Standalone Collector installation to retrieve metrics

If you don't use Kubernetes, or want to gather metrics from services not managed by Kubernetes, you can run the Collector as a standalone binary.

Create an API token

To interact with the Collector, you must create a service account. You must be a member of a team with the SysAdmin role to create a new service account. Chronosphere recommends creating a restricted service account with a write-only scope. Use the generated API token in your Kubernetes Secret to authenticate with the Collector.

⚠️

Store your API token in a secure location. If you lose your token, you must create a new service account.

Download the Collector binary and make it executable

Chronosphere supports Chronosphere Collector versions for a year from release. You can find a full list of versions and release dates in the Collector release notes located in the Chronosphere Observability Platform. To view the release notes, in the navigation menu select Help > Release notes, and then click the Collector tab.

Download one of the platform-specific binaries:
Modify the permissions on the binary to make it executable. For example, to change permissions on the linux-amd64 version of the binary:
```
chmod +x chronocollector-linux-amd64
```

Define the Collector configuration file

Download the example configuration file (opens in a new tab) and modify any configuration values based on your needs.

For applicable fields, you can set environment variables instead of specifying them in the configuration file. For example, to specify the listenAddress, set an environment variable named ${LISTEN_ADDRESS} with the value you want to use, such as:

listenAddress: "${LISTEN_ADDRESS:0.0.0.0:3029}"

The following list includes key fields for which you can specify values:

Backend: You must add a gateway backend to specify your Chronosphere instance.
```
backend:
   type: gateway
   gateway:
      address: {MY_COMPANY}.chronosphere.io:443
      insecure: false
      cert: ""
      certSkipVerify: false
```
Replace MY_COMPANY with your company name prefixed to your Chronosphere instance that ends in .chronosphere.io:443. For example, MY_COMPANY.chronosphere.io:443.
Global Labels: If you don't need to apply global labels to all metrics, remove the labels key. If you need global labels and are scraping metrics from a Prometheus endpoint, you can apply labels to each scrape target.

For example, to add host and rack to each metric:
```
static_configs:
   - targets: ['0.0.0.0:9100']
      labels:
         host: 'prod-server'
         rack: '14'
```
Each Prometheus scrape configuration type has a different way of specifying global labels. Refer to the Prometheus scrape configuration documentation (opens in a new tab) for more details.
Discovery: To scrape a static Prometheus endpoint, configure a scrape setting. Refer to the Prometheus service discovery documentation for more details.

Run the Collector

You can run the Collector standalone with a binary, with Docker, or as a Systemd service. To run the Collector, pass the configuration file and generated API token.

Replace the following values before running the startup commands:

MY_COMPANY: Your company name prefixed to your Chronosphere instance that ends in .chronosphere.io:443. For example, MY_COMPANY.chronosphere.io:443.
PATH_TO_FILE: The path to your Collector configuration file.
API_TOKEN: The API token generated from your service account.
VERSION: Version of the Collector that you want to run, prefixed with v.

Run standalone with a binary

GATEWAY_ADDRESS=MY_COMPANY.chronosphere.io:443 API_TOKEN=API_TOKEN PATH_TO_FILE/COLLECTOR_BINARY -f PATH_TO_FILE/config.yml

Run standalone with Docker

docker run -v $(pwd)/PATH_TO_FILE/config.yml:/etc/chronocollector/config.yml -e "GATEWAY_ADDRESS=MY_COMPANY.chronosphere.io:443" -e "API_TOKEN=API_TOKEN" gcr.io/chronosphereio/chronocollector:vVERSION

Run standalone as a service

You can install the Collector as a Systemd service on Linux systems:

Create a chronocollector user and group:

sudo groupadd -f chronocollector
sudo useradd -g chronocollector --no-create-home --shell /bin/false chronocollector

Change the ownership of the directory containing the Collector configuration file. The following command expects the Collector config.yml file in the /etc/chronocollector directory:
```
sudo chown -R chronocollector:chronocollector /etc/chronocollector
```

Copy the Collector binary to the /usr/bin directory and modify the permissions:

sudo cp ./chronocollector /usr/bin/
sudo chown chronocollector:chronocollector /usr/bin/chronocollector

In the /usr/lib/systemd/system directory, create and define a script such as chronocollector.service for the Collector service:

[Unit]
Description=Chronocollector
Documentation=https://docs.chronosphere.io/ingest/metrics-traces/collector
Wants=network-online.target
After=network-online.target

[Service]
User=chronocollector
Group=chronocollector
Type=simple
Restart=on-failure

Environment="GATEWAY_ADDRESS=MY_COMPANY.chronosphere.io:443"
Environment="API_TOKEN=API_TOKEN"

ExecStart=/usr/bin/chronocollector -f /etc/chronocollector/config.yml

[Install]
WantedBy=multi-user.target

Change the permissions of the chronocollector.service script you created:
```
sudo chmod 664 /usr/lib/systemd/system/chronocollector.service
```

Enable and verify the chronocollector service:

sudo systemctl daemon-reload
sudo systemctl start chronocollector
sudo systemctl status chronocollector
sudo systemctl enable chronocollector.service

Kubernetes (traces)Upgrade