Let’s be frank: Sometimes (read: very often) nothing beats the command line with its set of classic utilities. One of those classics is the Unix command tail, which many developers and IT professionals use on a daily basis. When invoked with the command line option
-f (follow), it allows you to monitor a file. You can watch live as lines are being added. It is a perfect tool to get a real-time view into what is written to a log file, making it one of the greatest little utilities for debugging and troubleshooting a program or system. (On Windows, the PowerShell command Get-Content provides similar functionality when invoked with the parameters -Tail and -Wait.)
Back in the good old days when most things a developer or sysadmin had to worry about happened locally on one or a few remote machines, good old tail was simple and efficient. It was a very common thing to see a small (or large) collection of terminal windows lined up on the screens of IT pros, each running a single
tail-f session monitoring different files, often remotely via a ssh connection. Because tail is such a handy tool many enhanced versions were developed over time in order to overcome some of the original’s limitations. There are variations out there that can monitor multiple files at once, provide filtering and color highlighting of text patterns, or display the text output on the desktop background.
And of course you could always pipe the output from tail into other commands to process it further.
A Thing of the Past?
However, in this day and age tail has lost a lot of its appeal. The era of very affordable clouds is here, and even small companies run distributed systems that consist of many components—the emphasis here is on many. Each server likely runs multiple VMs or containers, each with one or more applications. It is impossible to log into each and every system to fire up a tail session for each and every log file, let alone the fact that you wouldn’t be able to look at all of them in parallel anyway. (Well, maybe unless you are Argus Panoptes …but then you would have other things to do.) In short, when you have diverse services running on elastic servers in the cloud, you can’t—and shouldn’t—be logging into individual hosts.
And yet, wouldn’t it be great to have a tail-like, intuitive, real-time view of all those log events?
Enter Loggly Live Tail. It’s a real-time command-line tail that consolidates all your logs into one stream. In other words, it’s a live feed of all the log data that you send to Loggly.
Live Tail enables you to monitor changes of multiple log sources at once and also provides filtering and highlighting so that you can focus on what matters to you without being drowned in a Matrix-like overload of data. As Live Tail is a true command-line application it provides you the ability to pipe its output through any other command-line utilities. It not only runs on Linux, Mac OS X and other Unix-like operating systems, but also on the Windows command prompt and its more mighty sibling, Windows PowerShell.
Sharing Information in Real-Time: Live Tail and HipChat
If that is not enough “modernization” of good old tail -f, let me tell you that you can also send the output from Live Tail to a HipChat room in parallel. (And we will add other products from other vendors soon!) Sharing a live view of what’s going on in your logs—all your logs—with your team members can be extremely powerful when it comes to troubleshooting.
“Given enough eyeballs, all bugs are shallow.” We all know this famous law, and now it has just become much easier to have many experts looking at the same log events as they happen.
Fire It Up!
The Live Tail feature is part of the Enterprise subscription tier. If you don’t have an Enterprise subscription, you can upgrade by contacting your Loggly account manager. If you’re on a Standard or Pro subscription and not ready to upgrade your account but want to try Live Tail, you’ll be able to sign up for a special 14-day trial when you click on the Live Tail option in the Loggly menu bar. Live Tail is not available as a part of the Lite subscription. See our pricing page for details.
The technical documentation for Loggly Live Tail is here.
— tampajohn (@tampajohn) January 13, 2016
John Adams, Chief Architect at Bringhub and Live Tail beta tester:
“I cannot express how happy I am with Live Tail and Loggly in general.
We’re a polyglot product company that has a wide variety of technologies in play, from PHP APIs, to Golang microservices, to Iron.io workers processes, to Node Running in AWS lambda as well as Angular apps. With so much going on, logging can certainly be a bear, especially once you count the number of servers/services that we need to log from. Things get especially sticky when we want to track a single request through our entire pipeline. That’s where Live Tail really shines!
Since we’re able to uniquely identify a request from a specific client through our entire stack, we’re able to (in near real time) see exactly what’s going on when something undesirable occurs, regardless of what server in what cluster in what environment in what region the request may be touching. This truly is an amazingly helpful tool that I wish I had years ago!”