use nginx for handling and proxy request to API and Web UI #15
Conversation
📝 WalkthroughWalkthroughAdded an nginx reverse-proxy service to Docker Compose, removed host port mappings from Changes
Sequence DiagramsequenceDiagram
participant Client
participant nginx as nginx-proxy<br/>(port 80)
participant api as swetrix-api<br/>(port 5005)
participant app as swetrix<br/>(port 3000)
Client->>nginx: HTTP request (port 80)
rect rgba(200,220,240,0.5)
Note over nginx: Route decision
alt Path starts with /api
nginx->>api: proxy_pass to swetrix-api:5005 (HTTP/1.1)
api-->>nginx: response
else Other paths (root)
nginx->>app: proxy_pass to swetrix:3000 (HTTP/1.1)
app-->>nginx: response
end
end
nginx-->>Client: HTTP response
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Poem
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing touches
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
Fix all issues with AI Agents 🤖
In @nginx/config:
- Around line 6-12: The /api/ location block is missing several proxy headers
and operational directives; inside the location /api/ block add proxy_set_header
X-Forwarded-Proto $scheme and proxy_set_header Upgrade $http_upgrade along with
proxy_set_header Connection $connection_upgrade (and ensure you add the
top-level map $http_upgrade -> $connection_upgrade if WebSocket support is
required), and also add client_max_body_size with an appropriate limit plus
timeout directives proxy_connect_timeout, proxy_send_timeout, and
proxy_read_timeout to make the proxy resilient.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
nginx/config (1)
14-20: Add missing proxy headers and consider operational directives.The root location has the same gaps as the /api/ location:
Major concerns:
- X-Forwarded-Proto header missing (critical for HTTPS setups)
- Connection and Upgrade headers missing (required if the frontend uses WebSockets for real-time updates)
Recommended improvements:
- Add client_max_body_size directive
- Add timeout configurations for better reliability
🔎 Proposed enhancements
location / { proxy_pass http://swetrix:3000; proxy_http_version 1.1; proxy_set_header Host $host; proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; + proxy_set_header X-Forwarded-Proto $scheme; + proxy_set_header X-Forwarded-Host $host; + + # WebSocket support (if needed) + proxy_set_header Upgrade $http_upgrade; + proxy_set_header Connection $connection_upgrade; + + # Operational settings + client_max_body_size 10M; + proxy_connect_timeout 60s; + proxy_send_timeout 60s; + proxy_read_timeout 60s; }
🧹 Nitpick comments (2)
compose.yaml (2)
117-119: Replace deprecatedlinkswithdepends_on.The
linksdirective is deprecated in Docker Compose v2+. Usedepends_oninstead for service dependencies, which provides the same networking functionality with clearer semantics.🔎 Proposed refactor
nginx-proxy: image: nginx:1.29.4-alpine restart: always - links: - - "swetrix-api" - - "swetrix" + depends_on: + swetrix: + condition: service_healthy + swetrix-api: + condition: service_healthy ports: - "80:80"
114-123: Consider adding a health check for the nginx-proxy service.Adding a health check improves service orchestration and enables better monitoring of the reverse proxy.
🔎 Proposed enhancement
nginx-proxy: image: nginx:1.29.4-alpine restart: always links: - "swetrix-api" - "swetrix" ports: - "80:80" volumes: - ./nginx/config:/etc/nginx/conf.d/default.conf + healthcheck: + test: ["CMD-SHELL", "wget --no-verbose --tries=1 --spider http://localhost:80/ || exit 1"] + interval: 30s + timeout: 10s + retries: 3 + start_period: 10s
📜 Review details
Configuration used: defaults
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (2)
compose.yamlnginx/config
🔇 Additional comments (3)
nginx/config (1)
2-3: LGTM! Standard listen configuration.The IPv4 and IPv6 listen directives on port 80 are correctly configured for the single-domain architecture.
compose.yaml (2)
9-9: LGTM! API_URL correctly configured for single-domain setup.The hardcoded
/apivalue aligns with the nginx routing configuration that proxies/api/requests to the backend API service.
115-115: No action required. The image versionnginx:1.29.4-alpineis valid—nginx 1.29.4 (mainline) was released on December 9, 2025. If using a mainline release in production is a concern, consider pinning to the latest stable version (1.28.1 as of January 2026) instead, but the current version specification itself is not invalid.
| location /api/ { | ||
| proxy_pass http://swetrix-api:5005/; | ||
| proxy_http_version 1.1; | ||
| proxy_set_header Host $host; | ||
| proxy_set_header X-Real-IP $remote_addr; | ||
| proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; | ||
| } |
There was a problem hiding this comment.
Add missing proxy headers and consider operational directives.
The /api/ location is missing several important proxy headers and operational configurations:
Major concerns:
- X-Forwarded-Proto header is missing, which is critical for HTTPS awareness if SSL termination is added in the future (e.g., behind a load balancer or when adding TLS)
- Connection and Upgrade headers are missing, which are required for WebSocket support—verify whether the API uses WebSockets for real-time features
Recommended improvements:
- Add client_max_body_size directive to control upload limits (prevents large payload attacks)
- Add timeout configurations (proxy_connect_timeout, proxy_send_timeout, proxy_read_timeout) for better resilience
🔎 Proposed enhancements
location /api/ {
proxy_pass http://swetrix-api:5005/;
proxy_http_version 1.1;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+ proxy_set_header X-Forwarded-Proto $scheme;
+ proxy_set_header X-Forwarded-Host $host;
+
+ # WebSocket support (if needed)
+ proxy_set_header Upgrade $http_upgrade;
+ proxy_set_header Connection $connection_upgrade;
+
+ # Operational settings
+ client_max_body_size 10M;
+ proxy_connect_timeout 60s;
+ proxy_send_timeout 60s;
+ proxy_read_timeout 60s;
}If WebSocket support is needed, also add this map directive at the top level (outside server block):
map $http_upgrade $connection_upgrade {
default upgrade;
'' close;
}🤖 Prompt for AI Agents
In @nginx/config around lines 6-12, The /api/ location block is missing several
proxy headers and operational directives; inside the location /api/ block add
proxy_set_header X-Forwarded-Proto $scheme and proxy_set_header Upgrade
$http_upgrade along with proxy_set_header Connection $connection_upgrade (and
ensure you add the top-level map $http_upgrade -> $connection_upgrade if
WebSocket support is required), and also add client_max_body_size with an
appropriate limit plus timeout directives proxy_connect_timeout,
proxy_send_timeout, and proxy_read_timeout to make the proxy resilient.
|
wow that look fantastic! thanks for the PR, will be merged before the v5 release soon! |
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Fix all issues with AI agents
In `@compose.yaml`:
- Line 9: The frontend is configured with API_URL=/backend but nginx only
proxies location /api/ to the API service, causing requests to /backend to hit
the frontend; fix by either changing API_URL to "/api" in the compose
environment (update the API_URL value) or add a matching nginx location block
(add location /backend { proxy_pass http://swetrix-api:5005/; } in nginx/config)
so that requests to /backend are forwarded to swetrix-api:5005.
compose.yaml
Outdated
| - "80:3000" | ||
| environment: | ||
| - API_URL=${API_URL} | ||
| - API_URL=/backend |
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
cd /tmp && find . -name "compose.yaml" -o -name "compose.yml" 2>/dev/null | head -5Repository: Swetrix/selfhosting
Length of output: 45
🏁 Script executed:
# List structure to understand the repository layout
git ls-files | head -20Repository: Swetrix/selfhosting
Length of output: 287
🏁 Script executed:
# Find nginx configuration files
fd -t f -i 'nginx' | head -20
fd -t f 'config' | grep -i nginx
ls -la nginx/ 2>/dev/null || echo "No nginx directory at root"Repository: Swetrix/selfhosting
Length of output: 239
🏁 Script executed:
# Search for nginx configuration content
rg -n 'location|proxy_pass' --type-list | head -5
rg -n 'location|proxy_pass' 2>/dev/null | head -30Repository: Swetrix/selfhosting
Length of output: 882
🏁 Script executed:
cat -n nginx/configRepository: Swetrix/selfhosting
Length of output: 731
🏁 Script executed:
cat -n compose.yaml | grep -A 5 -B 5 "API_URL"Repository: Swetrix/selfhosting
Length of output: 423
Fix nginx configuration to proxy /backend path or change API_URL to /api.
The API_URL is hardcoded to /backend, but the nginx config only proxies /api/ to the API service. There is no location /backend block in nginx/config. This mismatch will cause API requests from the frontend to /backend to be caught by the catch-all location / directive and proxied to the frontend container itself, resulting in API failures.
Either update nginx/config to add location /backend { proxy_pass http://swetrix-api:5005/; } or change API_URL to /api.
🤖 Prompt for AI Agents
In `@compose.yaml` at line 9, The frontend is configured with API_URL=/backend but
nginx only proxies location /api/ to the API service, causing requests to
/backend to hit the frontend; fix by either changing API_URL to "/api" in the
compose environment (update the API_URL value) or add a matching nginx location
block (add location /backend { proxy_pass http://swetrix-api:5005/; } in
nginx/config) so that requests to /backend are forwarded to swetrix-api:5005.
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Fix all issues with AI agents
In `@compose.yaml`:
- Line 9: In the swetrix-fe container environment block in compose.yaml, replace
the unsupported environment key BASE_URL with API_URL so the frontend can locate
the backend (i.e., change the mapping from BASE_URL=${BASE_URL} to
API_URL=${BASE_URL} or set API_URL to the public Swetrix API base URL); update
the environment entry for the swetrix-fe service accordingly.
In `@README.md`:
- Around line 23-25: Replace the non-descriptive link text "here" that points to
https://docs.swetrix.com/selfhosting/configuring with a descriptive phrase like
"Self-hosting configuration guide" (or "Swetrix self-hosting configuration
guide") so the sentence "A table explaining what each value means can be found
here" becomes accessible and lint-compliant; update the markdown link text while
keeping the same URL and leave the surrounding sentences about BASE_URL and
docker unchanged to satisfy MD059.
🧹 Nitpick comments (1)
compose.yaml (1)
114-128: Consider mounting the nginx config read-only.
Prevents accidental writes in the container and signals intent.🔒 Suggested change
- - ./nginx/config:/etc/nginx/conf.d/default.conf + - ./nginx/config:/etc/nginx/conf.d/default.conf:ro
| - "80:3000" | ||
| environment: | ||
| - API_URL=${API_URL} | ||
| - BASE_URL=${BASE_URL} |
There was a problem hiding this comment.
🧩 Analysis chain
🌐 Web query:
swetrix-fe v4.1.0 BASE_URL environment variable support documentation
💡 Result:
Swetrix’s self-hosting docs for the frontend container do not document (or list) any BASE_URL environment variable. The only frontend env var currently documented is:
API_URL(URL of your self-hosted Swetrix API instance) [1]
So, if you’re looking for “swetrix-fe v4.1.0 BASE_URL support documentation”, it doesn’t appear to exist in the official Swetrix self-hosting documentation at this time; BASE_URL is not mentioned as a supported/configurable frontend env var there [1].
Source(s)
[1] Swetrix Docs — Self-hosting “Configuring” (Frontend env vars table) https://docs.swetrix.com/selfhosting/configuring
🌐 Web query:
swetrix-fe v4.1.0 release notes API_URL API endpoints configuration
💡 Result:
I couldn’t find any official Swetrix/swETRIX-frontend (“swetrix-fe”) v4.1.0 release notes; the public Swetrix Community Edition releases shown in the main repo go up to v4.0.6 (Oct 31, 2025), so v4.1.0 doesn’t appear to be a published OSS release/tag. [1]
API_URL (frontend) configuration
For self-hosting, the frontend container’s API_URL must point to the publicly reachable base URL of your swetrix-api backend, e.g. https://swetrix-api.mydomain.com (docs show default http://localhost:8080/). It must be reachable from users’ browsers (don’t use a Docker-internal hostname). [2]
API endpoints (what the frontend / clients call)
Swetrix’s APIs are documented as:
- Events ingestion API (pageviews/custom/heartbeat):
POST /log,POST /log/custom,POST /log/hb(cloud examples usehttps://api.swetrix.com/...; for self-hosting, these are on your own backend base URL). [3] - Statistics API:
GET /v1/log,GET /v1/log/sessions, etc. [4] - Admin API (project/org management):
GET /v1/project,POST /v1/project, etc. [5]
If you share where you saw “swetrix-fe v4.1.0” (Git tag, Docker tag, changelog link), I can trace the exact change that impacted API_URL/endpoints.
References:
[1] (github.com) [2] (docs.swetrix.com) [3] (docs.swetrix.com) [4] (docs.swetrix.com) [5] (docs.swetrix.com)
Citations:
- 1: https://github.com/Swetrix/swetrix?utm_source=openai
- 2: https://docs.swetrix.com/selfhosting/configuring?utm_source=openai
- 3: https://docs.swetrix.com/events-api?utm_source=openai
- 4: https://docs.swetrix.com/statistics-api?utm_source=openai
- 5: https://docs.swetrix.com/admin-api?utm_source=openai
Use API_URL instead of BASE_URL for the swetrix-fe container.
Official Swetrix self-hosting documentation only documents API_URL as the supported frontend environment variable. BASE_URL is not recognized and will cause the frontend to fail to resolve API endpoints. Change line 9 to API_URL=${BASE_URL} (or set the value directly to the publicly reachable base URL of your Swetrix API instance).
🤖 Prompt for AI Agents
In `@compose.yaml` at line 9, In the swetrix-fe container environment block in
compose.yaml, replace the unsupported environment key BASE_URL with API_URL so
the frontend can locate the backend (i.e., change the mapping from
BASE_URL=${BASE_URL} to API_URL=${BASE_URL} or set API_URL to the public Swetrix
API base URL); update the environment entry for the swetrix-fe service
accordingly.
| 3. Configure the environment variables for your Swetrix instance. It can be easily done by running `./configure.sh` script, which will ask you to provide the necessary values (including `BASE_URL`) and generate a `.env` file with them. A table explaining what each value means can be found [here](https://docs.swetrix.com/selfhosting/configuring). | ||
| 4. Run `docker compose up -d` to start the Swetrix services. | ||
| 5. After that, you will be able to access Swetrix web portal on the port you specified in `swetrix` category in `compose.yaml` (by default, it's set to `80`). | ||
| 5. After that, you will be able to access Swetrix web portal at the URL you set in `BASE_URL` (by default, `http://localhost`). |
There was a problem hiding this comment.
Use descriptive link text instead of “here.”
Improves accessibility and lint compliance while keeping the same destination.
✏️ Proposed doc tweak
-3. Configure the environment variables for your Swetrix instance. It can be easily done by running `./configure.sh` script, which will ask you to provide the necessary values (including `BASE_URL`) and generate a `.env` file with them. A table explaining what each value means can be found [here](https://docs.swetrix.com/selfhosting/configuring).
+3. Configure the environment variables for your Swetrix instance. It can be easily done by running `./configure.sh` script, which will ask you to provide the necessary values (including `BASE_URL`) and generate a `.env` file with them. See the [configuration variable reference](https://docs.swetrix.com/selfhosting/configuring).As per static analysis hints, this addresses MD059.
📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| 3. Configure the environment variables for your Swetrix instance. It can be easily done by running `./configure.sh` script, which will ask you to provide the necessary values (including `BASE_URL`) and generate a `.env` file with them. A table explaining what each value means can be found [here](https://docs.swetrix.com/selfhosting/configuring). | |
| 4. Run `docker compose up -d` to start the Swetrix services. | |
| 5. After that, you will be able to access Swetrix web portal on the port you specified in `swetrix` category in `compose.yaml` (by default, it's set to `80`). | |
| 5. After that, you will be able to access Swetrix web portal at the URL you set in `BASE_URL` (by default, `http://localhost`). | |
| 3. Configure the environment variables for your Swetrix instance. It can be easily done by running `./configure.sh` script, which will ask you to provide the necessary values (including `BASE_URL`) and generate a `.env` file with them. See the [configuration variable reference](https://docs.swetrix.com/selfhosting/configuring). | |
| 4. Run `docker compose up -d` to start the Swetrix services. | |
| 5. After that, you will be able to access Swetrix web portal at the URL you set in `BASE_URL` (by default, `http://localhost`). |
🧰 Tools
🪛 markdownlint-cli2 (0.18.1)
23-23: Link text should be descriptive
(MD059, descriptive-link-text)
🤖 Prompt for AI Agents
In `@README.md` around lines 23 - 25, Replace the non-descriptive link text "here"
that points to https://docs.swetrix.com/selfhosting/configuring with a
descriptive phrase like "Self-hosting configuration guide" (or "Swetrix
self-hosting configuration guide") so the sentence "A table explaining what each
value means can be found here" becomes accessible and lint-compliant; update the
markdown link text while keeping the same URL and leave the surrounding
sentences about BASE_URL and docker unchanged to satisfy MD059.
2 participants
This fix issue Swetrix/swetrix#460, by using nginx to proxy request then forward to either API or FE. It help Swetrix can be deploy on single domain with API is on
/api/instead of separate domainSummary by CodeRabbit
Chores
Documentation
✏️ Tip: You can customize this high-level summary in your review settings.