Skip to main content Skip to complementary content

Collecting microservice logs

When an artifact is deployed, every log from its execution is collected in Log Server and accessible from the GUI.

About this task

Default microservice logs are configured in the embedded log4j2.xml file. During deployment, the embedded log4j2.xml is updated to collect logs to Talend Cloud. Log collection is enabled by default.

When recovering artifacts, every log is collected again.

To configure a custom behavior for log collection, edit the org.talend.ipaas.rt.dsrunner.cfg file in the <RemoteEngineInstallationDirectory>/etc directory.

Procedure

  • To disable log collection, set the value of the ms.log.collection.enabled property to false.
  • To change the log4j2 configuration, edit the ms.custom.log4j2.conf property.
    During deployment, if the ms.custom.log4j2.conf has been configured, the microservice service will use a copy of this file as is, and will not configure it for collection to Talend Cloud.
  • To change the Talend LogServer host and port, edit the ms.log.collection.host and ms.log.collection.port.
    The default values are localhost and 7788.
  • To change the delay between Talend LogServer reconnections, edit the ms.log.collection.reconnection.delay property.
    The default value is 10000 (ms).
Troubleshooting:
  • Problem:

    Microservices, being always-on, continuously send logs to a socket on a 7788-based port. Each microservice deployment occupies one worker thread for logging. The thread pool size is defined as ms.worker.thread.number.

    Saturation can occur if a single microservice opens multiple connections, overwhelming the thread pool.

  • Recommended solution:

    Increase the thread pool size (ms.worker.thread.number) to accommodate your deployment needs. For example, setting the size to 50 can mitigate the problem effectively with minimal memory or performance impact.

Did this page help you?

If you find any issues with this page or its content – a typo, a missing step, or a technical error – please let us know!