feat: add example how to use nginx to avoid overloading Artifactory b…

…y scraping data by Prometheus. (#79)

README.md

@@ -35,12 +35,23 @@ $ ./artifactory_exporter <flags>
 ### Docker
 
 Set the credentials in `env_file_name` and you can deploy this exporter using the [peimanja/artifactory_exporter](https://registry.hub.docker.com/r/peimanja/artifactory_exporter/) Docker image:
-:
 
 ```bash
 $ docker run --env-file=env_file_name -p 9531:9531 peimanja/artifactory_exporter:latest <flags>
 ```
 
+### Docker Compose
+
+Running the exporter against an Artifactory instance with millions of artifacts will cause performance issues in case Prometheus will scrape too often.
+
+To avoid such situations you can run nginx in front of the exporter and use it as a cache. The Artifactory responses will be cached in nginx and kept valid for `PROXY_CACHE_VALID` seconds. After that time any new request from Prometheus will re-request metrics from Artifactory and store again in the nginx cache.
+
+Set the credentials in an environment file as described in the *Docker* section and store the file as `.env` next to `docker-compose.yml` and run the following command:
+
+```bash
+docker-compose up -d
+```
+
 ## Install with Helm
 
 [Helm](https://helm.sh) must be installed to use the charts.

docker-compose.yml

@@ -0,0 +1,30 @@
+version: "3.8"
+
+services:
+  autoheal:
+    image: willfarrell/autoheal:latest
+    restart: always
+    volumes:
+      - '/var/run/docker.sock:/var/run/docker.sock'
+
+  proxy:
+    depends_on:
+      - exporter
+    image: decentralize/caching-proxy
+    restart: always
+    environment:
+      UPSTREAM: "http://exporter:9531"
+      MAX_SIZE: "100m"
+      PROXY_READ_TIMEOUT: "2400s"
+      PROXY_CACHE_VALID: "60s"
+    ports:
+      - 8088:80/tcp
+
+  exporter:
+    image: peimanja/artifactory_exporter
+    restart: always
+    ports:
+      - 9531:9531
+    command: --log.level=debug
+    env_file:
+      - .env