pkoci Posted February 21 Share Posted February 21 Hi, I have recently changed my ISP which supports IPv6 and I have been assigned the /56 prefix by the ISP. I would like to make my NAS including docker containers fully accessible by both protocols. Is it possible for the containers to be able to reach the host via IPv6, the way they can on IPv4. I would almost say it doesn't work via IPv6. By the way, is there any way to prevent the NAS from getting a different IPv6 (not prefix) after each reboot outside of static network settings? What are your recommended practices for IPv6? Thank you Quote Link to comment
MAM59 Posted February 21 Share Posted February 21 >I have recently changed my ISP which supports IPv6 and I have been assigned the /56 prefix by the ISP. Strange, normally you get /64 (at home) or /48 (companies). /56 is in between, rather uncommon. >Is it possible for the containers to be able to reach the host via IPv6, the way they can on IPv4. Sure, even better because you don't need that evil NAT (hiding everything behind a single address). But, of course it is different, depending on WHAT you like to access from the outside. > By the way, is there any way to prevent the NAS from getting a different IPv6 (not prefix) after each reboot outside of static network settings? This should not happen unless you either have turned on "privacy extension" (which is hard to do on unraid, if even possible at all), or a strange dhcpv6 server (in your router???). Normally the local address is calculated based on the MAC address of the nic, so it should stay the same every reboot. (or are you using many NICs aggragated (bonded) ? Then the address will be one of them, but may change if a line goes down) Too bad, V6 contains several ways to "get an address", it depends on the settings of your isp, router and local servers. So it is hard to diagnose the origin without more details. >What are your recommended practices for IPv6? Use it, be happy 🙂 Get in touch with it and drop V4 someday... Quote Link to comment
pkoci Posted March 24 Author Share Posted March 24 Hi, I have been testing IPv6 for some time and found some issues. Macvlan docker containers cannot communicate with the host via IPv6 although this option is enabled in the docker settings. This works via IPv4. Additionally, once in a while ULA address is assigned to the docker network see below. Mar 24 18:12:10 UNRAID avahi-daemon[9726]: Registering new address record for fd11:8a45:f39:54d:2d0:XXXX:2cf2 on eth0.*. Mar 24 18:13:09 UNRAID avahi-daemon[9726]: Withdrawing address record for fd11:8a45:f39:54d:2d0:XXXX:2cf2 on eth0.Mar 24 18:12:10 UNRAID avahi-daemon[9726]: Registering new address record for fd11:8a45:f39:54d:2d0:XXXX:2cf2 on eth0.*. Mar 24 18:13:09 UNRAID avahi-daemon[9726]: Withdrawing address record for fd11:8a45:f39:54d:2d0:XXXX:2cf2 on eth0. Furthermore, I cannot visit the web UI via https using Ipv6 and myunraid.net (similar link like below). Again IPv4 works and even http via IPv6. https://2a03-xxxx-xxxx-xxxx-xxxx-xxxx-xxxx-xxxx.mycert.myunraid.net:64443/ Any ideas? Quote Link to comment
pkoci Posted March 26 Author Share Posted March 26 Any idea what's going on? Mar 26 13:48:09 UNRAID dhcpcd[22784]: eth0: adding address fd11:8a45:f39:54d:2d0:b4ff:fe02:2cf2/64 Mar 26 13:48:09 UNRAID avahi-daemon[23854]: Registering new address record for fd11:8a45:f39:54d:2d0:b4ff:fe02:2cf2 on eth0.*. Mar 26 13:48:09 UNRAID dhcpcd[22784]: eth0: adding route to fd11:8a45:f39:54d:6700::/64 Mar 26 13:48:15 UNRAID avahi-daemon[23854]: Withdrawing address record for fd11:8a45:f39:54d:2d0:b4ff:fe02:2cf2 on eth0. Mar 26 14:18:11 UNRAID dhcpcd[22784]: eth0: expired address fd11:8a45:f39:54d:2d0:b4ff:fe02:2cf2/64 Mar 26 14:18:11 UNRAID dhcpcd[22784]: eth0: part of a Router Advertisement expired Mar 26 14:18:11 UNRAID dhcpcd[22784]: eth0: deleting route to fd11:8a45:f39:54d:6700::/64 Mar 26 15:23:19 UNRAID dhcpcd[22784]: eth0: adding address fd11:8a45:f39:54d:2d0:b4ff:fe02:2cf2/64 Mar 26 15:23:19 UNRAID avahi-daemon[23854]: Registering new address record for fd11:8a45:f39:54d:2d0:b4ff:fe02:2cf2 on eth0.*. Mar 26 15:23:19 UNRAID dhcpcd[22784]: eth0: adding route to fd11:8a45:f39:54d:6700::/64 Mar 26 15:23:25 UNRAID avahi-daemon[23854]: Withdrawing address record for fd11:8a45:f39:54d:2d0:b4ff:fe02:2cf2 on eth0. Mar 26 15:35:43 UNRAID avahi-daemon[23854]: Registering new address record for fd11:8a45:f39:54d:2d0:b4ff:fe02:2cf2 on eth0.*. Mar 26 15:44:01 UNRAID avahi-daemon[23854]: Withdrawing address record for fd11:8a45:f39:54d:2d0:b4ff:fe02:2cf2 on eth0. Mar 26 15:44:06 UNRAID avahi-daemon[23854]: Registering new address record for fd11:8a45:f39:54d:2d0:b4ff:fe02:2cf2 on eth0.*. Mar 26 15:45:19 UNRAID avahi-daemon[23854]: Withdrawing address record for fd11:8a45:f39:54d:2d0:b4ff:fe02:2cf2 on eth0. Mar 26 16:15:18 UNRAID dhcpcd[22784]: eth0: pid 0 deleted address fd11:8a45:f39:54d:2d0:b4ff:fe02:2cf2/64 Mar 26 16:15:19 UNRAID dhcpcd[22784]: eth0: part of a Router Advertisement expired Mar 26 16:15:19 UNRAID dhcpcd[22784]: eth0: deleting route to fd11:8a45:f39:54d:6700::/64 Quote Link to comment
itimpi Posted March 26 Share Posted March 26 That is not atypical for a docker container continually crashing and attempting to restart. You should be able to see if this is the case by looking at the uptime of your containers on the Docket tab. Quote Link to comment
pkoci Posted April 3 Author Share Posted April 3 On 3/26/2024 at 11:08 PM, itimpi said: That is not atypical for a docker container continually crashing and attempting to restart. You should be able to see if this is the case by looking at the uptime of your containers on the Docket tab. I have checked the uptime of containers and found no crashing. Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.