We all love Jenkins. It’s flexible, scalable, has unbelievable community support (more than 1010 plugins available) and is very easy to get started with. No wonder Jenkins is the CI/CD server used in at least 70% of IT and R&D organisations around the globe.
Once you start using Jenkins you quickly get hooked. It’s so easy to automate any development or system task, add a button and let your users push it whenever needed.
And your Jenkins instance begins to grow. Very soon you have views, nested views, pipeline views, dozens of plugins for every little thing, jobs for dev, jobs for QA, jobs for project managers, an ever growing bunch of slave nodes, you name it.Everyone in the organisation is using Jenkins, everyone falls in love with it as much as you originally did.
But then the day comes and your Jenkins that was so lively and fast when you installed it and ran your first job suddenly starts showing signs of fatigue. The UI takes time to load, jobs get stuck for unclear reasons, plugins conflict with one another, nodes get lost, disks fill up…
But users still expect the same level of service they grew accustomed to. You – the Jenkins admin – start getting emails and phone calls – sometimes in very uncomfortable hours.
And that’s when the need for monitoring becomes evident. You realise that your users would be happier and your own life easier if you could get alerts before things get actually broken. Alerts about disk space, memory consumption, lost build slaves and application errors.
But we all know that #monitoring sucks. Moreover – if you’re a Jenkins admin in an enterprise – chances are that you don’t have the access to existing monitoring infrastructure. Instead of being able to configure things the way you want them you’ll have to explain your needs to sysadmins and they configure it for you. And you don’t always know all your needs at once. So it can take a lot of time and pain to get that Jenkins instance monitoring right.
But what if I told you that there is a tool that is laughably easy to configure, can run on your commodity Windows machine and is already configured to check all the basic metrics of your Jenkins server. Doesn’t this sound like any enterprise Jenkins admin’s wet dream?
I think it does. And the great thing – now there is such a tool on the market!
A few days ago I had the pleasure to meet with Tamir Gefen and David Cohen of ALMToolBox. What they presented to me was the new version of ALM Vitality tool with added Jenkins monitoring support.
ALM Vitality was originally built for application-aware monitoring of ClearCase and ClearQuest and it seems to be doing a great job for those two as well. God knows I’ve spent a couple of years managing these monsters and their performance bottleneck analysis and resolution can be a major pain in the butt.
But for this session we were specifically looking at ALM Vitality for Jenkins and it really seems a nice little tool that hits a sweet spot for hassle-free Jenkins monitoring.
It is very easy to configure. Basically all it needs is an access to your Jenkins server and the credentials to the application itself. After that you’re up and running. Your Jenkins is monitored for disk space, application availability and stuck jobs. More checks (like lost slaves and memory consumption) are coming soon according to Tamir. Beside having access to your system state through a clean and simple web UI (how about putting a screen on the wall for all to see?) one can also configure email notifications for when things go wrong.
All in all, I certainly believe ALM Vitality is a great addition to a Jenkins admin toolbox.
You can watch a webcast on ALMToolox website for more details.
And if you decide to try the tool out – tell them I sent you 😉