Rack Servers

hmartens · ‎06-24-2009

We've recently got a Poweredge 2970 with a SAS card connected to a Powervault MD3000. While installing Centos 5.3 64bit it reboots the server towards the end of installation and it comes up with the error: HyperTransport error cause a system reset: Embedded I/O Bridge Device 2. Check system logs to get more info. I've managed to install Centos 5.3 64bit while SAS cables is not installed...but then now and then the server would just reboot with the same error. Has anyone come accross the same problem? Is it software problem or hardward? Hope you can help Hennie

Marcus.pitre · ‎08-11-2009

We have had the same problem with the same server, except we are running Exchange 2007 on Server 2008. Without reason the box will restart. Have you found a solution yet?

hmartens · ‎08-12-2009

Dell is aware of this problem and states that it's purely a software problem(firmware). They have escalated this issue to their hardware supplier and they are working on it. In the meantime Dell has replaced the motherboard and it didn't fix the problem. It's still doing it. They emailed me yesterday saying that there is a new DSET that I must download and run so and send them the report so they can check. Hopefully the issue will be resolved soon.

EliBruner · ‎08-22-2009

I have the same issue:

HyperTransport error caused a system reset. Embedded I/O Bridge Device 2

Please check the system event log for details. F1 to continue.

The Dell Tech is looking at the DSET Report now. He's telling me I will probably need to update BIOS, drivers, and firmware.

When we're done, I'll post the results.

EliBruner · ‎08-22-2009

The DSET Report was analyzed. All necessary updated were already performed. We attempted to update the NIC firmware. However, the existing NIC firmware was the same version.

Because it is a Saturday, the Tech. told me he will need to work with the "Early Life Support Group". I guess they work with strange and unique problems. But, only Monday-Friday.

He's going to contact me again on Monday.

My PowerEdge 2970 is running SBS 2008 Premium. It has 2 Quad-Core AMD Opteron processors at 3.1GHz and 16GB System Mem. (This system was manufactured 7.25.09 ; It's brand new.) Because this is our Exchange Server (07). We really need to get this fixed so it doesn't happen again. This type of problem could cause a corruption in the Exchange database. I hope it doesn't happen again. /crossfingers

The original reset happened during system migration from SBS 03 to SBS 08 (different servers). I was migrating shared files from the old server to this "New" one. Luckly there were no corruption problems. The Dell Tech. metioned there may be a problem with the Northbridge or the embedded NIC.

I guess I'll have to wait for Monday to see if there is a sure fix. We'll see.

rsnair · ‎09-23-2009

Eli,

Was there any response/fix from Dell? I just saw the same issue on my new Poweredge 2970.

Please let us know.

Thanks - Rajesh.

hmartens · ‎09-23-2009

Yes it was fixed. They replaced the motherboard and it still gave the same error. So eventually they replaced one of the powersupplies and that seems to have sorted out the problem. It's been running now over a month with no problems. So I guess your best bet is to get them to replace the powersupplies.

rsnair · ‎09-23-2009

Just received an email back from hmartens (pasted below). Thanks a ton for replying back to me. Really appreciate it.

-Rajesh.

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Re: HyperTransport error caused a system reset. Embedded I/O Bridge Device 2
By hmartens in PowerEdge Rack HW

Yes it was fixed. They replaced the motherboard and it still gave the same error. So eventually they replaced one of the powersupplies and that seems to have sorted out the problem. It's been running now over a month with no problems. So I guess your best bet is to get them to replace the powersupplies.

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

astronomic · ‎01-11-2010

Is there a confirmed fix for this issue? Is this definitely a power supply problem, or a firmware issue? I have been putting up lots of these and over 25% of the servers crash with the same error.

Thanks!

turbinecore · ‎02-06-2010

I have also experienced the same error. The System Event Log reads:

Severity: Critical
Date and Time: Wed Feb 03 13:48:27 2010
Description:
Chipset Err: Critical Event sensor, PCI PERR (Bus
0 Device 7 Function 0) was asserted

Has anyone seen the HyperTransport Error and found this same description in the SEL?

Also, any fixes?

Thanks.