As we continue to expand Logentries’ powerful search and metric capabilities, we are also looking for new ways to allow our Users to send us critical data. In the newest version of the Windows Agent users can now follow process metrics and send them to Logentries. This feature enables even more insight into your application by allowing you to monitor your processes resource allocation in Logentries. Additionally, with Logentries powerful real-time alerting capabilities you will be instantly alerted when your process stops logging.
To allow for this feature, the Logentries service has added a brand new tab to the Windows Agent. In this new view you will see a list of all your current processes in your environment. To start collecting information on a given process you simply have to click the checkbox. By default, the Windows Agent sends your metrics every 15 seconds, but this can be easily modified to fit the user’s preferences. You can also refresh the list of processes by pressing the Refresh button if a process that you wish to monitor is not been listed.
Once selected, the Windows Agent will begin to send these process metrics to your logentries-stats.log, which is located in the Host registered to the Windows Agent.
In your Logentries System Stats log you will now be able to see log events detailing the resource usage of any given process. Here you can monitor the CPU, Memory and IO usage of a particular process.
Now that you have these metrics available you can start taking advantage of Logentries powerful log search and analytics features to create Dashboards, which will give you actionable insights into your process resource management.
To make things interesting, I started to monitor the Internet Explorer process and then proceeded to put IE under high load by using http://peacekeeper.futuremark.com/. This resulted in some great reporting metrics; as you can see below, we can clearly identify when the CPU for this process starts to come under load and then eventually recedes.