Windows server 2016 system occasionally fails to receive network data and automatically recovers after a short period of time

Question

Problem:

Windows server 2016 system occasionally fails to receive network data and automatically recovers after a short period of time.

Environment:

indows 10 Kernel Version 14393 MP (32 procs) Free x64

Product: Server, suite: TerminalServer DataCenter SingleUserTS

Edition build lab: 14393.6897.amd64fre.rs1_release.240404-1613

Description：

Recently, I encountered a problem that a Windows server system occasionally has network anomalies. The problem scenario is that there is a server in our LAN environment that periodically sends ping detection data (ICMP data packets) to each terminal. If a terminal fails the ping detection for two consecutive detection cycles, it is considered that the terminal is disconnected. In the past month, the terminal has experienced ping detection failures many times, but other terminals in the LAN are normal. The specific phenomenon when the terminal fails to ping is that the system cannot receive network data, which lasts for about 20-120 seconds, and then the fault will automatically recover.

I have tried to use wireshark to capture packets both on the server 2016(IP: 168.160.0.31) with the network problem and the detection server(IP: 168.160.0.42). From the captured packet data, I can see that the server 2016 system can send data packets normally, but cannot receive the response data returned by the server. The packet capture on the server side shows that data sending and receiving are normal;

server 2016's packet data with wireshark:

the detection server's packet data with wireshark:

My server 2016 operating system runs in a virtual cloud desktop environment. I contacted the cloud desktop supplier for technical support. The feedback we received was that the network receiving cache of the operating system was full, resulting in the inability to receive data packets. However, when the system has a ping problem, there is no high network load, and I dumped the system dump file when the terminal network is abnormal. When analyzing the dump file, I found that when the problem occurred, the network card driver status was also normal. The receiving queue was checked through the ndis extension command and there was no abnormality. Therefore, I don’t know where the problem is. Can you help analyze the dump file to help locate the cause of the problem?

In the dump file, the network card in running status:

Check the receiving queue information of the network card driver as follows:

In summary, please help analyze the dump file to see if the system receive buffer is full and what the possible reasons are. I look forward to receiving your feedback.

Dump file download address: https://pan.baidu.com/s/1Y4nEwBnUrZiGxJABr7Lciw?pwd=tx1p