Continuous Delivery requires the capability to quickly spot errors while executing a release pipeline and easily investigate them. Determining the root cause of a pipeline failure can be quite hard to work out by just looking at the symptoms. However, if the failed execution can be compared against a previous successful run, then the differences can pinpoint the cause quickly.
The release dashboard in vRealize Code Stream provides the capability to compare the results of multiple pipeline executions. By comparing a failed pipeline to a benchmarked one, the release team can quickly narrow down to the changes that broke the pipeline run. The comparison is applicable on finished, cancelled or failed pipeline executions. The example below shows a missing jar file in the pipeline execution that failed.
However, sometimes the error may not have been caused by a configuration or artifact change. Investigating these errors would require access to the log files on the host machine where the execution failed. vRealize Code Stream provides an easy and time-saving way to collect and browse logs, right in the UI, to troubleshoot a pipeline failure.
These logs are stored in the configured JFrog Artifactory instance that is embedded with vRealize Code Stream. The collection context is preserved by associating the logs with the corresponding execution information e.g. pipeline execution id, pipeline stage id, host and filepath. This speeds up access to the logs and make troubleshooting a lot easier.
Here are some additional resources to help you learn more about vRealize Code Stream