Sematext now offers quick navigation from your logs to metrics with a simple click.
Why correlate logs and metrics?
Logs give you details about events such as error messages to inform a problem has occurred or something important has happened at a specific point in time. If you were receiving a significant amount of errors on your host you can see this in your logs. The challenge you may have here is finding the starting point of these issues in the piles of logs that are generated by your infrastructure and applications. It can take time to dig in and find the starting point of the root cause.
So how can Sematext help you with this? By correlating your logs with metrics you can see the big picture and spot where and when problems started.
Let’s take os.host example;
You might start receiving a significant amount of error and warning logs on your server. You could start by looking at the logs received from this host in the past N days and try to spot the reason behind it. The problem may be that you don’t know exactly when the real issue started in all the noise. So to come up with a confident conclusion you need to check metrics for this host, such as CPU, disk, memory usage, and processes packages in this host.
If you have a flow of these measurements for long periods of time, you can see when these metrics started to rise at a certain time outside of the normal range. And you can see from the logs that a certain event happened at that time and conclude that this event put things in motion and your vital host metrics started to rise slightly without you noticing.
Now let’s see how you can easily do that with Sematext!
You can easily navigate from logs to metrics from any field listed in Sematext Common Schema.
To do that simply click on the value of the field (e.g.: os.host)
You will be navigated to the Infra App that has os.host details such as metrics, containers, processors, and packages where you can check network traffic, disk IO rate, memory swap metrics and so on… And pinpoint the starting point of a particular issue.
For more information please refer to the logs table quick actions docs.