Thanks, That worked. I think a push would make more sense as it would resolve many network topology challenges and enable a lot of flexibility.
Pete On Wed, Apr 27, 2016 at 11:24 PM, Stéphane Graber <[email protected]> wrote: > On Wed, Apr 27, 2016 at 11:01:42PM -0400, Pete Osborne wrote: > > Hi, > > > > I am trying to setup an image server for my development team. I'm trying > to > > push a container from one host to a remote so that other users can use > the > > image. I happen to be pushing over an openvpn connection. My local IP is > > 192.168.2.158 but my IP for the VPN is 172.16.0.5. > > > > sudo lxc remote add gloin 192.168.0.196 > > Certificate fingerprint: ok (y/n)? y > > Admin password for gloin: > > Client certificate stored at server: gloin > > > > sudo lxc publish development gloin: --alias=development > > error: Get > > > https://192.168.2.158:8443/1.0/images/2c00f2ae68df008fe1977325512153a45156baf3b9c70b7b8be4b10bfd004e77?secret=ef5283bbfde3a7a4ca32efd8c4af99631b4df3f6523d4bcf72ea935d9426c > > f97: Unable to connect to: 192.168.2.158:8443 > > > > I assume this is the correct approach but it appears that lxd is trying > to > > connect back on the wrong IP, is there a way to tell lxd to use a > different > > IP or interface? > > Easiest way to force it would be to set core.https_address to be > 172.16.0.5. > We have logic in a bunch of places to try all IPs one by one, but I > guess we've missed that particular code path. > > We also have a plan to allow for push rather than pull, which would fix > this in a possibly cleaner way too. > > > > > Thanks, > > Pete > > > -- > Stéphane Graber > Ubuntu developer > http://www.ubuntu.com > > _______________________________________________ > lxc-users mailing list > [email protected] > http://lists.linuxcontainers.org/listinfo/lxc-users >
_______________________________________________ lxc-users mailing list [email protected] http://lists.linuxcontainers.org/listinfo/lxc-users
