 842: Path 'vmhba33:C2:T0:L0': Type: 0x0, ANSI rev: 5, TPGS: 0 (none)

Oct 27 10:20:19 db-esx2 vmkernel: 21:18:35:02.624 cpu8:23097)ScsiScan: 105: Path 'vmhba33:C2:T0:L0': Peripheral qualifier 0x1 not supported

Oct 27 10:20:19 db-esx2 vmkernel: 21:18:35:02.629 cpu8:23097)ScsiScan: 839: Path 'vmhba33:C3:T0:L0': Vendor: 'DELL ' Model: 'MD3000i ' Rev: '0735'

Oct 27 10:20:19 db-esx2 vmkernel: 21:18:35:02.629 cpu8:23097)ScsiScan: 842: Path 'vmhba33:C3:T0:L0': Type: 0x0, ANSI rev: 5, TPGS: 0 (none)

scan 839 is returning Vendor DELL etc..

others 0x0 and 0x1... what are the other two though? 105 and 842? what are these numbers?

0 Kudos
Hot Shot
Hot Shot

If you do NOT have any luns assigned as Lun0 then you will get this error. So always start presenting luns with Lun0 aka: Lun0 - First Disk you assign to it, Lun1 - Second Disk etc. You don't neccessarly have to reassign your currently lun, you could create a new lun and present it as Lun 0. So try that if you are able to create a new Lun. Otherwise if you have existing VM's running on the current Lun5 you would have to stop them, unregister them from the inventory, change the lunID from 5 to 0 and then reregister them I believe. Also, make sure also you dont skip any lun numbers ... aka go Lun0,Lun1,Lun2,Lun3 etc.

View solution in original post

0 Kudos
Contributor
Contributor

You are right i do not have any LUN 0 .. i will test and update.

0 Kudos
VMware Employee
VMware Employee

Hi armaniseal,

Those messages behind the path are all SCSI standard conform, check t10.org, but the important excerpt is also in the KB I linked to. As for the VMKernel Log line:

Date

Hostname

Source

Uptime

CPU the event was logged from

World (Process) that logged the event

Severity

Component / Module

Line Number (in the sourcecode of that module)

I still recommend to get your storage vendor to have a look at this to reference against the array log. Don't forget to let the community know what you found out once you get an answer Smiley Happy

Cheers

Valentin

0 Kudos
Contributor
Contributor

You are right i have a second environment IDENTICAL and the LUN number is 0 and there are no such errors!

0 Kudos
Hot Shot
Hot Shot

Very cool, thanks for the update.

0 Kudos
Contributor
Contributor

Hi All,

You all guys are going in the wrong direction. Valentin, what you said is correct. At least you should engage your storage vendor into it.

The original problem is not with LUN 0 or LUN 5 but the actual problem is with Storage firmware. Just check the storage firmware it should start with 06.xx.xx.xx which shows its first generation firmware. This firmware version of MD 3000i is not supported by VSphere ESX4.

To avoid the error you are getting in vmkernel log, you need to upgrade your firmware to second generation firmware which start with 07.xx.xx.xx.

I was also getting the same error and after searching through web and engaging the storage vendor, I came to know this solution.

The peripheral device error you are getting is nowhere related to LUN number. It is just showing you that your device type is not accessible by ESX Server.

I am posting a link to one of the KB article of VMware which will help you to understand your error better.

http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=101227...

Regards.

Mehul

0 Kudos