I perf stuck after a while

[ashraffouda]

I have iperf server running on a machine for some reason I have to restart the iperf server for clients to be able to connect after keeping it running for a few days
btw, I'm running version 3.14
it gives operation timeout even if I'm running the client on the same machine that runs the server

~ # iperf --client 185.206.122.32 --port 300 --bandwidth 0 --interval 20 -R --json
{
	"start":	{
		"connected":	[],
		"version":	"iperf 3.14",
		"system_info":	"Linux zero-os 5.10.55-Zero-OS #2 SMP PREEMPT Thu Apr 6 12:09:33 UTC 2023 x86_64"
	},
	"intervals":	[],
	"end":	{
	},
	"error":	"unable to connect to server - server may have stopped running or use a different port, firewall issue, etc.: Operation timed out"
}

and this what happened after restarting it

~ # iperf --client 185.206.122.32 --port 300 --bandwidth 0 --interval 20 -R --json
{
	"start":	{
		"connected":	[{
				"socket":	5,
				"local_host":	"10.4.0.26",
				"local_port":	60292,
				"remote_host":	"185.206.122.32",
				"remote_port":	300
			}],
		"version":	"iperf 3.14",
		"system_info":	"Linux zero-os 5.10.55-Zero-OS #2 SMP PREEMPT Thu Apr 6 12:09:33 UTC 2023 x86_64",
		"timestamp":	{
			"time":	"Mon, 18 Aug 2025 13:58:14 UTC",
			"timesecs":	1755525494
		},
		"connecting_to":	{
			"host":	"185.206.122.32",
			"port":	300
		},
		"cookie":	"pgyo7keo3puqpszlura5hisje5gogbipeytl",
		"tcp_mss_default":	1448,
		"target_bitrate":	0,
		"fq_rate":	0,
		"sock_bufsize":	0,
		"sndbuf_actual":	16384,
		"rcvbuf_actual":	131072,
		"test_start":	{
			"protocol":	"TCP",
			"num_streams":	1,
			"blksize":	131072,
			"omit":	0,
			"duration":	10,
			"bytes":	0,
			"blocks":	0,
			"reverse":	1,
			"tos":	0,
			"target_bitrate":	0,
			"bidir":	0,
			"fqrate":	0
		}
	},
	"intervals":	[{
			"streams":	[{
					"socket":	5,
					"start":	0,
					"end":	10.000047,
					"seconds":	10.000046730041504,
					"bytes":	1136859552,
					"bits_per_second":	909483391.58033645,
					"omitted":	false,
					"sender":	false
				}],
			"sum":	{
				"start":	0,
				"end":	10.000047,
				"seconds":	10.000046730041504,
				"bytes":	1136859552,
				"bits_per_second":	909483391.58033645,
				"omitted":	false,
				"sender":	false
			}
		}],
	"end":	{
		"streams":	[{
				"sender":	{
					"socket":	5,
					"start":	0,
					"end":	10.003553,
					"seconds":	10.003553,
					"bytes":	1139933184,
					"bits_per_second":	911622647.67328167,
					"retransmits":	174,
					"max_snd_cwnd":	0,
					"max_snd_wnd":	0,
					"max_rtt":	0,
					"min_rtt":	0,
					"mean_rtt":	0,
					"sender":	false
				},
				"receiver":	{
					"socket":	5,
					"start":	0,
					"end":	10.000047,
					"seconds":	10.000047,
					"bytes":	1136859552,
					"bits_per_second":	909483367.028175,
					"sender":	false
				}
			}],
		"sum_sent":	{
			"start":	0,
			"end":	10.003553,
			"seconds":	10.003553,
			"bytes":	1139933184,
			"bits_per_second":	911622647.67328167,
			"retransmits":	174,
			"sender":	false
		},
		"sum_received":	{
			"start":	0,
			"end":	10.000047,
			"seconds":	10.000047,
			"bytes":	1136859552,
			"bits_per_second":	909483367.028175,
			"sender":	false
		},
		"cpu_utilization_percent":	{
			"host_total":	14.612327573918883,
			"host_user":	1.1938754967692382,
			"host_system":	13.41846207300936,
			"remote_total":	2.2722474443139133,
			"remote_user":	0.072733990325059755,
			"remote_system":	2.1994934611382968
		},
		"sender_tcp_congestion":	"cubic",
		"receiver_tcp_congestion":	"cubic"
	}
}

[davidBar-On]

Some questions for more information:

1. Are there any messages on the server side?
2. Can you run a client on the server side to make to make sure that the server is not responding and that this is not a network problem?
3. During the few days the server works, how much time it is running tests? What is the throughput of these tests? I am asking since there was a memory leak problem that impacted the server for long tests.
4. Can you try upgrading to newer iperf3 version?