This repository configures an HTTPS reverse proxy with a server sidecar for collecting metrics.
- Nginx 1.11.10-alpine
- See Nginx documentation on HTTPS configuration and optimization
- The
ngx_http_stub_status_moduleis enabled athttp://<hostname>:<port>/nginx_statusfor basic status data used in metrics reporting
- Telegraf 1.2.1-alpine
- See InfluxData's documentation on the Nginx plugin for Nginx
- Outputs to InfluxDB to the default retention policy on a database "nginx"
- SSL certificate on the proxy host for HTTPS
- InfluxDB server for recording metrics
- Optional: retention policies and continuous queries created on a database "nginx"
- Syslog server for logging
Configure environment variables, or a .env:
| Variable | Description |
|---|---|
AGENT_HOSTNAME |
Name used in tag values for "host" in InfluxDB database |
AGENT_INTERVAL |
Data collection interval for Telegraf Nginx plugin |
INFLUXDB_HOST |
InfluxDB host and transport protocol (e.g., http://10.0.0.1) |
INFLUXDB_PORT |
InfluxDB port that server is listening on (e.g., 8086) |
SYSLOG_ADDRESS |
Syslog host, transport protocol, and port (e.g., tcp://10.0.0.2:514) |
SSL_CERTIFICATE |
Path to SSL certificate (e.g., /etc/ssl/certs/server.crt) |
SSL_CERTIFICATE_KEY |
Path to private key (e.g., /etc/ssl/private/server.key) |
SSL_TRUSTED_CERTIFICATE |
Path to trusted CA certificates PEM format (e.g., /etc/ssl/private/ca-certs.pem) |
DH_PARAMS |
Path to DH parameters for DHE ciphers (e.g., /etc/ssl/certs/dhparam.pem) |
To build and start containers:
docker-compose up -d
Telegraf communicates its metrics to an InfluxDB database and retention policy. If the database "nginx" does not exist, InfluxDB will create it, but will not configure retention policies or downsampling continuous queries.
If immediate database configuration is desired, before starting containers, open up an influx console on your InfluxDB host and input the following:
# create the nginx database with optional default retention policy
CREATE DATABASE "nginx" # WITH [DURATION <duration>] [REPLICATION <replication>] [NAME <retention-policy-name>]
# create additional retention policies
CREATE RETENTION POLICY <retention-policy-name> ON "nginx" WITH DURATION <duration> REPLICATION <replication> NAME <retention-policy-name>
# create continuous queries
CREATE CONTINUOUS QUERY <cq_name> ON "nginx" BEGIN SELECT mean(accepts) AS accepts, mean(active) AS active, mean(handled) AS handled, mean(reading) AS reading, mean(requests) AS requests, mean(waiting) AS waiting, mean(writing) AS writing INTO nginx.<target-retention-policy-name>.:MEASUREMENT FROM nginx.<source-retention-policy-name>.nginx GROUP BY time(<interval>), * END