VMware Cloud Community
kp238
Contributor
Contributor

Trouble with IBM x3650 M2 and ESX 4.0.0 Build 164009

I have 8 new IBM x3650 M2 (P/N: 794792G) servers installed with ESX 4.0.0 Build 164009. The servers come from IBM with UEFI Firmware 1.01 (with fixed problem "Interrupt routing was not valid for VMWARE 4.0" - ).

The Service Console show this error:

0:00:00:02.582 cpu0:4096)CosCommon: 2103: Couldn't setup IRQ 11 with vector 193

0:00:00:02.582 cpu0:4096)CosCommon: 2103: Couldn't setup IRQ 11 with vector 193

Who has an idea for solution?

IBM say the problem is ESX 4.0 software and not the IBM hardware.

Reply
0 Kudos
47 Replies
jarsenea
Contributor
Contributor

Just so you know, I have two new x3650 M2's with the same issue (different vector # however).

This KB ->

There are simply too many devices using IRQ11. The KB indicates that you have to tell the BIOS to have these devices use different IRQs, but the x3650 M2 bios doesn't provide any option to do so.. I haven't had any issues so I don't think it's a big deal, but it is frustrating seeing that red message appear when you boot ESX Smiley Happy

Reply
0 Kudos
vmroyale
Immortal
Immortal

Hello. Did either of you ever get this resolved? I'm running into the same issue. Thanks.

Brian Atkinson | vExpert | VMTN Moderator | Author of "VCP5-DCV VMware Certified Professional-Data Center Virtualization on vSphere 5.5 Study Guide: VCP-550" | @vmroyale | http://vmroyale.com
Reply
0 Kudos
jarsenea
Contributor
Contributor

I didn't get it resolved, no...

I've been running these ESX servers for just under a month on vSphere (essentially since May 21) and I haven't had any issues. I'm thinking it may be more of a warning than anything else.

Reply
0 Kudos
MCJJJJ
Contributor
Contributor

I've got the same issue. It would be nice if someone could post whether or not we should worry about this message.

-J

Reply
0 Kudos
jidhin
VMware Employee
VMware Employee

please upgrade the BIOS to latest version.

If you found this information useful, please consider awarding points for "Correct" or "Helpful". Thanks!
Reply
0 Kudos
MCJJJJ
Contributor
Contributor

Sorry if I was unclear, but by saying I have the same problem I meant that I have updated to the latest uEFI/BIOS that IBM has on their website, just like the original post mentions. It is version D6E124AUS or, as referred to above 1.01 and can be found here:

http://www-947.ibm.com/systems/support/supportsite.wss/docdisplay?lndocid=MIGR-5079767&brandind=5000...

Unless there is an update that is only available by calling IBM support I'm not sure what you are referring to. Is there a newer uEFI/BIOS that I should be running?

Reply
0 Kudos
MCJJJJ
Contributor
Contributor

I spoke to VMware and they are aware of this issue and are working with IBM on a solution.

Reply
0 Kudos
vmroyale
Immortal
Immortal

Thanks for the replies.

Brian Atkinson | vExpert | VMTN Moderator | Author of "VCP5-DCV VMware Certified Professional-Data Center Virtualization on vSphere 5.5 Study Guide: VCP-550" | @vmroyale | http://vmroyale.com
Reply
0 Kudos
kp238
Contributor
Contributor

With UEFI Firmware 1.00 boot ESX 4.0.0 Build 164009 without IRQ 11 failure messages.

This UEFI Firmware 1.00 is in UpdateXpress System Pack for Microsoft Windows v1.01 included.

Attention: 2 of my 8 servers with downgraded UEFI Firmware 1.00 an IMM Firmware 1.03 irreparably and IBM has replaced the motherboards.

Reply
0 Kudos
renpen
Contributor
Contributor

I had the same issue with my 3650M2. Downgraged UEFI one version to ver 1.00 and issue went away.

KP238, which Raid controller do you have in your system - BR10i or MR10i?

Have you been able to install ESX 3.5 on your system. I cannot - just hangs on initializing scheduler screen

Reply
0 Kudos
NM777
Contributor
Contributor

I have the same error but with different vector 209. The server is IBM blade HS22 and when i updated the bios to the latest uEFI/BIOS, the error appears in the ESX 4 console.

0:00:00:02.582 cpu0:4096)CosCommon: 2103: Couldn't setup IRQ 11 with vector 209

0:00:00:02.582 cpu0:4096)CosCommon: 2103: Couldn't setup IRQ 11 with vector 209

output of interrupts:

# cat /proc/vmware/interrupts | grep 'irq *11'

0x79: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 <COS irq 11 (PCI level)>

0x81: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 <COS irq 11 (PCI level)>

0x89: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 <COS irq 11 (PCI level)>

0x91: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 <COS irq 11 (PCI level)>

0x99: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 <COS irq 11 (PCI level)>

0xa1: 97 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 <COS irq 11 (PCI level)>, VMK qla2xxx

0xa9: 327 0 651 0 0 1673 0 0 0 0 0 0 0 0 0 1777 <COS irq 11 (PCI level)>, VMK uhci_hcd:usb5, VMK ehci_hcd:usb1

thanks

Reply
0 Kudos
MCJJJJ
Contributor
Contributor

Just thought I'd share that I decided to go to ESXi rather than use ESX and I haven't had these problems. Why switch? Well, VMware recommends ESXi over the older ESX platform. Check out their blog article for more info. I know some might not agree (no console and I'm sure many other reasons), but we decided we'd rather align ourselves with VMware's future goals. Not sure if this will really help anyone, but I figured it might be good info to post here.

Reply
0 Kudos
trurodh
Enthusiast
Enthusiast

Has anyone ever heard anything more on this issue? I have the same issue and I have disabled a few things and the errors still show up. The server seems to work fine but can't imagine that being a good thing anyway. It is definetly something to do with the 1.01 UEFI compared to the 1.0. I will have to open a ticket with both IBM and VMware I suppose. If anyone in this thread has heard back from either let me know. I will do the same.

Thanks.

Rod

Reply
0 Kudos
jarsenea
Contributor
Contributor

I know this isn't a real solution, but seeing as ESX will be phased out within the next year or so (and based on my conversations with a few VMware engineers), I've migrated my servers to ESXi from ESX. No irq errors are appearing after having moved to ESXi and (in my environment anyway) I haven't lost any functionality going to ESXi.

Reply
0 Kudos
NM777
Contributor
Contributor

I have ticket with both ibm and vmware. They are still working with this issue.

Noel Maldonado

Reply
0 Kudos
trurodh
Enthusiast
Enthusiast

Hi Noel

Did either of them say anything up front. What I mean is did they say it would cause any issues having the interrupts over allocating? Did you have to back rev to 1.0 on the UEIF? The server appears to be ok with the problem.

Thanks for the reply!

Rod

Reply
0 Kudos
NM777
Contributor
Contributor

Hi Rod

Well, IBM replicated the problem in their labs and is being manage by product engineering/development. Suppport told us it would not cause any issues, but we will not go in production with this error. We did not back rev to 1.0 on the hs22 blades.

Thks

Noel

Noel Maldonado

Reply
0 Kudos
Staert
Contributor
Contributor

Any news on this issue?? I'm struggling with the same problem with ESX 4.0 Build 171294 and UEFI V1.02

Thanks in advance..

Reply
0 Kudos
NM777
Contributor
Contributor

IBM support is working in a new EFI firmware to resolve this issue.

Noel Maldonado

Reply
0 Kudos