Ever been frustrated skimming through logs across your app servers trying to find out why in the world your app keeps crashing? Not only are you already frustrated that your app is crashing, but now you have to dig deep into those log files trying to find any hint of a cause. And it gets compounded as you scale in the cloud.
What if you could get your logs combined into one easy to use interface? Better yet, what if you could collect metrics based on your logs or even identify security attacks? Sounds like a pretty cool idea doesnt it? Sure there are a ton of options out there like Splunk, LogEntries, or Loggly. Go ahead and add Oohlalog to your list of options.
I know, this is starting to sound like a bit of a plug, but hear me out. Logging is that one item a developer doesnt really care that much about. It's something for IT to worry about right? Logs are there in the event of an emergency right? Wrong! You can do some pretty cool stuff with logs. Use it for custom counter metrics, alerts to critical issues, or even leverage your logs to make business decisions by tracking signups, visits, ui interactions, and more. Some of you are saying, I've got google analytics for that, or New Relic. These are both great tools and very powerful. And in many cases they tend to do the job you want. but there are gaps in the data these services collect that could be better used by you, the developer, or you the CMO. Even your support staff could leverage your log files to help diagnose customer issues. It seriously is that easy.
Ok, now you are sorta intrigued right? But at the same time thinking, how involved is this going to be? Its super simple. Oohlalog leverages existing log interfaces for most languages, and even drills into common frameworks like Rails , or Grails (Log4j). It can be as simple as adding a plugin to your app and setting your API key.
When we set off to create a service like oohlalog, we set off as people in need of the tool ourselves. We made it easy to drill down into stacktraces and even link to real code in our github projects. We wanted real time charts tailored to matching log patterns to give us a more detailed view as to how our apps were behaving. The minute management and support staff started using it ( people with little or no programming expertise ) we realized, the potential for log data was much higher than we would have imagined ourselves. Imagine leaving a note on an error log with a comment trail, explaining the cause for the problem. Lets say that error crops up a year later. Immediately being able to see notes on the potential cause and communicating that to your future self sounds pretty sweet right?