We are very excited to announce that from version 4.1 onwards, AppDynamics will support first class monitoring of Apache Web Servers. Apache Web Server has been the most popular web server for some time and is often the gateway to highly distributed applications. Needless to say, monitoring Apache’s availability and performance is critical to gauging the health of a business application.
What does it do?
AppDynamics Web Server Agent lets you
- Discovers web servers in your application topology
- Check the health of your Apache web server(s)
- Alert in the case of any anomalies
- Perform a root cause analysis quickly
Figure 1: Application Flow Map with Web Server
The figure above shows a partial screenshot of an Application Flow Map in AppDynamics. Notice the Apache web server which was discovered automatically and its KPIs – load and average response time.
But how do I do these things?
Bird’s eye view
AppDynamics Application Flow Map will automatically show you all the web servers in the flow map along with their KPIs (as shown in the picture above). The Application Flow Map is constructed using all the traffic that is coming in and flowing between different tiers of your distributed application. Sometimes, you may want to focus only on your web servers and the Tier dashboard allows you to do just that. The Tier Flow Map represents the aggregate traffic flowing through all your web servers, however, if you wanted to focus on one particular web server instance, you could accomplish that by going to the node level dashboard. The gif below shows the different level of details you can view while cycling through these views.
Figure 2: Drilling down from Application Flow Map to Node Dashboard
Outliers and baselining
The web server agent automatically captures three KPIs – load, response time, and errors. AppDynamics also calculates baselines for these KPIs so that you can look instantly look at the outliers (shown below).
Figure 3: Key Performance Indicators
What about Business Transactions?
AppDynamics automatically categorizes all the incoming traffic into Business Transactions (BT) and it’s no different for the web server agent. Your BTs now start at the web server and you get all the goodness that accompanies it like – transaction scorecards, transaction analysis, snapshots etc. More information about BTs can be found here.
For instance, the gif below shows how to troubleshoot the anomaly seen in Figure 3 above around 4:32 PM by going from the dashboard to BT snapshot in a few clicks.
Figure 4: Troubleshooting
So, web servers have BT snapshots?
AppDynamics captures snapshots for all BTs flowing through your application. More information about BT snapshots can be found here. Since these BTs now start at the web server tier, the same snapshots will have a web server segment. Valuable troubleshooting information is included in these snapshots. Specifically for Apache, these snapshots will include the Apache modules where most time is being spent. For instance, Figure 5 shows an Apache segment in a snapshot. Apache’s mod_proxy is taking up most time (222 ms) in this snapshot in the Content Handler stage of Http request processing in Apache. Clicking on the drill down button in the bottom half will take you to the next tier where this http request is routed (Shipping Portal).
Figure 5: Snapshots with Apache Modules and Exit calls
Alright, I am sold. How do I get my hands on this?
Apache Web server agent is now generally available. You can download the installer package from AppDynamics Download Zone.