Logging and Metrics infrastructure
Makefile generates a Docker Compose file which sets up the Kill Bill stack along with services for logging and recording metrics.
To launch the full stack, run:
Makefilewill look for the Docker container IP via
docker-machine ip default. Adapt to your setup if necessary.
docker-compose logswon’t work, because all logs are forwarded to Elasticsearch. The easiest way to find logs for a specific container is to search for the container ID in Kibana.
You need several GB of RAM to run the entire stack in the host: on Mac-OS, open Virtual Box and check the VM has 2GB or more of RAM (you need to poweroff the VM first).
The Elastic Stack setup (Elasticseach, Logstash, Kibana) is provided in the
It will give you the ability to analyze any data set by using the searching/aggregation capabilities of Elasticseach and the visualization power of Kibana.
Logstash listens on 1514 (Syslog protocol). For debugging purposes, the stdout plugin is enabled:
docker logs -f elk_logstash_1
The Kibana UI is available at
http://<IP_ADDRESS>:5601. Port 9200 for Elasticsearch is open for available to the host for Sense.
docker-compose.gi.yml file sets up Telegraf, InfluxDB, cAdvisor and Grafana.
The Grafana UI is available at
http://<IP_ADDRESS>:3000 (default credentials:
When running the full stack, Grafana uses the MariaDB container as the storage backend. To configure it (default database credentials
CREATE DATABASE grafana; USE grafana; CREATE TABLE `session` ( `key` CHAR(16) NOT NULL, `data` BLOB, `expiry` INT(11) UNSIGNED NOT NULL, PRIMARY KEY (`key`) ) ENGINE=MyISAM DEFAULT CHARSET=utf8;
You also need to add the InfluxDB datasources. Got to http://<IP_ADDRESS>:3000/datasources/new: