Why does the Dell R330 server restart?

Question

Dell

There are several Dell R330 servers with an E3-1270v5 processor, all of them have the same problem, the server restarts periodically and spontaneously while cursing in the ipmi log:
57 | 03/31/2017 | 22:47:26 | Unknown #0x2e | | asserted
58 | 03/31/2017 | 22:47:26 | Processor #0x0d | Transition to non-recoverable | asserted
59 | 03/31/2017 | 22:47:26 | Unknown #0x28 | | asserted
5a | 03/31/2017 | 22:47:26 | Unknown #0x28 | | Asserted
5b | 03/31/2017 | 22:47:26 | Unknown #0x28 | | asserted
5c | 03/31/2017 | 22:47:26 | Unknown #0x28 | | asserted
5d | 03/31/2017 | 22:47:26 | Unknown #0x28 | | asserted
5e | 03/31/2017 | 22:47:26 | Unknown #0x28 | | asserted
5f | 03/31/2017 | 22:47:26 | Unknown #0x28 | | asserted
60 | 03/31/2017 | 22:47:26 | Unknown #0x28 | | asserted
61 | 03/31/2017 | 22:47:26 | Unknown #0x28 | | asserted
62 | 03/31/2017 | 22:47:26 | Unknown #0x28 | | asserted
63 | 03/31/2017 | 22:47:26 | Unknown #0x28 | | asserted
64 | 03/31/2017 | 22:47:26 | Unknown #0x28 | | asserted
65 | 03/31/2017 | 22:47:26 | Unknown #0x28 | | asserted
66 | 03/31/2017 | 22:47:26 | Unknown #0x28 | | asserted
67 | 03/31/2017 | 22:47:26 | Unknown #0x28 | | asserted
68 | 03/31/2017 | 22:47:26 | Unknown #0x28 | | asserted
69 | 03/31/2017 | 22:47:27 | Unknown #0x28 | | asserted
6a | 03/31/2017 | 22:47:27 | Unknown #0x28 | | asserted
6b | 03/31/2017 | 22:47:27 | Unknown #0x28 | | asserted
6c | 03/31/2017 | 22:47:27 | Unknown #0x28 | | asserted
6d | 03/31/2017 | 22:47:27 | Unknown #0x28 | | asserted
6e | 03/31/2017 | 22:47:27 | Unknown #0x28 | | asserted
6f | 03/31/2017 | 22:47:27 | Unknown #0x28 | | Asserted
70 | 03/31/2017 | 22:47:27 | Unknown #0x28 | | Asserted
Firmwares for BIOS, IDrac, RAID controller, etc. fresh
Reboot occurs without different kernel-panic, nothing is written to the crash, the behavior is twitchy in terms of power, but no power problems are noticed.
Has anyone experienced something similar?

Answer 1

Something is wrong with the processor IMHO:
58 | 03/31/2017 | 22:47:26 | Processor #0x0d | Transition to non-recoverable | Asserted
Probably overheating. Look at the detailed ipmi message.
People tried to find a solution, IMHO the maximum that turned out linux.debian.user.narkive.com/DuJ14QIu/machine-che...
Reboots because, apparently, this is how ipmi is configured to respond to failures. Another option - the people sin on a bug in the kernel, because of which the kernel does not see a hardware failure.
In I-net, a little on the topic "dell Unknown # 0x28".