mirror of
https://github.com/Ocelot-Social-Community/Ocelot-Social.git
synced 2025-12-13 07:46:06 +00:00
44 lines
1.2 KiB
Markdown
44 lines
1.2 KiB
Markdown
# Maintenance mode
|
|
|
|
> Despite our best efforts, systems sometimes require downtime for a variety of reasons.
|
|
|
|
Quote from [here](https://www.nrmitchi.com/2017/11/easy-maintenance-mode-in-kubernetes/)
|
|
|
|
We use our maintenance mode for manual database backup and restore. Also we
|
|
bring the database into maintenance mode for manual database migrations.
|
|
|
|
## Deploy the service
|
|
|
|
We prepared sample configuration, so you can simply run:
|
|
```sh
|
|
# in folder deployment/
|
|
kubectl apply -f human-connection/maintenance
|
|
```
|
|
|
|
This will fire up a maintenance service.
|
|
|
|
## Bring application into maintenance mode
|
|
|
|
Now if you want to have a controlled downtime and you want to bring your
|
|
application into maintenance mode, you can edit your global ingress server.
|
|
|
|
E.g. in file `deployment/digital-ocean/https/ingress.yaml` change the following:
|
|
```yaml
|
|
...
|
|
|
|
- host: nitro-staging.human-connection.org
|
|
http:
|
|
paths:
|
|
- path: /
|
|
backend:
|
|
# serviceName: nitro-web
|
|
serviceName: maintenance
|
|
# servicePort: 3000
|
|
servicePort: 80
|
|
```
|
|
|
|
Then run ` kubectl apply -f deployment/digital-ocean/https/ingress.yaml`. If you
|
|
want to deactivate the maintenance server, just undo the edit and apply the
|
|
configuration again.
|
|
|