Still working on ansible-deployment, I found that the dev-environment
is created using ubuntu-18.04 but production is using ubuntu-20.04
instead. In order to fix this I'd like to know what's the new policy
for images management because many containers in the lxd-hosts are
now also using ubuntu-20.04 instead of ubuntu-18.04 but I don't
see an "osgeo-ubuntu-20.04" image installed on them.
The osgeo7 lxd-host has only a osgeo-debian-10 image published.
The osgeo4 lxd-host has osgeo-ubuntu-18.04 and osgeo-debian-10 images.
Is it ok to make images available locally in production or are
we delegating images holding to the osgeo9 host instead ?
Still working on ansible-deployment, I found that the dev-environment is
created using ubuntu-18.04 but production is using ubuntu-20.04 instead.
In
order to fix this I'd like to know what's the new policy for images
management because many containers in the lxd-hosts are now also using
ubuntu-20.04 instead of ubuntu-18.04 but I don't see an "osgeo-ubuntu-
20.04" image installed on them.
The osgeo7 lxd-host has only a osgeo-debian-10 image published.
The osgeo4 lxd-host has osgeo-ubuntu-18.04 and osgeo-debian-10 images.
Is it ok to make images available locally in production or are we
delegating
images holding to the osgeo9 host instead ?
--strk;
[Regina Obe]
I have been building the newer images on osgeo9 and then copying as needed
to the production servers.
I was planning to have osgeo9 be the image authority.
That said there is no policy -- want to start one?
On Fri, Feb 18, 2022 at 01:51:45PM -0500, Regina Obe wrote:
> Still working on ansible-deployment, I found that the dev-environment is
> created using ubuntu-18.04 but production is using ubuntu-20.04 instead.
In
> order to fix this I'd like to know what's the new policy for images
> management because many containers in the lxd-hosts are now also using
> ubuntu-20.04 instead of ubuntu-18.04 but I don't see an "osgeo-ubuntu-
> 20.04" image installed on them.
>
> The osgeo7 lxd-host has only a osgeo-debian-10 image published.
> The osgeo4 lxd-host has osgeo-ubuntu-18.04 and osgeo-debian-10 images.
>
> Is it ok to make images available locally in production or are we
delegating
> images holding to the osgeo9 host instead ?
>
> --strk;
>
[Regina Obe]
I have been building the newer images on osgeo9 and then copying as needed
to the production servers.
I was planning to have osgeo9 be the image authority.
That said there is no policy -- want to start one?
The idea of deploying osgeo9 first, as an image repository, is appealing.
> > Is it ok to make images available locally in production or are we
> delegating
> > images holding to the osgeo9 host instead ?
> >
> > --strk;
> >
> [Regina Obe]
> I have been building the newer images on osgeo9 and then copying as
> needed to the production servers.
>
> I was planning to have osgeo9 be the image authority.
>
> That said there is no policy -- want to start one?
The idea of deploying osgeo9 first, as an image repository, is appealing.
--strk;
[Regina Obe]
Just a side note. The osgeo-ubuntu-20 image I did build with an ansible
script which is under the images role
I guess I even did define osgeo9 as having the images role and used that to
build and push the osgeo-ubuntu-20 to it.
That script installs the ldap auth, configures it, adds sudoers.
I think the debian one will be similar so I'll attempt to build a fresh
debian-11 images and a debian-12.