What can cause a lost carrier?

Question

Computer networks

Hello All!
I encountered such a problem, at indefinite intervals a lost carrier error occurs:

1921-grt01-nya#sh int gig 0/0
GigabitEthernet0/0 is up, line protocol is up
Hardware is CN Gigabit Ethernet, address is 84b8.0239.99a0 (bia 84b8.0239.99a0)
Description: pp21-p1
Internet address is xxxx /30
MTU 1500 bytes, BW 100000 Kbit/sec, DLY 100 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
Full Duplex, 100Mbps, media type is RJ45
output flow-control is unsupported, input flow-control is unsupported
ARP type: ARPA, ARP Timeout 04:00:00
Last input 00:00:14, output 00:00:00, output hang never
Last clearing of "show interface " counters 4d19h
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 26
Queueing strategy: fifo
Output queue: 0/40 (size/max)
5 minute input rate 135000 bits/sec, 15 packets/sec
5 minute output rate 8000 bits/sec, 11 packets/sec
5707773 packets input, 3204418960 bytes, 0 no buffer
Received 10201 broadcasts (0 IP multicasts)
0 runts, 0 giants, 0 throttles
78 input errors, 12 CRC, 27 frame, 0 overrun, 0 ignored
0 watchdog, 0 multicast, 0 pause input
3805486 packets output, 384755047 bytes, 0 underruns
0 output errors, 0 collisions, 0 interface resets
0 unknown protocol drops
0 babbles, 0 late collision, 0 deferred
7 lost carrier, 0 no carrier, 0 pause output
0 output buffer failures, 0 output buffers swapped out

This is data for three days, and input errors, CRC and frame, do not increase constantly, they occurred during one of the lost carrier events, but do not grow anymore. Lost carrier errors themselves occur without any explainable event.
The connection diagram to the ISP switch is as follows: edge router -> patch panel main cross -> patch panel inductory cross -> lightning protection -> ISP switch.
Of course, the cable has been tested, at least by a basic tester, unfortunately there is no one that can measure the signal transmission characteristics in the cable.
The Switch ISP is in an unheated cabinet, we have frosts down to -35, we have not yet been able to establish a correlation between temperature and the number of errors that occur, at first glance their number increases when the temperature drops below -15, but there is little input data.
There is a suspicion of the possible influence of passive lightning protection on the occurrence of this error, an event is planned to temporarily bypass it to eliminate this factor.
What else can be done before the complete re-installation of the entire cable path?
Any advice, and even more so your own experience of dealing with a lost carrier, is welcome.
I can’t find if CISCO has a debug for such events so that I can see in more detail what happens with this lost carrier error event?
UPD Question closed.
The fault was and is, the D-Link switch, as it turned out, the ISP found a bunch of tickets with the same problem, lost carrier, they were all solved by switching to no negotiation.
Lightning protection, FTP cables, low temperatures, all this was not to blame, D-Link knows something about standards that other companies do not.

Answer 1

Ahem. Electronics is guaranteed to work at temperatures from +5. Warm up the closet.

Answer 2

In principle, I don’t see anything critical if carrier loss occurs briefly for 10-20 seconds. Most likely this is due to static electricity that accumulates on the cable, especially in cold weather, when the humidity is very low, especially if there is light snow, especially during a thunderstorm.
Lightning protection, if it is normal, then there are no problems with it.
Actually two options.
1) Use an FTP cable with shield connectors and properly ground the switches. The option, I tell you, is not very ...
2) Use optics. Super! Moreover, it is now cheaper than dirt by 100 megabits. If it is not possible to install SFP modules, then simply install transmitters on both sides, for example, D-link DMC-300SC. The cable can be bought immediately terminated to the desired length.

Answer 3

UPD Question closed.
The fault was and is, the D-Link switch, as it turned out, the ISP found a bunch of tickets with the same problem, lost carrier, they were all solved by switching to no negotiation.
Lightning protection, FTP cables, low temperatures, all this was not to blame, D-Link knows something about standards that other companies do not.