cant ping machine on local /30 subnet.

azonsayoner · Yesterday 07:36:35

I have 2 machines connected directly via a single network cable.
Arch linux client – Arch linux server on /30 subnet . IP's 8.8.8.1 & 8.8.8.2
both machines have different ethenet NICS, and also have wifi NICs with a different ip
Both wifi and ethernet are managed by network manager on both machines.

Yesterday everything was working and I was able mount my files from my server onto client with nfs
Booted one machine into windows to do some bs, boot back and cant mount my files.
Cant ping machines on this local subnet.

PING 8.8.8.2 (8.8.8.2) 56(84) bytes of data.
From 8.8.8.1 icmp_seq=1 Destination Host Unreachable.....

tcpdump running on client can see arping from server. tcpdump on server cannot see arping from client

Ips are correct, NICs are correct and UP.

Ip neigh show :
8.8.8.2 dev enp0s31f6 FAILED

Booted back into windows, was able to ping the other machine through etherenet, so its not the cable.
tried chaining IPs, tried disabling iptables, tried flushing ARP cache.

Have scowered the internet, have asked chatgpt. Nothing to fix this magical problem yet.

Any help is really appreciated!

-thc · Yesterday 08:19:27

8.8.8.x belongs to Google. Please use 10.x.x.x instead (Yes - even if it's isolated).

Disable "Fast boot" on any Windows dual boot installation following the Wiki article.

azonsayoner · Yesterday 09:15:20

Thank you, already tried another ip, and just tried specifically 10.10.10.1 & 10.10.10.2 with now luck
Also boot win from different media

Lone_Wolf · Yesterday 10:48:42

Please post the outputs of

$ ping -l 3 -r <other-side-of-connection-as-ipv4-address>
$ ip -4 route

azonsayoner · Yesterday 10:59:57

hi thanks for helping!
here are the 2 outputs

PING 10.10.10.1 (10.10.10.1) 56(84) bytes of data.
From 10.10.10.2 icmp_seq=12 Destination Host Unreachable
From 10.10.10.2 icmp_seq=13 Destination Host Unreachable
From 10.10.10.2 icmp_seq=14 Destination Host Unreachable
From 10.10.10.2 icmp_seq=15 Destination Host Unreachable
^C
--- 10.10.10.1 ping statistics ---
16 packets transmitted, 0 received, +4 errors, 100% packet loss, time 12995ms
pipe 4

default via 88.0.0.1 dev wlp3s0 proto dhcp src 88.0.0.3 metric 600
10.10.10.0/30 dev eno1 proto kernel scope link src 10.10.10.2 metric 100
88.0.0.0/8 dev wlp3s0 proto kernel scope link src 88.0.0.3 metric 600

seth · Yesterday 12:33:17

Booted one machine into windows to do some bs, boot back and cant mount my files.

3rd link below. Mandatory.
Disable it (it's NOT the BIOS setting!) and reboot windows and linux twice for voodo reasons.

Ifff that's not it (there're many ffs a to that ifff), can we also get this from the other side as well as

sudo ethtool eno1

When connecting the wired network to some dhcp routerswitchmodem combo it works w/o problem?

azonsayoner · Today 02:47:37

Thank you, will do when I get home today.
Is it still mandatory considering I am booting both arch and windows from separate removable medias lol?

Last edited by azonsayoner (Today 02:48:48)

azonsayoner · Today 06:10:11

Hi, disabled on windows. reboot 2 times, no luck.
heres output of both systems with sudo ethtool eno1:

Client/Workstation:
Supported ports: [ TP ]
Supported link modes: 10baseT/Half 10baseT/Full
   100baseT/Half 100baseT/Full
   1000baseT/Full
Supported pause frame use: Symmetric Receive-only
Supports auto-negotiation: Yes
Supported FEC modes: Not reported
Advertised link modes: 10baseT/Half 10baseT/Full
   100baseT/Half 100baseT/Full
   1000baseT/Full
Advertised pause frame use: Symmetric Receive-only
Advertised auto-negotiation: Yes
Advertised FEC modes: Not reported
Link partner advertised link modes: 10baseT/Half 10baseT/Full
   100baseT/Half 100baseT/Full
   1000baseT/Full
Link partner advertised pause frame use: Symmetric Receive-only
Link partner advertised auto-negotiation: Yes
Link partner advertised FEC modes: Not reported
Speed: 1000Mb/s
Duplex: Full
Auto-negotiation: on
Port: Twisted Pair
PHYAD: 1
Transceiver: internal
MDI-X: on (auto)
Supports Wake-on: pumbg
Wake-on: g
Current message level: 0x00000007 (7)
drv probe link
Link detected: yes

Server:
Supported ports: [ TP ]
Supported link modes: 10baseT/Half 10baseT/Full
   100baseT/Half 100baseT/Full
   1000baseT/Full
Supported pause frame use: Symmetric Receive-only
Supports auto-negotiation: Yes
Supported FEC modes: Not reported
Advertised link modes: 10baseT/Half 10baseT/Full
   100baseT/Half 100baseT/Full
   1000baseT/Full
Advertised pause frame use: Symmetric Receive-only
Advertised auto-negotiation: Yes
Advertised FEC modes: Not reported
Link partner advertised link modes: 10baseT/Half 10baseT/Full
   100baseT/Half 100baseT/Full
   1000baseT/Full
Link partner advertised pause frame use: Symmetric Receive-only
Link partner advertised auto-negotiation: Yes
Link partner advertised FEC modes: Not reported
Speed: 1000Mb/s
Duplex: Full
Auto-negotiation: on
Port: Twisted Pair
PHYAD: 1
Transceiver: internal
MDI-X: off (auto)
Supports Wake-on: pumbg
Wake-on: g
Current message level: 0x00000007 (7)
drv probe link
Link detected: yes

seth · Today 06:10:42

The installation of the location is largely irrelevant - "removable" *might* make the firmware to reset the hardware but there's no guarantee for that to happen or work.
You cannot run two OS in parallel and context "windows to do some bs, boot back" is eerily suspicious.

Edit:
Please use [code][/code] tags. Edit your post in this regard.
Ftr. what are client and server, which is the system you ran windows on and what exactly is the other system?
Also what does the network config (ip a; ip r) on the other system look like now?

Have you tried to replug the cable (on either end)?

Last edited by seth (Today 06:17:48)

-thc · Today 06:46:57

From everything I've seen so far, there seems to be an elementary communication failure.

OSI level 1 seems to be O.K. (both NICs report a successful link at the same speed).
OSI level "2.5" (ARP) is not O.K. - the other host can't be reached because it's Ethernet address is unknown.

You can try to install "tcpdump" on both machines and monitor the traffic while you try to ping the "server".
This should show the ARP traffic and all possible failures.

Last edited by -thc (Today 06:47:29)

azonsayoner · Today 08:24:30

Thankyou both for your help.

More info on setup:
Workstation & Server both running arch. In process of migrating workstation from windows to arch, so I been booting between them frequently.
Server is normal arch server, running some docker containers for web/media servers, reverse proxy, file browser, fail2ban ect.
Definitely tried unplugging cable, tried another cable, tried blink test from both machines.

As said in my first post, tcp dump on client and arping from workstation shows traffic, while tcpdump on workstation arping from server shows nothing.

here are the outputs of ip a and ip r for both machines:

Workstation:

~$ ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host noprefixroute 
       valid_lft forever preferred_lft forever
2: wlp3s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
    link/ether 24:0a:64:49:af:b2 brd ff:ff:ff:ff:ff:ff
    altname wlx240a6449afb2
    inet 88.0.0.3/8 brd 88.255.255.255 scope global dynamic noprefixroute wlp3s0
       valid_lft 78215sec preferred_lft 78215sec
    inet6 fe80::e771:33f3:916b:770/64 scope link noprefixroute 
       valid_lft forever preferred_lft forever
3: eno1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
    link/ether ac:22:0b:c6:61:02 brd ff:ff:ff:ff:ff:ff
    altname enp0s25
    altname enxac220bc66102
    inet 10.10.10.2/30 brd 10.10.10.3 scope global noprefixroute eno1
       valid_lft forever preferred_lft forever
    inet6 fe80::ae22:bff:fec6:6102/64 scope link proto kernel_ll 
       valid_lft forever preferred_lft forever
~$ ip r
default via 88.0.0.1 dev wlp3s0 proto dhcp src 88.0.0.3 metric 600 
10.10.10.0/30 dev eno1 proto kernel scope link src 10.10.10.2 metric 100 
88.0.0.0/8 dev wlp3s0 proto kernel scope link src 88.0.0.3 metric 600

Server:

ip r
default via 88.0.0.1 dev wlp4s0 proto dhcp src 88.0.0.2 metric 600 
10.10.10.0/30 dev enp0s31f6 proto kernel scope link src 10.10.10.1 metric 100 
88.0.0.0/8 dev wlp4s0 proto kernel scope link src 88.0.0.2 metric 600 

ip a
2: enp0s31f6: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
    link/ether 54:e1:ad:20:b9:5b brd ff:ff:ff:ff:ff:ff
    altname enx54e1ad20b95b
    inet 10.10.10.1/30 brd 10.10.10.3 scope global noprefixroute enp0s31f6
       valid_lft forever preferred_lft forever
    inet6 fe80::bfc4:7f79:c961:a8cc/64 scope link noprefixroute 
       valid_lft forever preferred_lft forever

ps server has lots of docker networks creating interfaces that I havent been to bothered to organize well.

Thanks for the help!

-thc · Today 09:21:43

azonsayoner wrote:

As said in my first post, tcp dump on client and arping from workstation shows traffic, while tcpdump on workstation arping from server shows nothing.

Sorry - clearly missed that.

Such an "one-sided" communication error strongly suggests a netfilter or network misconfiguration issue on the server.

azonsayoner · Today 09:54:06

I agree.
Strange that my windows boot can ping the adress while server remains unchanged.
I have disabled iptables, on servr with no luck. Im not sure where to start.

-thc · Today 11:59:14

Yes - that is strange. But the alternative would be an even stranger netfilter or configuration issue on the client that prevents ARP packets from leaving the interface.

Can you try to boot the client from an Arch installation ISO, configure the LAN interface manually and try to ping the server?

Arch Linux

#1 Yesterday 07:36:35

cant ping machine on local /30 subnet.

#2 Yesterday 08:19:27

Re: cant ping machine on local /30 subnet.

#3 Yesterday 09:15:20

Re: cant ping machine on local /30 subnet.

#4 Yesterday 10:48:42

Re: cant ping machine on local /30 subnet.

#5 Yesterday 10:59:57

Re: cant ping machine on local /30 subnet.

#6 Yesterday 12:33:17

Re: cant ping machine on local /30 subnet.

#7 Today 02:47:37

Re: cant ping machine on local /30 subnet.

#8 Today 06:10:11

Re: cant ping machine on local /30 subnet.

#9 Today 06:10:42

Re: cant ping machine on local /30 subnet.

#10 Today 06:46:57

Re: cant ping machine on local /30 subnet.

#11 Today 08:24:30

Re: cant ping machine on local /30 subnet.

#12 Today 09:21:43

Re: cant ping machine on local /30 subnet.

#13 Today 09:54:06

Re: cant ping machine on local /30 subnet.

#14 Today 11:59:14

Re: cant ping machine on local /30 subnet.

Board footer