T110ii power supply

Question

My customer has a T110 Poweredge tower running Centos linux I set it up in August 2010. There were a number of issues with it at the time and eventually dell replaced the power supply under warranty Since then it has been running without a problem until this week. The warranty has now expired. Three times in the last few days the server has restarted itself during the night, usually interrupting the backup. Centos logs are normal up to the point of the restart with no clues as to the reason. Then the restart is logged and runs to completion, the server is then completely operational on the network and accessible. Once is one of those things, three times is something else. The system health led on the front panel is flashing, but the centos hardware diagnostics do not show any issues. I suspect another power supply problem and have been looking for a replacement 305W psu. However it appears there are two possibilities R480P and N238P - the dell site does not tell me which psu it has for 5HYXS4J and a replacement psu is not listed as a dell spare part. Two questions, would random restarts indicate a psu problem and second, does it have the R480P or N238P and are they interchangeable? Regards, Peter Lawrie

theflash1932 · Accepted Answer

That would certainly be enough to make the light blink amber, as it is a hardware error that would require your attention.  Since it is much less likely for a processor to fail, I would probably assume the motherboard at this point, but it is possible that the voltage regulators on the processor itself are not working properly.

plawrie · Answer

Further to my question about the power supply for a Dell T110 poweredge

It was suggested that I install the Dell OMSA for linux which I have just done -

Although I can't access OMSA through a browser, I think it has found a problem which is being logged in the /var/log/messages - someone please advise, is this motherboard or psu?

Would this cause the system health led to blink amber?

Nov 3 18:30:39 localhost Server Administrator: Instrumentation Service EventID: 1000 Server Administrator starting

Nov 3 18:30:39 localhost Server Administrator: Instrumentation Service EventID: 1154 Voltage sensor detected a failure value Sensor location: CPU1 0.75 VTT PG Chassis location: Main System Chassis Previous state was: Unknown Discrete voltage state: Bad

Nov 3 18:30:39 localhost Server Administrator: Instrumentation Service EventID: 1012 IPMI status Interface: OS

Nov 3 18:30:39 localhost Server Administrator: Instrumentation Service EventID: 1001 Server Administrator startup complete

Nov 3 18:32:41 localhost kernel: IPMI System Interface driver.

Nov 3 18:32:41 localhost kernel: ipmi_si: Trying SMBIOS-specified kcs state machine at i/o address 0xca8, slave address 0x20, irq 0

Nov 3 18:32:42 localhost kernel: ipmi: Found new BMC (man_id: 0x0002a2, prod_id: 0x0100, dev_id: 0x20)

Nov 3 18:32:42 localhost kernel: IPMI kcs interface initialized

Nov 3 18:32:43 localhost instsvcdrv: disabled kipmi0 thread

plawrie · Answer

Theflash1932 Thanks, armed with this, I'll telephone Dell on Monday and see what the options are. Peter

PowerEdge Hardware General

Was this post helpful?