Hardware Failures in 2020 - Post yours!

andrewbedia

Active Member
Jan 11, 2013
667
207
43
Technically, I've missed the deadline, but in 2020 I had:

A Ryzen 9 3900X fail (stock, integrated memory controller died / failing all memtests)
Broke my chair
Broke an original Corsair M60 (all of the buttons suddenly stopped working)
 

mbosma

Member
Dec 4, 2018
69
39
18
We've had 3 of these Supermicro X11DPi-NT fail now, all seem to have issues in the vrm area.
Seems to be tied to hw version 1.21, replacement boards had version 2.0.

I still have a bunch of these with hw version 1.21 in production.
Let's hope they last longer than these ones did.View attachment 13350
To follow up on this topic, we had two more of these motherboards last month fail and decided it was enough.
After contacting Supermicro they stated with the following:
“The relevant ECO#21032 applied to All X11DPI-NT (Rev. 1.21) In Stock & New Production @2018-07-10, so enclosed items with invoice/release dated back in July 2018 or earlier can be accepted to CS from 1020. All items with invoice/release dated after July 2018 should be rejected. Revision 2.0 is not necessary as long as the boards have latest ECO implemented.”
It seems to affect both X11DPI-N and X11DPI-NT before July 2018.

For reference: The motherboard IPMI will power up and you'll be able to use the IPMI as usual the server simply won't power on. In my cases the server just crashed with an error in the IPMI log stating that CPU1 temp was critical and the system just wasn't able to boot with a cpu installed in that socket anymore.
I hope this helps someone in trying to debug this issue.
 
  • Like
Reactions: gb00s

edge

Active Member
Apr 22, 2013
111
34
28
I don't think this is a hardware failure so much as a software WAD (works as designed).

I moved in 2018 and in doing so I went from fully hardwired to having wifi bridges to my wife's home office and my home office. I always had great connectivity but my wife said hers stank like a dead skunk in your car engine compartment.

It came to a head a week ago when she was dropping out of her company board meeting (she is cfo). I did a speed test from my office and got a consistent 320 down and 30 up,

Fired up wifi analyzer in the room below her office and noticed roku spiking on the channel for her wifi bridge. Roku was broadcasting its wifi direct for casting and for its remote. Then it dawned on me- in our last house all the roku were hardwired.

It turns out hardwired roku just jam channel 36, but wifi connected roku wifi direct broadcast on the channel they use to connect to the wifi router and thus jam that channel. It makes sense as a roku only has one antenna so one channel...

Pulled a cable to the family room roku and suddenly the wifi in that room is great and the wifi bridge in my wife's home office is limited by our cable rate.

The works as designed of wifi direct as implemented by roku is a nasty invisible fail, hardware or not. Not so much about the one channel as it upping its broadcast to over power the access point signal.
 

cbarbie

New Member
Mar 10, 2011
18
5
3
US-VA
Had a ZyXEL GS1510-24 driving the 1st floor network for years with no problem. 1-2 times a month the entire 1st floor would drop off the network. Checked the switches on 1st & 2nd floor and they were running fine. Checked all the cables. Replaced lines between floors. This went on for a few months. It turned out the power supply was failing.
 

Fritz

Well-Known Member
Apr 6, 2015
2,298
567
113
67
I learned the hard way that 5TB 2.5 shucked HD's weren't meant to run 24/7.
 
  • Like
Reactions: Markess

Markess

Well-Known Member
May 19, 2018
631
302
63
This month, I had an Asus P9D-I...not necessarily fail outright. More like its having a nervous breakdown or being petulant. Random restarts, sometimes after running for hours or even days. I've tried changing out the usual suspects OS, CPU, RAM, cooling, power supply, disks and power/data cables. So, I can only assume that the motherboard only works until it just doesn't feel like it anymore.

Unless its a faulty power switch...oh heck....more testing.
 
Last edited:

Markess

Well-Known Member
May 19, 2018
631
302
63
I learned the hard way that 5TB 2.5 shucked HD's weren't meant to run 24/7.
I have to say, "Thank You @Fritz!" Your trailblazing experiences have saved me from making more than one mistake over the last few months. After you posted about your experience with these in another thread, I put the two I had back in their housings and use them for offline backup.
 
  • Like
Reactions: Fritz