Docker Logging Driver
The article describes how to implement a unified logging system for your Docker containers. An application in a production environment requires to register certain events or problems during its runtime.
The old-fashioned way is to write these messages into a log file, but that inherits certain problems. Specifically, when we try to perform some analysis over the registers, or on the other hand, if the application has multiple instances running, the scenario becomes even more complex.
On Docker v1.6, the concept of logging drivers was introduced. The Docker engine is aware of the output interfaces that manage the application messages.
For Docker v1.8, we have implemented a native Fluentd Docker logging driver. Now, you are able to have a unified and structured logging system with the simplicity and high performance of Fluentd.
NOTE: Currently, the Fluentd logging driver doesn't support sub-second precision.
Getting Started
Using the Docker logging mechanism with Fluentd is a straightforward step. To get started, make sure you have the following prerequisites:
A basic understanding of Fluentd
A basic understanding of Docker
A basic understanding of Docker logging drivers
Docker v1.8+
For simplicity, the Fluentd is launched as a standard process, not as a container.
Please refer to Docker Logging via EFK (Elasticsearch + Fluentd + Kibana) Stack with Docker Compose for a fully containerized tutorial.
Step 1: Create the Fluentd Configuration File
The first step is to prepare Fluentd to listen for the messages coming from the Docker containers. For demonstration purposes, we will instruct Fluentd to write the messages to the standard output. Later, you will find how to accomplish the same by aggregating the logs into a MongoDB instance.
Create demo.conf with the following configuration:
Step 2: Start Fluentd
Now, start an instance of Fluentd like this:
On successful start, you should see the Fluentd startup logs:
Step 3: Start Docker Container with Fluentd Driver
By default, the Fluentd logging driver will try to find a local Fluentd instance (Step # 2) listening for connections on the TCP port 24224. Note that the container will not start if it cannot connect to the Fluentd instance.
The following command will run a base Ubuntu container and print some messages to the standard output:
Note that we have launched the container specifying the Fluentd logging driver i.e. --log-driver=fluentd.
Step 4: Confirm
Now, you should see the incoming messages from the container in Fluentd logs:
At this point, you will notice that the incoming messages are in JSON format, have a timestamp, are tagged with the container_id and contain general information from the source container along with the message.
Additional Step 1: Parse Log Message
The application log is stored in the "log" field in the record. You can parse this log before sending it to the destinations by using filter_parser.
Then you provide the log message with JSON format:
About --log-opt tag=..., please refer at Driver Options section.
Original Event (without filter plugin):
Filtered Event:
Additional Step 2: Concatenate Multiple Lines Log Messages
The application log is stored in the log field of the record. You can concatenate these logs by using fluent-plugin-concat filter before sending it to the destinations.
At first, you need to create custom docker image due to install the fluent-plugin-concat gem in the Fluentd container.
Create Dockerfile with the following content:
Build the custom image:
Then, create the configuration file demo.conf with the following content:
Launch the Fluentd container:
Then you provide the log message contains newlines:
Original Events (without filter plugin):
Filtered Events:
If the logs are typical stacktraces, consider using detect-exceptions plugin instead.
NOTE: For plugins with the simple file structure, such as fluent-plugin-concat, plugins directory can be used instead of creating custom docker image.
Prepare the plugins directory and copy the plugin file:
Launch the Fluentd container with plugins directory mounted:
Driver Options
The Fluentd Logging Driver supports following options through the --log-opt Docker command-line argument:
fluentd-addresstag
fluentd-address
fluentd-addressSpecifies the optional address (<ip>:<port>) for Fluentd.
Example:
tag
tagLog tags are a major requirement for Fluentd as they allow for identifying the source of incoming data and take routing decisions. By default, the Fluentd logging driver uses the container_id as a tag (64 character ID). You can change its value with the tag option like this:
Additionally, this option allows to specify some internal variables such as {{.ID}}, {{.FullID}} or {{.Name}} like this:
Development Environments
For a real-world use-case, you would want to use something other than the Fluentd standard output to store Docker container messages, such as Elasticsearch, MongoDB, HDFS, S3, Google Cloud Storage, and so on.
This document describes how to set up a multi-container logging environment via EFK (Elasticsearch, Fluentd, Kibana) with Docker Compose.
Production Environments
In a production environment, you must use one of the container orchestration tools. Currently, Kubernetes has better integration with Fluentd, and we're working on making better integrations with other tools as well.
If this article is incorrect or outdated, or omits critical information, please let us know. Fluentd is an open-source project under Cloud Native Computing Foundation (CNCF). All components are available under the Apache 2 License.
Last updated
Was this helpful?