PSU Issue on DL Server - confusing!!

[edikat]

Hi

I purchased a used DL560 G1 Quad CPU (3.0GHZ) server with Dual PSU. Within 30 minutes of powering up I heard a "pop" and smelt some smoke.

System indicated a PSU failure but kept going on remaining one PSU. The PSU was definitely faulty as I tried by itself and also checked the server status for the PSU that stated "faulty".

I ordered a second PSU, and plugged it in and got redundancy again.

After another 20 hours or so I heard another "pop" and smelt some smoke again (just like the first time but maybe not as loud).

The server shutdown with a "PSU fault" Red light (at least I think it was the PSU light and not the server health light).
I thought **bleep**.. both PSU have gone??

After recycling the power however the system has booted up fine with both PSU working....

Any ideas what happened?
Something definitely blew and the smell is quite strong!

/var/log/messages shows:

I think during bootup:

Sep 28 08:39:27 localhost hpasmd[2593]: ProLiant System Health Monitor loading
Sep 28 08:41:06 localhost hpasmd[2593]: WARNING: hpasmd: System Power Supply: General Failure (Power Supply 2)
Sep 28 08:41:06 localhost hpasmd[2593]: WARNING: hpasmd: System Power Supplies Not Redundant
Sep 28 08:41:08 localhost hpasmd[2593]: NOTICE: hpasmd: System Power Supply: General Failure (Power Supply 2) has been repaired
Sep 28 08:41:08 localhost hpasmd[2593]: NOTICE: hpasmd: System Power Supplies Not Redundant has been repaired

after a power on:

Power supply #1
Present : Yes
Redundant: Yes
Condition: Ok
Hotplug : Supported
Power supply #2
Present : Yes
Redundant: Yes
Condition: Ok
Hotplug : Supported


All the server health lights on the front panel are "green" as well.

Many thanks!
Ed

[edikat]

OK it was c969 under the PCI-X riser cage that had blown... but the server still works ok!?

[Abhishek Bangalore]

Hello,

I understand that there was a smell from the server and now it is working fine after rebooting.

Still I would suggest you to run the offline diagnostic on the server in order to check all the parts are working fine.

Please refer the steps provided below in order to do the same:

Please run offline diagnostic on the server to confirm the hardware issue, To run offline diagnostic you need to boot the server with the smart start cd

We need to schedule a downtime for around 2 hrs to run the offline diagnostics.Schedule a downtime & run the offline diagnostics using the smart start cd.

However we you can also download the Smart Start cd from the below given link HP SmartStart CD

http://h20000.www2.hp.com/bizsupport/TechSupport/SoftwareDescription.jsp?lang=en&cc=us&prodTypeId=15351&prodSeriesId=316583&prodNameId=3288105&swEnvOID=1005&swLang=8&mode=2&taskId=135&swItem=MTX-d7672226b6284520b9e292b9bd

1. Unzip the file and burn the resulting .iso image to a CD.2. Boot the server from this CD.3. Select the language and click "Continue"4. Accept the License Agreement.5. Select "Maintain Server"6. Click "Diagnose Server"7. Click the "Test" Tab at the top.8. Click the "Complete Test" Tab. Change the no of loops to 2.11. Select "All Devices" under "Select a device to test"12. Click the "Begin Testing" button on the lower right hand corner.13. Once the Testing is complete, Click on the "Log" Tab at the Top and then the "Test Log" Tab.14. Click on the "Save" Button and save the report to a floppy or a USB media or take the screenshots of the results for the future purpose

Survery report:
After the step 13, please select the tab Survey tab
Select Advanced and then select All.
Now capture the survey logs and Click on the "Save" Button and save the report to a floppy or a USB media.

Regards

Abhishek