Happy New Year everyone!
It's been a while since our last release notes. However, you may have noticed some changes already. As said last time, our team was working really hard to deliver our new integrations framework. And the good news is: this is now released and fully operational!
So let's dive in and see what we delivered over the past few weeks.
The integrations page
40+ integrations are currently visible today.
There are 2 main steps to follow if you want to have a integration fully operational:
Setupstep: this is what you have to do to setup on your servers or services before integrating with Logmatic.io
Installstep: once the
setupis done, you now can - potentially configure and - install the integration. Installing an integration means Logmatic.io will auto-configure your platform so it provides for instance new dashboards, alerts, notification endpoints, etc... Not all the integrations are installable at the moment but we are working on it.
In the near future, we are planning to add some with third-party applications to make the logmatic.io experience even more seamless and valuable with your tools of choice.
Btw, there is a search bar to lookup your integration and if we can't find it you'll see a suggest integration button, please use it, we're always eager for your feedback.
Suggest a missing integration
Install it whenever you use a regular log shipper such as Rsyslog, Syslog-NG, NXLog... that defines
severity concepts the Infrastructure integration provides an overview dashboard over all your applicative infrastructure.
Infrastructure overview dashboard
The Source Code integration has been developed to work against most programming languages such as Java (Log4J2, SLF4J, etc...), PHP, NodeJS, PHP, etc...
The programming languages section
This integration has 2 roles:
- Normalising the notions of
severityto make different application written in different languages comparable
- Providing an overview dashboard so you can quickly troubleshoot all your services
A Java and a NodeJS application displayed in the Source Code overview dashboard
Whichever HTTP server, routing or proxy technologies or service you are using - Apache, Nginx, Haproxy, Varnish, AWS ELB, Heroku router, etc...- it comes with access logs. And these are useful to:
- Analyse the most important or simply slow URLs
- Monitor your API calls
- Geo locate your users
- Understand which OS, devices and browser are using your services
- Track errors and broken links
- See which bots are coming and what pages are they looking (SEO monitoring)
Web Server / Load balancer section in the integration page
The HTTP Web Server integration works accordingly with all the various technologies to provide:
- A normalised way to analyse access logs even if formats are different
- 3 dashboards to understand what goes on from a performance or an error tracking perspective
HTTP Web Access performance overview
You are a Heroku user? This integration has been designed for you!
In one click, Logmatic.io starts tokenising Heroku Routing logs, applicative logs and https://devcenter.heroku.com/articles/metrics to provide an efficient way to troubleshoot and see all the activity of your 10s or 100s of dynos.
Heroku overview dashboard
You can now rename any attribute of your incoming log entries trough processing pipelines. The new Rename Attribute processor takes list of attributes to rename to a target one. These can get preserved or erased and in case of conflicts the processor can also build arrays of values or simply take the last one.
Rename some servernames attributes into a standard one
By default, Logmatic.io displays the timeline and any associated analytic in the timezone of the user. So a user in Paris sees the timeline with a 6 hours difference from one located in New York.
From now on, if for some reasons you are in Paris but you want to see the logs as you were located in New York you can now change your timezone in your user profile as illustrated here:
Change your timezone
You can now ensure to clean up all the filters currently defined in one click. The only actual difference compared to before behaviour is the ability to clean the content of the search bar at the same time.
Clear the filters and search bar in one click
You can now display tagsets content as a regular column in the list events view:
Showing the "Response Time" tagset as a regular column
You may have noticed it already but we totally redesigned the homepage to lead you over the main concepts of the application.
The new homepage
A Attributes naming convention is at your disposal to help you decide how you should probably name your attributes while defining the JSON format you have some kind of control on or while defining your parsing rules over raw logs.
Integrations are actually already heavily depending on it as the one of their goals is to make different log format coming from similar technologies comparable (eg HTTP Web Access Integration).
It was decided to make this article public so companies with multiple teams collaborating together have a good starting point to work over various log formats.