Don’t let your Heroku apps fail silently….

Heroku error codes we’re coming to get you!!!!

Logentries has come out of private beta on Heroku and is now available to all as a Heroku add on. Logentries provides log managenemt as a service with in built Heroku intelligence, so your applications don’t fail silently.

Heroku App Management Tips

“Ok, so why should I care?”

Well, with the standard Heroku logging you get access to only the last 500 events, which is not great if you need to go and look at an issue that a customer reported over night for example.

However, thanks to Logplex (Heroku’s logging infrastructure – which, by the way, has been open sourced) Logentries can gather and store all your Heroku log data for as long as you like – this includes your application, API and Heroku system logs – so you can see exactly what’s going on across your entire stack at any point in time.

Thanks to Logplex, this is all automatically configured with a single click (i.e. selecting the add on 🙂 ). At that point your log events are immediately sent to a newly created Logentries account. Here you can keep your logs for as long as you want (by the way we are the only logging add on with unlimited storage) – this is useful if you need to store your logs indefinitely, or do postmortem analysis of system issues.

Once you’ve selected the add on you can make use of all the Logentries features like visualisation, tagging, searching etc. to get some useful information out of your logs.

Ok that’s pretty useful – I might configure Logentries in case I need to do some debugging…

That’s not the best bit. As well as the standard log management features, Logentries comes with in-built Heroku system intelligence.

“Eh, so what’s that I hear you ask?”

So we have preconfigured the Logentries add on to highlight any Heroku error codes that appear in your logs. This allows you to easily identify these in your log data when they occur.

“That’s pretty neat, right?”

Well, we’ve gone one better, thanks to some input from the guys at Heroku, we automatically notify you if any of, what are deemed, the more serious errors occur (e.g. app crash, memory quota exceeded, failed requests…)

Because we have notifications preconfigured – you don’t need to do anything – it’s setup and ready to go out of the box. Emails are sent to you if any serious errors are happening under the hood of your Heroku applications so you no longer fail silently.

Note, because we capture events at the syslog level, we also get system events produced by the Heroku infrastructure, which you will not get if you are using exception tracking software that require ruby gems to grab exceptions from your application code. That way we get you the whole picture at the different levels of the stack.

That means if your application is struggling under load, dropping requests or using too much memory you get notified immediately. And when you log in to your Logentries account you can clearly see whats going on as these events are automatically highlighted.

Heroku App Management Data Logging

Furthermore, if there are issues on the Heroku side (e.g. maintenance, platform issues) you can also get notified and keep an eye on their status site!

Without notifications you may not realize your application is down unless you go and check the logs or (more likely) until your customers start screaming at you… using Logentries you get immediate notifications if anything fishy is going on (e.g. requests start getting dropped) so you can react before it becomes a major problem (e.g. an app crash).

“Wow now I can sleep again at night – thanks Logentries!!!”

No problem… just don’t fail silently 😉 … check out our add-on now

Heroku with Logentries
Posted in Heroku

Leave a Reply