0

My scenario is the following: I have a postgres 12 installed on ubuntu 20.04 server and a docker container running in another host with ipv6 enabled.

I can't reach postgres server from inside docker container using ipv6.

these commands works from within the container:

$ ping POSTGRES_SERVER_IPv4
$ ping POSTGRES_SERVER_IPv6
$ telnet POSTGRES_SERVER_IPv4 5432

but this does not work from inside the container (but it works from the docker host):

$ telnet POSTGRES_SERVER_IPv6 5432

i've already set listen_address = '*' and "host all all md5" in postgres pg_hba.conf.

EDIT:

i've realized that my network adapter is using two IPv6, the static one that I defined in netplan and another one that is within my network range, but I didn't assigned it.

jguilhermeam
  • 121
  • 1
  • 1
  • 10

1 Answers1

2

In your netplan configuration, try to add:

accept-ra: no

example:

network:
  version: 2
  ethernets:
    ens192:
      accept-ra: no
      addresses: ...
Gerson Barreiros
  • 36
  • 3
  • thanks, it solved. apparently my adapter was getting another ipv6 from my RA, but the curious thing is that the problem was only ocurring inside docker container or using putty (ssh). – jguilhermeam Aug 20 '20 at 21:29