electron-rare/conversations
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

⚰️(storage) remove S3 media storage

Browse Source 
Unused for now and not clear if we will need it later.
This commit is contained in:
Quentin BEY
2025-08-26 17:00:25 +02:00
parent 86bcebadf8
commit 697b040a40
17 changed files with 2 additions and 322 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/architecture.md
-1
View File
@@ -10,5 +10,4 @@ flowchart TD
Back -- REST API --> Yserver
Back --> DB("Database (PostgreSQL)")
Back <--> Celery --> DB
Back ----> S3("Minio (S3)")
```
docs/env.md
-5
View File
@@ -19,11 +19,6 @@ These are the environment variables you can set for the `conversations-backend`
| DB_PORT | port of the database | 5432 |
| MEDIA_BASE_URL | | |
| STORAGES_STATICFILES_BACKEND | | whitenoise.storage.CompressedManifestStaticFilesStorage |
| AWS_S3_ENDPOINT_URL | S3 endpoint | |
| AWS_S3_ACCESS_KEY_ID | access id for s3 endpoint | |
| AWS_S3_SECRET_ACCESS_KEY | access key for s3 endpoint | |
| AWS_S3_REGION_NAME | region name for s3 endpoint | |
| AWS_STORAGE_BUCKET_NAME | bucket name for s3 endpoint | conversations-media-storage |
| DOCUMENT_IMAGE_MAX_SIZE | maximum size of document in bytes | 10485760 |
| LANGUAGE_CODE | default language | en-us |
| API_USERS_LIST_THROTTLE_RATE_SUSTAINED | throttle rate for api | 180/hour |
docs/examples/conversations.values.yaml
-19
View File
@@ -48,10 +48,6 @@ backend:
POSTGRES_USER: dinum
POSTGRES_PASSWORD: pass
REDIS_URL: redis://default:pass@redis-master:6379/1
AWS_S3_ENDPOINT_URL: http://minio.conversations.svc.cluster.local:9000
AWS_S3_ACCESS_KEY_ID: root
AWS_S3_SECRET_ACCESS_KEY: password
AWS_STORAGE_BUCKET_NAME: conversations-media-storage
STORAGES_STATICFILES_BACKEND: django.contrib.staticfiles.storage.StaticFilesStorage
migrate:
@@ -117,18 +113,3 @@ ingress:
ingressAdmin:
enabled: true
host: conversations.127.0.0.1.nip.io
ingressMedia:
enabled: true
host: conversations.127.0.0.1.nip.io
annotations:
nginx.ingress.kubernetes.io/auth-url: https://conversations.127.0.0.1.nip.io/api/v1.0/documents/media-auth/
nginx.ingress.kubernetes.io/auth-response-headers: "Authorization, X-Amz-Date, X-Amz-Content-SHA256"
nginx.ingress.kubernetes.io/upstream-vhost: minio.conversations.svc.cluster.local:9000
nginx.ingress.kubernetes.io/rewrite-target: /conversations-media-storage/$1
serviceMedia:
host: minio.conversations.svc.cluster.local
port: 9000
docs/installation.md
-20
View File
@@ -8,7 +8,6 @@ This document is a step-by-step guide that describes how to install Conversation
- an OIDC provider (if you don't have one, we provide an example)
- a PostgreSQL server (if you don't have one, we provide an example)
- a Memcached server (if you don't have one, we provide an example)
- a S3 bucket (if you don't have one, we provide an example)
### Test cluster
@@ -173,22 +172,6 @@ POSTGRES_USER: dinum
POSTGRES_PASSWORD: pass
```
### Find s3 bucket connection values
Conversations uses an s3 bucket to store documents, so if you have a provider obtain the necessary information to use it. If you don't, you can install a local minio testing environment as follow:
```
$ helm install minio oci://registry-1.docker.io/bitnamicharts/minio -f examples/minio.values.yaml
$ kubectl get po
NAME READY STATUS RESTARTS AGE
keycloak-0 1/1 Running 0 38m
keycloak-postgresql-0 1/1 Running 0 38m
minio-84f5c66895-bbhsk 1/1 Running 0 42s
minio-provisioning-2b5sq 0/1 Completed 0 42s
postgresql-0 1/1 Running 0 24m
redis-master-0 1/1 Running 0 10m
```
## Deployment
Now you are ready to deploy Conversations without AI. AI requires more dependencies (OpenAI API). To deploy Conversations you need to provide all previous information to the helm chart.
@@ -205,8 +188,6 @@ conversations-conversations-backend-migrate-c949s 0/1 Completed
conversations-conversations-frontend-6749f644f7-p5s42 1/1 Running 0 79s
keycloak-0 1/1 Running 0 48m
keycloak-postgresql-0 1/1 Running 0 48m
minio-84f5c66895-bbhsk 1/1 Running 0 10m
minio-provisioning-2b5sq 0/1 Completed 0 10m
postgresql-0 1/1 Running 0 34m
redis-master-0 1/1 Running 0 20m
```
@@ -220,7 +201,6 @@ $ kubectl get ingress
NAME CLASS HOSTS ADDRESS PORTS AGE
conversations-conversations <none> conversations.127.0.0.1.nip.io localhost 80, 443 114s
conversations-conversations-admin <none> conversations.127.0.0.1.nip.io localhost 80, 443 114s
conversations-conversations-media <none> conversations.127.0.0.1.nip.io localhost 80, 443 114s
conversations-conversations-ws <none> conversations.127.0.0.1.nip.io localhost 80, 443 114s
keycloak <none> keycloak.127.0.0.1.nip.io localhost 80 49m
```
docs/system-requirements.md
-2
View File
@@ -41,7 +41,6 @@ Production deployments differ significantly from development environments. The t
| PostgreSQL | **2 – 8 GB** | Higher `shared_buffers` and connection pooling for concurrent users |
| OIDC Provider (optional) | **Variable** | Any OIDC-compatible provider (Keycloak, Auth0, Azure AD, etc.) - external or self-hosted |
| Redis | **256 MB – 2 GB** | Session storage and caching; scales with active user sessions |
| Object Storage (optional) | **External or self-hosted** | Can use AWS S3, Azure Blob, Google Cloud Storage, or self-hosted MinIO |
| Django API (+ Celery) | **1 – 3 GB** | Production workloads with background tasks and higher concurrency |
| Static Files (Nginx) | **< 200 MB** | Serves Next.js build output and static assets; no development overhead |
| Nginx (Load Balancer) | **< 200 MB** | Reverse proxy, SSL termination, static file serving |
@@ -50,7 +49,6 @@ Production deployments differ significantly from development environments. The t
- **Frontend**: Uses pre-built Next.js static assets served by Nginx (no Node.js runtime needed)
- **Authentication**: Any OIDC-compatible provider can be used instead of self-hosted Keycloak
- **Object Storage**: External services (S3, Azure Blob) or self-hosted solutions (MinIO) are both viable
- **Database**: Consider PostgreSQL clustering or managed database services for high availability
- **Scaling**: Horizontal scaling is recommended for Django API service