Traditionally sysadmins used bash, WLST/Jython scripts to monitor weblogic based applications including Oracle SOA Suite and Oracle Service Bus. There are few disadvantages with this approach:
- Need to maintain multiple scripts to monitor a single domain.
- Sysadmins needs to be proficient in 1 or 2 scripting language. Developing script may take few hours to few days.
- Will not have access to historical monitoring data for doing trend analysis.
With the advent of modern monitoring tools like Prometheus and Grafana, above challenges are addressed. We can setup beautiful graphs and dashboards quickly and tools like Grafana also support alerting mechanism. Wavefront is an integrated solution that can store time series metrics, supports great visualizations, alerting, tracing and more. This blog post describes steps to integrate Weblogic monitoring with Wavefront.
Weblogic doesn’t expose time series metrics data directly. So, we need to install Jolokia on Weblogic for exposing JMX metrics as rest endpoints and use telegraf agent on VM for collecting these metrics and send to Wavefront. You can follow Wavefront documentation for installing telegraf and jolokia.
From version 220.127.116.11 onwards, SOA is providing IWS reports which contain lot of metrics about composite runtime. Unfortunately, these metrics are not available as JMX metrics. So, we have written couple of python/bash scripts to parse these reports and send them to Wavefront. Similarly, OSB metrics can also be sent to Wavefront. For more details and sample configurations refer to my github repository. I am thankful to my colleague Harisha for introducing me to Wavefront and helping me with setting up dashboards.
System, JVM and Thread Pool Metrics:
Datasource, JMS and SOA Metrics:
- 2023.05.23rabbitmqRabbitmq Connection Error: javax.net.ssl.SSLHandshakeException: Invalid ECDH ServerKeyExchange signature
- 2023.03.24rcuSOA Suite 18.104.22.168 installation: Got exception when auto configuring the schema component(s) with data obtained from shadow table
- 2023.03.09rabbitmqNot able to start Rabbitmq Cluster: Cannot declare a queue ‘~s’ on node ‘~s’: ~255p
- 2023.03.084lw commandsTelegraf agent not able to monitor zookeeper