IT4Change/Ocelot-Social
3
0
Fork 0
mirror of https://github.com/IT4Change/Ocelot-Social.git synced 2025-12-13 07:45:56 +00:00
Code Issues Packages Projects Releases Wiki Activity
58 Commits 5 Branches 2 Tags
Commit Graph

58 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Robert Schäfer
3eda8691bd
Typo 
@appinteractive you could use a translation service to easily stop typos
 2019-02-26 14:29:03 +01:00
Grzegorz Leoniec
086347e977
Merge pull request #11 from Human-Connection/improve-documentation 
Improved documentation for remote deployment
 2019-02-26 14:20:52 +01:00
Robert Schäfer
64bf1bb2a4
Update README.md 
Co-Authored-By: appinteractive <greg@app-interactive.de>
 2019-02-26 13:59:41 +01:00
Robert Schäfer
f13f1d39c9
Update README.md 
Co-Authored-By: appinteractive <greg@app-interactive.de>
 2019-02-26 13:59:24 +01:00
Robert Schäfer
32afabb1b2
Merge pull request #9 from Human-Connection/fix-secret-references 
Fix wrong secret references
 2019-02-26 13:54:58 +01:00
Robert Schäfer
9ca6f6da58
Merge pull request #10 from Human-Connection/fix-urls-for-staging 
Fix staging urls
 2019-02-26 13:54:29 +01:00
Grzegorz Leoniec
7c05aadcf7
Add some more information how to setup the configuration right before deploying to a remote cluster 2019-02-26 13:46:03 +01:00
Grzegorz Leoniec
1c20305c21
Fix staging urls 2019-02-26 13:36:43 +01:00
Grzegorz Leoniec
1e35278d8c
Fix wrong secret references 2019-02-26 13:33:34 +01:00
Robert Schäfer
da94d0b3a0
Merge pull request #8 from Human-Connection/fix-do-documentation 
Fix Digital Ocean Documentation
 2019-02-26 13:07:07 +01:00
Grzegorz Leoniec
f6c3841220
Fix Digital Ocean Documentation 
The order of commands was wrong as the dashboard was not correctly setup when calling `kubectl proxy`
 2019-02-26 13:04:59 +01:00
Grzegorz Leoniec
4d6c888253
Merge pull request #7 from Human-Connection/certmanager 
Certmanager
 2019-02-09 17:32:55 +01:00
Robert Schäfer
ec0b229e61 Tiny fix in README 2019-02-09 16:39:31 +01:00
Robert Schäfer
84da458ff1 Repeat HTTPS tutorial and update README 2019-02-09 16:04:22 +01:00
Robert Schäfer
9dfedfd317 ReadWriteOnce cannot scale the backend 
This kind of voume access apparently does not scale on a multi node
setup. Last time I tried, it worked, but last time was on just one node.
So I assume that's why I see the following error:
```
Multi-attach error for volume XY volume is already used by pod XZ
```
 2019-02-09 14:57:12 +01:00
Robert Schäfer
7e4a616a7a Git grep for quickstart and replace it 
@appinteractive I'm having serious issues with kubernetes and I'm not
able to connect with the cluster anymore. If your
```
kubectl get nodes
```
returns anything, could you try out the current deployment README?
 2019-02-08 15:48:20 +01:00
Robert Schäfer
6aea79608e Slight improvements of README 2019-02-08 14:15:16 +01:00
Robert Schäfer
85f9a5ef13 Save some money and decrease volume size 
We can resize it later
 2019-02-08 13:58:33 +01:00
Robert Schäfer
5d04c4ccc0 HTTPS works! 2019-02-08 11:03:18 +01:00
Robert Schäfer
9bb2361b18 Unverified documentation of TLS setup 2019-02-08 11:03:18 +01:00
Robert Schäfer
5d5db1ea47 Use single configmap for simplicity 2019-02-08 11:03:18 +01:00
Robert Schäfer
a25f806d0e Add missing kubernetes dashboard deployment 2019-02-08 11:03:18 +01:00
Robert Schäfer
066adb9b6e Bulk rename namespace staging->human-connection 2019-02-08 11:02:19 +01:00
Robert Schäfer
3bbe1f1471 Create ingress server and add documentation 2019-02-08 11:02:19 +01:00
Grzegorz Leoniec
15581de701
Merge pull request #4 from Human-Connection/add_db_migration_worker_deployment 
Add db migration worker deployment
 2019-02-06 22:51:51 +01:00
Robert Schäfer
e1e457abae Cleanly separate legacy migration part 
Use `kubectl patch -f ...`
 2019-02-05 18:04:56 +01:00
Robert Schäfer
99262a0d4d Fix README 2019-02-05 17:27:59 +01:00
Robert Schäfer
8c6bc72bd2 Update README 2019-02-04 22:05:24 +01:00
Robert Schäfer
1fbec5f525 Set NEO4J_URI to localhost 
Apparently, if you have a mult-container setup, the same pod is
accessible via localhost.
 2019-02-04 22:00:53 +01:00
Robert Schäfer
d876a3f442 Multiple container per pod setup 2019-02-04 20:44:07 +01:00
Robert Schäfer
81ae557be1 Create a persistent volume claim for uploads 2019-02-04 13:00:00 +01:00
Robert Schäfer
fb929da2cd Refactor db-migration-worker deployment 
Instead of creating a deployment with a replica set, we just create the
pod once. Also the pod should have everything in the future to run the
database migration. Ie. have `cypher-shell` to write directly to the
database in the current network. All required configuration is passed
manually to the `db-migration-worker`-pod directly. SSH-keys are copied
through a secrets file. This altogether made many configuration files
obsolete.
 2019-02-04 01:37:35 +01:00
Robert Schäfer
246a46c2e8 Remove obsolete volume in deployment 2019-02-03 14:01:19 +01:00
Robert Schäfer
672770ad18
Merge pull request #5 from Human-Connection/digital_ocean 
Fix certain configuration for Digital Ocean
 2019-02-02 21:22:47 +01:00
Robert Schäfer
abf623bd51 Rename .yml to .yaml files for consistency 
See:
https://stackoverflow.com/a/21059164

@appinteractive
 2019-02-02 18:47:17 +01:00
Robert Schäfer
0b075830bc Put many configuration files in one folder 2019-02-02 13:38:43 +01:00
Robert Schäfer
15f3915394 Adding dashboard deployment for digital ocean 2019-02-02 13:17:23 +01:00
Robert Schäfer
671826e060 Bundle all configuration in one folder staging/ 2019-02-02 12:40:29 +01:00
Robert Schäfer
af15ec6393 Fix certain configuration for Digital Ocean 2019-02-02 00:00:29 +01:00
Robert Schäfer
d8c23c3452 Reproduced the whole setup locally with success 2019-02-01 19:53:09 +01:00
Robert Schäfer
6fed4797ed Add commits to configuration to deploy :latest tag 
The recommended way to update a kubernetes deployment to a new image is
to change the image tag. However, our build server is configured to push
every commit of the `master` branch to docker hub to the respective
repository and tag `:latest`. So adding some configuration that can be
changed seems to be a trick to re-deploy the `:latest` image.

See here:
https://stackoverflow.com/a/51835397
 2019-02-01 19:23:00 +01:00
Robert Schäfer
ad3a974076 Base64 encode default secrets in template 
The JWT_SECRET is taken from the current default secret in
`Nitro-Backend`, the MONGODB_PASSWORD is just the label encoded.
 2019-02-01 01:56:12 +01:00
Robert Schäfer
9b5f88d7ac Kubernetes best practices: 
* put many configurations in one file
* expose services before you start pods

See: https://kubernetes.io/docs/concepts/configuration/overview/
 2019-02-01 01:25:05 +01:00
Robert Schäfer
e8b212f5cc Backend is accessible only from internal network 
I think this is better. For development it might be interesting to have
access to the backend but I think one should do development with
docker-compose not minikube and such.
 2019-02-01 01:09:41 +01:00
Robert Schäfer
7b981c06d0 Point web service to backend of internal network 2019-02-01 00:23:53 +01:00
Robert Schäfer
c6661def31 Improve README 2019-01-31 23:52:24 +01:00
Robert Schäfer
bbfe39e076 Use folders for groups of YAML files 
This keeps our configuration DRY and helps us to save keystrokes.
 2019-01-31 23:27:23 +01:00
Robert Schäfer
5cd0485117 Point the backend to the local neo4j service 2019-01-31 18:19:19 +01:00
Robert Schäfer
330fd9a8e9 Enable apoc to run neo4j import 2019-01-30 20:15:40 +01:00
Robert Schäfer
8a74f452a9 Create persistent volumes for minikube 
Running the import works in minikube!
 2019-01-30 19:34:45 +01:00