In this article I will introduce the free IBM Bluemix Monitoring and Analytics service and show some of the available information.
Monitoring and Analytics
From the Bluemix Dashboard select one of your node applications and then “Add a new Service” – scroll down until you find the DevOps section and select Monitoring and Analysis.
There are two opens – Free and not Free, in this case we are going to select free.
Once added you have to restage the application
Once the application is restaged we can access the service from the dashboard – and initially there will be very little of interest/use.
But over time you will start to see information build up about how your application is performing within Bluemix. In this case I am using an example from http://pscportal.mybluemix.net – which if you click on the link you will find nothing – that is because I purposefully have nothing running on the root. What is fascinating is how the monitoring shows this and other interesting information.
Once you have added the service you can bind it to all your applications and it will monitor all of them separately.
Investigating a crash
Apparently the application crashed overnight
So I wondered what the monitoring said
Apparently the application went from zero CPU usage to DEAD right at 3:24am. But what is also really cool and apparent is that the application self healed and restarted itself – how freaking-cloud-cool is that !!
The application is never available – well that is as I would expect – the root has no node path responding to it. There are other routes within the application which work – but not that one.
Looking at the logs we can see that something happened but not much information as to what. I do not have any logging built into the application yet – but I will have to work on that.
Unfortunately the whole suite of monitoring is not available for XPages yet – I bound the service to my XPages application and all that was visible was he availability. This is still better than nothing.
The not free version
“You can use the Request Summary dashboard to identify the requests with the slowest response times and use the widgets to drill down and identify blockages. To open the Request Summary dashboard, open theMonitoring and Analytics service, click on the Performance tab and click Diagnose.”
Doing “DevOps” means constant monitoring of you application. The trick will be to take action based on the monitoring – we will see if we can do that and I will keep watching these logs over time to see what happens to the applications. Admittedly there are next to no users at all – but hey we will see.