NOTE: The WebLogic Logging Exporter project has been archived. Now, the repository is read-only and all issues, pull requests, code, labels, milestones, and such, also have become read-only. Contributors with access to the repository only can fork or star the project.
Users are encouraged to use Fluentd or Logstash. If you use Fluentd to export logs to Elasticsearch or OpenSearch, then you may be interested in the WebLogic Kubernetes Operator documentation that describes how you can use Fluentd to export WebLogic logs to Elasticsearch.
The goal of this project is to provide an easy to configure, robust, and production-ready solution to access WLS log information through Elasticsearch and Kibana.
The WebLogic Logging Exporter adds a log event handler to WebLogic Server, such that WebLogic Server logs can be integrated into Elastic Stack in Kubernetes directly, by using the Elasticsearch REST API.
The current version of the WebLogic Logging Exporter is 1.0.1, which was released on Wednesday, January 27, 2021. This version supports pushing logs into Elasticsearch using the REST API.
The following features are planned for the next few releases:
You can download the WebLogic Logging Exporter already compiled for you from the releases page.
If you prefer, you can build the WebLogic Logging Exporter from the source code. To do this, you will need access to some WebLogic Server libraries. There are two ways to get these libraries:
You can use the Oracle Maven Synchronization plugin, which is included in your WebLogic Server installation, to install the necessary dependencies into your local Maven repository.
There are two steps:
push
goal to populate your local Maven repository from your WebLogic Server installation.To install the plugin, navigate to your WebLogic Server installation, then enter the commands (this example
assumes you installed WebLogic Server in /u01/wlshome
):
cd /u01/wlshome/oracle_common/plugins/maven/com/oracle/oracle-maven-sync/12.2.1
mvn install:install-file -DpomFile=oracle-maven-sync-12.2.1.pom -Dfile=oracle-maven-sync-12.2.1.jar
To populate your local Maven repository from your WebLogic Server installation, enter this command:
mvn com.oracle.maven:oracle-maven-sync:push -DoracleHome=/u01/wlshome
You can verify the dependencies were installed by looking in your local Maven repository which is
normally located at ~/.m2/repository/com/oracle/weblogic
.
Note: If you populated your local repository using the Oracle Maven Synchronization plugin, then this step is not required.
To access the Oracle Maven repository, refer to the documentation available here.
To build the WebLogic Logging Exporter, clone the project from GitHub and then build it with Maven:
git clone git@orahub.oraclecorp.com:oracle/wls-logging-exporter.git
mvn install
The weblogic-logging-exporter.jar
will be available under the target
directory.
This section outlines the steps that are required to add the WebLogic Logging Exporter to WebLogic Server.
Download or build the WebLogic Logging Exporter as described above.
Copy the weblogic-logging-exporter.jar
into a suitable location, e.g. into your domain directory.
Add a startup class to your domain configuration.
weblogic.logging.exporter.Startup
.You can verify this by checking for the update in your config.xml
which should be similar to this example:
<startup-class>
<name>LoggingExporterStartupClass</name>
<target>AdminServer</target>
<class-name>weblogic.logging.exporter.Startup</class-name>
</startup-class>
Add weblogic-logging-exporter.jar
and snakeyaml-1.27.jar
to your classpath.
This project requires snakeyaml
to parse the YAML configuration file. If you built the project locally,
you can find this JAR file in your local maven repository at ~/.m2/repository/org/yaml/snakeyaml/1.27/snakeyaml-1.27.jar
.
Otherwise, you can download it from Maven Central.
Place the file(s) in a suitable location, e.g. your domain directory.
Update the server classpath to include these file(s). This can be done by adding a statement to the end of your
setDomainEnv.sh
script in your domain's bin
directory as follows (this example assumes your domain
directory is /u01/base_domain
):
export CLASSPATH="/u01/base_domain/weblogic-logging-exporter.jar:/u01/base_domain/snakeyaml-1.27.jar:$CLASSPATH"
Create a configuration file for the WebLogic Logging Exporter.
There are two options currently - the version 1.x configuration, or the new version 2.x configuration - please
note that the 2.x configuration is alpha
and therefore subject to change as we get close to the 2.0 release.
a. Version 1.x configuration
Create a file named WebLogicLoggingExporter.yaml
in your domain's config
directory. You can copy the
sample provided in this project as a starting point. That sample
contains details of all of the available configuration options. A completed configuration file might look
like this:
publishHost: localhost
publishPort: 9200
domainUID: domain1
weblogicLoggingExporterEnabled: true
weblogicLoggingIndexName: domain1-wls
weblogicLoggingExporterSeverity: Notice
weblogicLoggingExporterBulkSize: 1
weblogicLoggingExporterFilters:
- filterExpression: 'severity > Warning'
Note that you must give a unique domainUID
to each domain. This value is used to filter logs by domain when you
send the logs from multiple domains to the same Elasticsearch server. If you are using the WebLogic Kubernetes
Operator, it is strongly recommended that you use the same domainUID
value that you use for the domain.
It is also strongly recommended that you consider using a different Elastcsearch index name for each domain.
b. Version 2.x configuration
If you prefer to place the configuration file in a different location, you can set the environment variable
WEBLOGIC_LOGGING_EXPORTER_CONFIG_FILE
to point to the location of the file.
If you want to write the JSON logs to a file instead of sending it elasticsearch directly use the following configuration file and adjust it to your needs. Make sure to rename it to WebLogicLoggingExporter.yaml.
Restart the servers to activate the changes. After restarting the servers, they will load the WebLogic Logging Exporter and start sending their logs to the specified Elasticsearch instance. You can then access them in Kibana as shown in the example below. You will need to create an index first and then go to the visualization page.
You can also use a curl command similar to the following example to verify that logs have been posted to Elasticsearch.
The default index name is wls
, and docs.count
should be greater than zero indicating that log entries
are being sent to Elasticsearch.
$ curl "localhost:9200/_cat/indices?v"
health status index uuid pri rep docs.count docs.deleted store.size pri.store.size
yellow open wls q4Q2v2dXTBOyYsHZMdDe3H 5 1 23 0 101kb 101kb
If you wish to test on your local machine, a sample is provided to run Elasticsearch and Kibana in Docker on your local machine.
This project welcomes contributions from the community. Before submitting a pull request, please review our contribution guide.
Please consult the security guide for our responsible security vulnerability disclosure process.
Copyright (c) 2017, 2021 Oracle and/or its affiliates.
Released under the Universal Permissive License v1.0 as shown at https://oss.oracle.com/licenses/upl/.