Curl (56) Recv failure: Connection reset by peer [TCP Retransmission] - networking

Question

Network topology
PC1:
Env: Centos 8.2 Linux compute-31 4.18.0-240.22.1.el8_3.x86_64 #1 SMP Thu Apr 8 19:01:30 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux
IP: 10.0.0.31
ping is ok
[root#compute-31 ~]# ping 10.10.10.82
PING 10.10.10.82 (10.10.10.82) 56(84) bytes of data.
64 bytes from 10.10.10.82: icmp_seq=1 ttl=63 time=0.741 ms
From 10.0.0.1: icmp_seq=2 Redirect Host(New nexthop: 10.0.0.210)
64 bytes from 10.10.10.82: icmp_seq=2 ttl=63 time=0.404 ms
64 bytes from 10.10.10.82: icmp_seq=3 ttl=63 time=0.429 ms
64 bytes from 10.10.10.82: icmp_seq=4 ttl=63 time=0.389 ms
64 bytes from 10.10.10.82: icmp_seq=5 ttl=63 time=0.412 ms
^C
--- 10.10.10.82 ping statistics ---
5 packets transmitted, 5 received, 0% packet loss, time 133ms
rtt min/avg/max/mdev = 0.389/0.475/0.741/0.133 ms
telnet is ok
[root#10.0.0.31 ~]# telnet 10.10.10.82 9100
Trying 10.10.10.82...
Connected to 10.10.10.82.
Escape character is '^]'.
curl is not ok
[root#10.0.0.31 ~]# curl -v http://10.10.10.82:9100/metrics
* Trying 10.10.10.82...
* TCP_NODELAY set
* Connected to 10.10.10.82 (10.10.10.82) port 9100 (#0)
> GET /metrics HTTP/1.1
> Host: 10.10.10.82:9100
> User-Agent: curl/7.61.1
> Accept: */*
>
* Recv failure: Connection reset by peer
* Closing connection 0
curl: (56) Recv failure: Connection reset by peer
traceroute is ok
[root#10.0.0.31 ~]# traceroute -n -I 10.10.10.82
traceroute to 10.10.10.82 (10.10.10.82), 30 hops max, 60 byte packets
1 10.0.0.1 0.172 ms 0.151 ms 0.146 ms
2 10.0.0.210 0.209 ms 0.289 ms 0.268 ms
3 10.10.10.82 0.600 ms 0.582 ms 0.590 ms
PC2:
Env: Windows 10
IP: 10.0.10.6
curl is ok
C:\Users\Jeffery>curl 10.10.10.82:9100/metrics
# HELP go_gc_duration_seconds A summary of the pause duration of garbage collection cycles.
# TYPE go_gc_duration_seconds summary
go_gc_duration_seconds{quantile="0"} 7.228e-06
go_gc_duration_seconds{quantile="0.25"} 8.679e-06
go_gc_duration_seconds{quantile="0.5"} 1.0364e-05
go_gc_duration_seconds{quantile="0.75"} 1.2266e-05
go_gc_duration_seconds{quantile="1"} 0.000641901
go_gc_duration_seconds_sum 0.139410267
go_gc_duration_seconds_count 11244
TRY1:
In the PC1(10.0.0.31), Change the route path to 10.0.0.210 ->10.10.10.82 by add new route route add -net 10.10.10.0 netmask 255.255.255.0 gw 10.0.0.210, problem solved! But why? What happend to 10.0.0.1?
PS: The old path is 10.0.0.1 ->10.0.0.210 -> 10.10.10.82
TRY2:
I found somethings in tcpdump, but I didn't find what caused it.
[root#10.0.0.31 ~]# tcpdump -n -i eno2 host 10.0.0.31 and 10.10.10.82 -w tcpdum.10.0.0.31