June
a7772aa154
Having this option available allows for using a flake which isn't in (or
upwards of) the directory the command gets executed in and allows for
using remote flakes.
Also archive the flake to use first and then operate on the archive.
This allows for easily getting the deployment_configuration.json from
the archive and also ensures that once the archiving suceeds, there
shouldn't be issues with the flakes source anymore.
Since now the deployment_configuration.json will always be taken from
the root of the flakes archive and therefore from the root of the flakes
repo, this is a breaking change, since previously it was taken from the
current working directory.
The idea of archiving the flake first and operating on the archive comes
from bij:
|
||
---|---|---|
.forgejo/workflows | ||
src/infra_rebuild | ||
tests | ||
.gitignore | ||
CONTRIBUTING.md | ||
COPYING | ||
flake.lock | ||
flake.nix | ||
pyproject.toml | ||
README.md |
infra-rebuild
infra-rebuild mirrors nixos-rebuild closely in its command-line interface, but focuses on making infrastructure deployment more convenient. It is built for infrastructure defined in a flake, where each hosts configuration is represented by a different nixosConfiguration
in the flakes output.
For building the configuration of e.g. two hosts called web-01
and git
one can simply run:
infra-rebuild build web-01 git
For deploying the hosts - using the switch operation - it's as simple as:
infra-rebuild switch web-01 git
Here infra-rebuild will simply run nixos-rebuild switch
with the target-host
option set for each host.
Because infra-rebuild tries to be usable without any configuration, by default it queries the hosts FQDN from its nixosConfiguration
and uses that for the target host.
However to override aspects of the target host for specific or all hosts, infra-rebuild also accepts a deployment_configuration.json
as configuration.
Configuration
infra-rebuild accepts optional configuration in a deployment_configuration.json
present in the flake repos root.
The following keys are available to be set for configuring various aspects of deployment for specific or all hosts:
default.targetPort
: A default port to use for connecting to all host.default.targetUser
: A default user to use for connecting to all host.hosts.<host>.targetPort
: The port to use for connecting to<host>
.hosts.<host>.targetUser
: The user to use for connecting to<host>
.hosts.<host>.targetHostname
: The hostname to use for connecting to<host>
.
Example
An example deployment_configuration.json
might look like this then:
{
"default": {
"targetPort": 2222,
"targetUser": "deploy"
},
"hosts": {
"web-01": {
"targetUser": "web-deploy",
"targetHostname": "web-01-intern.infra.example",
"targetPort": 22
}
},
"git": {
"targetUser": "git-deploy"
}
}
Order of Precedence
The order of precedence from least to greatest for targetPort
and targetUser
is the following:
- system default
default.targetPort/User
hosts.<host>.targetPort/User
The order of precedence from least to greatest for targetHostname
is the following:
- FQDN from hosts nixosConfiguration
hosts.<host>.targetHostname
License
This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with this program. If not, see https://www.gnu.org/licenses/.