I can not ping or SSH to hosts on my native community, together with my very own router, from my MBP. I am utilizing a wired connection in my CalDigit thunderbolt hub, and WiFi is disabled, however the issue persists no matter which one I take advantage of (or each). The whole lot else works: web, SSH’ing to hosts I run within the cloud, and many others.
➜ ~ ping 192.168.1.1
PING 192.168.1.1 (192.168.1.1): 56 information bytes
ping: sendto: No path to host
ping: sendto: No path to host
Request timeout for icmp_seq 0
ping: sendto: No path to host
Request timeout for icmp_seq 1
ping: sendto: No path to host
Request timeout for icmp_seq 2
Since every little thing else works, it appears unusual to wish to substantiate this, however the routing desk appears to be like fantastic.
➜ ~ netstat -nr
Routing tables
Web:
Vacation spot Gateway Flags Netif Expire
default 192.168.1.1 UGScg en11
127 127.0.0.1 UCS lo0
127.0.0.1 127.0.0.1 UH lo0
169.254 hyperlink#17 UCS en11 !
192.168.1 hyperlink#17 UCS en11 !
192.168.1.1/32 hyperlink#17 UCS en11 !
192.168.1.1 68:d7:9a:6a:39:fe UHLWIir en11 1134
192.168.1.2 dc:a6:32:38:8b:30 UHLWIi en11 1189
192.168.1.26 78:c8:81:81:5:a8 UHLWI en11 44
192.168.1.31 9c:6b:0:14:6a:7a UHLWI en11 230
192.168.1.255 ff:ff:ff:ff:ff:ff UHLWbI en11 !
224.0.0/4 hyperlink#17 UmCS en11 !
224.0.0.251 1:0:5e:0:0:fb UHmLWI en11
255.255.255.255/32 hyperlink#17 UCS en11 !
I’ve a few Linux containers, a take a look at server (.31), and a PiHole (.2) for native DNS. If I attempt to ping or SSH to both one, I get “no path to host.” I am flummoxed. It is a native LAN. There is no routing concerned. The PiHole is even linked in the identical change as my wired connection to my MBP. After configuring their networks with an IP handle and the fitting netmask, they need to simply arp one another and… speak. (And you may see within the routing desk that my Mac is arp’ing these hosts simply fantastic.)
One thing within the Mac’s IP stack is stopping both ping or SSH even getting off the machine. I’ve tried capturing the ping with tcpdump, however I do not see any packets. They don’t seem to be even getting put “on the wire” by my Mac. It isn’t even attempting.
On the Mac, I normally use an off-site DNS (Cloudflare or OpenDNS). Nevertheless, if I add the PiHole to my checklist of DNS resolvers on this machine, swiftly I can ping AND SSH to that host, however I nonetheless cannot ping the router or the opposite field. So macOS is making some exception when the host is a resolver, and I do not know what that is about.
Here is the true kicker: If I DON’T have the PiHole in my checklist of resolvers, I can not ping it, SSH to it, or use it for DNS lookup, HOWEVER, I CAN hit the PiHole admin internet app operating on it! So there’s some exception-to-the-exception for port 80.
Here is one other kicker: If I add the router (192.168.1.1) to my checklist of DNS resolvers, I can ping it too. Similar with the take a look at server.
One other: Though I can not ping or SSH to both Linux field, they can each ping the MBP simply fantastic.
I am not utilizing any VPN. (By no means have.)
I am not utilizing the Mac’s firewall.
What could possibly be inflicting my Mac to right away fail pings and SSH makes an attempt with “no path to host” on my native LAN, and never even attempt to put these packets on the wire (until the host is listed as a DNS resolver, apparently)?