1

After looking at technical documents, I could not figure out cause of memory channel A not working on intel serverboard s1200v3rps. So, I send my complete server to dealer center to initiate warranty return.

My dealer checked the system and just now sent me pictures of serverboard and CPU and they appear to be burnt.(Pictures attached). Now, my dealer has informed me that since there appears to burning, warranty might be rejected for both motherboard and CPU.

I have been using this system for past 2 and half years with heavy usage and only couple of days ago, Memory channel A not working was first sign of trouble. I regularly monitor cpu temperature during usage and cpu core temperatures did not rise above 70 C during usage. So, I want to know if it is normal for cpu and serverboard pins to get black due to heavy usage. If yes, will Intel still reject warranty claim since I have used server too heavily. If no, what are the possible reason my serverboard and CPU have got burnt so that I can take better care of my other server.

Motherboard is Intel S1200V3RPS and cpu is Intel Xeon 1245v3. serverboard

enter image description here

closed as off-topic by Ramhound, n8te, PeterH, G-Man, Toto Oct 9 '18 at 12:02

This question appears to be off-topic. The users who voted to close gave this specific reason:

  • "This question is not about computer hardware or software, within the scope defined in the help center." – Ramhound, PeterH, Toto
If this question can be reworded to fit the rules in the help center, please edit the question.

  • 3
    It's impossible for us to answer a question, asking us if Intel will reject a warranty claim, we are not Intel. An answer that speculates if Intel will reject or honor a warranty claim would not be very helpful. You also already have an answer to that question, answered by somebody with physical access to the hardware, we don't have access to your hardware. Likewise, I don't see in the image, where your CPU socket would be burnt. However, I see a ton of dust, which indicates a lack of preventive maintenance on the hardware. – Ramhound Oct 5 '18 at 11:37
  • Thank you Ramhound for your input. I clean out my server once every 2 months. I'll do it more frequently for the other server now onwards. There appears to be discoloration on the cpu pins on the left side of the image and some discoloration on the top right side of motherboard socket. Are you suggesting those discoloration are simply dust? I was informed that they are burnt. – lonstud Oct 5 '18 at 11:59
  • I see caked on dust on the motherboard – Ramhound Oct 5 '18 at 12:07
  • Thank you Ramhound for making your point clear about dust on the motherboard. I'll appreaciate if you or somebody else can look past that and offer any advice about cpu pins and cpu socket on motherboard and what might have caused that discoloration on both. – lonstud Oct 5 '18 at 12:11
  • 1
    I would need physical access to the hardware to conclude the real reason for the discoloration. I can tell you, your socket isn't burned, because more than your memory wouldn't be working if that happened. The discoloration is also unrelated to the hardware failure. I would also say that the discoloration on the CPU could likely be polished out. You should ignore the discoloration on the socket. Any damage to the pins on a motherboard isn't under warranty typically. – Ramhound Oct 5 '18 at 15:15
2

No, it is definitely abnormal.

Airflow inside a server is so critical that many manufacturers have very specific instructions as to the placement of air dams, and put dummy spacers in vacant memory slots to make sure cooling air is properly directed. Please check the Technical Product Specifications and the configuration documentation to make sure the manufacturer's airflow design is not impeded.

High content (>90%) isopropyl alcohol is often used to clean contacts after properly disconnecting power sources and taking anti-static precautions. Also use demineralized or deionized water and lint-free wipes to avoid making foreign particle issues worse.

APC has many different devices; some are just switching power supplies and not a true Uninterruptible Power Supply.

  • Thank you K7AAY for your thoughts. I have sent both components for warranty but I shall try cleaning cpu as suggested by you if they reject warranty claim due to improper mainainence. APC device, which I am using, keeps my server running even during power outage for upto 45 minutes. I am assuming APC device with me is UPS. But since my other server is connected to same APC device, I have discarded possibility of external power surge for this issue. I'll also check Technical Product Specification for the motherboard and make sure my other server implements proper air dams. – lonstud Oct 5 '18 at 16:21
  • I grew up in Lightning Central; the southern gulf coast of Florida. lightningsafety.com/nlsi_info/lightningmaps/US_FD_Lightning.pdf The county computers (in the highest building in the county) were powered by a motor-generator set with a flywheel rotating on a wooden axle to completely isolate the computers from landline power, it was that bad. I have seen "UPS" systems fail in surge protection while continuing to switch power when landline power failed. Don't assume anything about your UPS; read its specs. – K7AAY Oct 5 '18 at 17:22
1

A modern CPU cannot burn as it would have thermal throttling/thermal shutdown.

A too-sudden overload would mean PSU problem or power surge or lightning strike (do you have UPS?). Only in the first case is your computer warranty engaged. In the other cases, it is rather your home insurance that is involved.

It is only your fault if you have removed a sticker saying "Warranty void if removed", or have over-clocked the CPU or have removed and incorrectly reseated your CPU.

In any case, you are rather in the hands of your manufacturer, since he can claim circumstances that are not covered by the warranty and you have no way of proving the contrary. Good luck.

  • I am using APC ups and usage of UPS does not go above 40%. I have connected two servers to a single UPS and other server is working fine. Only after memory channel A stopped working couple of days ago after working fine for 2 and a half years did I first remove and reseated my CPU. At that time also, I could see discoloration on both processor(top left side of image) and motherboard cpu pins(top right side of image). Do you think this discoloration happened due to reasons stated by you of top-sudden overload or can you think some other reasons which would help me in protecting other server. – lonstud Oct 5 '18 at 13:02
  • Whatever I say, it is your manufacturer that decides. I know of a case where a UPS didn't prevent a burn-out, so it depends on the UPS. There are two many variables here for me to give a better answer. – harrymc Oct 5 '18 at 13:05

Not the answer you're looking for? Browse other questions tagged or ask your own question.