HPE 1920 sometimes stops responding to ping requests

[DSV12]

G'day 4all!

I have exactly the same problem as described here: https://community.hpe.com/t5/Web-and-Unmanaged/HP-1920-temporary-no-reply-to-ICMP-request/td-p/6873613 (the solution was not found there).

HPE1920 (PoE) with the latest version of the firmware - 5.20.R1117, two VLAN's. ip binded only to the management VLAN 1 interface.

I used PingInfoView from NirSoft (timeout 1000ms, 5 seconds between the pings) for pinging switches.

What's happening: HPE 1920 (8 and 24 ports models) sometimes stops responding to ping requests, and usually it is lost several consecutive replies: from 5 to 10! (this means that the switch was not replied within almost a full minute!). Transit traffic is NOT lost, only the replies from the switch disappear:

09.08.2017 10:01:46,sw-hp1920.domain.com,192.168.169.211,Timeout
09.08.2017 10:01:52,sw-hp1920.domain.com,192.168.169.211,Timeout
09.08.2017 10:01:58,sw-hp1920.domain.com,192.168.169.211,Timeout
09.08.2017 10:02:04,sw-hp1920.domain.com,192.168.169.211,Timeout
09.08.2017 10:02:10,sw-hp1920.domain.com,192.168.169.211,Timeout
09.08.2017 10:02:16,sw-hp1920.domain.com,192.168.169.211,Timeout
09.08.2017 10:02:22,sw-hp1920.domain.com,192.168.169.211,Timeout
09.08.2017 10:02:28,sw-hp1920.domain.com,192.168.169.211,Timeout
09.08.2017 10:02:34,sw-hp1920.domain.com,192.168.169.211,Timeout
09.08.2017 10:02:40,sw-hp1920.domain.com,192.168.169.211,Timeout

According to my observations, the loss of responses from the switches does not directly depend on the load on the network, they happened both during working hours and at night (eg at 2:00 AM).

Now we have three HPE1920 switches, and with all there is such a problem. Soon we will have several dozen HP 1920, I would like to understand and eliminate the reason for their such behavior.

TIA,

Sergei