One of the many ways the IBM Runtimes Node.js team stays efficient and on top of all tasks is constantly maintaining and monitoring all our jobs and processes both within IBM and in the Node.js community. However, sometimes this can require quite a bit of manpower to ensure we are always aware of any changes or failures on our systems in the community. A team member would have to manually check each new build on the community for IBM specific failures and raise the corresponding work item, with enough information for us to deal with them

In an effort to increase efficiency, we’ve automated this process, taking the manual element for the most part, out of problem discovery. We have created scripts, which will query the community Jenkins and our RTC Dashboard at specified intervals, to monitor these community jobs, checking for any IBM Specific failures we don’t already have a work item for. When there is a failure, a work item containing enough information to be able to quickly make an initial analysis and decision on what to do (e.g extracts of the console log, the pull request or commit being tested etc) is generated. We are notified instantly by email as well as an updated dashboard and a team member can quickly respond to the issue.

As a result, we are now on top of our community processes and have cut down the time it takes us to respond to a failure greatly. We as a team, can now ensure our time is more efficiently spent solving problems and making quality improvements to the IBM SDK for Node.js



Screenshot of Dashboard, showing number of unresolved community issues.

 



Example of an automatically generated work item’s description

 

Join The Discussion

Your email address will not be published. Required fields are marked *