Stackify Retrace (formerly known simply as Stackify) is a solid IT management tool that provides coverage for those burdened with both infrastructure management and application performance management (APM) duties.
However, while Stackify Retrace (which begins at $50 per month per server) can serve as an infrastructure manager, the company touts its product as "built for developers by developers) and it shows.
We find its deep app insight and exceptional data presentation for APM so impressive that it merits our Editors' Choice award in the APM category.
The company also offers a code profiling tool called Prefix for desktop code, but that's not the focus of this review.
(It's also important to note that while Stackify Retrace appears in our infrastructure management roundup, it does not win our Editors' Choice award in that category, a distinction we give instead to MMSoft Pulseway).
Stackify has extensively rewritten this product, and has renamed it Retrace.
It now supports Java on parity with Microsoft .NET.
Retrace now supports monitoring non-web apps.
As mentioned, pricing for Stackify starts as low as $50 per month per server and $10 for pre-production environments, monitoring without code profiling is $15 per month and Retrace APM for small servers is $25 per month per server; Stackify focuses on using logs as a key source of information for the health of a server more than any of the other products tested.
Logs have always been critically important for monitoring and diagnosing app problems.
Stackify Retrace leverages the logs available on most modern server operating systems, including Microsoft Windows and Linux, to provide its insight into the inner workings and interactions between apps and the supporting environment.
In leveraging this information, Stackify Retrace is following the strategy of companies like Splunk, which have built entire platforms on top of log file analysis to better adapt it for real-time operational intelligence.
Stackify's deep app insight can be seen in the figure below, which shows in one screen a graphical and tabular overview of their HelloStackify.Web demo app.
Since many cloud-based apps use some type of database to store information, it's important to have the ability to see actual database transactions such as the list of SQL queries in the lower left-hand corner of the screen.
The figure below also shows web requests in both a graphical form and in a tabular list so you can see the actual data behind the graphics.
You can see these same graphics on the Stackify Sandbox website.
The depth of app information provided by Stackify Retrace totally outshined the other products in the APM category, which is why it's our Editors' Choice for APM along with LogicMonitor.
Installation
Installing a local agent was a breeze using the Stackify installer.
It does require access to the Internet, but with the supplied links it only takes a few minutes.
The downside here is you must install the agent using the supplied key on every machine you wish to monitor.
That's not a big deal for 10 or even 20 machines, but beyond that it could get tedious without some type of automation.
Fortunately, Stackify Retrace now supports installation via command line scripting with all options as arguments, including the key.
This makes larger scale implementation significantly easier.
Other products in this category such as MMSoft Pulseway and Idera Uptime Infrastructure Monitor don't require the use of a key, making them somewhat easier to install.
For those who want to get fancier with installation, Retrace supports Chef Cookbooks, which is an automation standard that can help with configuration and monitoring.
Stackify also includes built-in support for Amazon Web Servides (AWS), Microsoft Azure, and Docker container technology, which provides the possibility of different deployment strategies as well.
Configuring error notifications consists of selecting a specific monitored item and then choosing what conditions will trigger that notification.
Stackify uses a notification group that consists of one or more contacts associated with a specific app, server template, or resource monitor.
Resource monitors include Azure queues, SQL queries, log queries and a website monitor.
Each notification group also has adjustable time limits for sending an email or SMS reminder.
Configuring monitoring for custom apps required adding code to the app at the time we reviewed Retrace.
That's to create logs which can, in turn, be monitored by Stackify.
Stackify Retrace supports Microsoft .NET-based applications as well as Java and non-web apps .
The time it takes to add a few lines of code to an existing application will be well worth it in terms of the insight you will get into the inner workings of data moving between the different pieces of your system.
The key here is to pinpoint bottlenecks and resource issues in order to make the system run better.
Management Interface
The primary operator interface is clean and uncluttered.
The figure below shows the Monitoring screen for the server named SMSERV2012R2.
Stackify makes use of the sparkline graphical element introduced by Edward Tufte to show trends in a small amount of screen real estate.
Any server issues show up on the single status line as either a warning, critical issue or outage.
Opening the details page for any server lets you quickly identify any alert items, and then one click on the issue of concern brings up a more detailed page to help you diagnose the issue.
I liked the way Stackify presented a lot of information in a compact form when compared to either Idera Uptime Infrastructure Monitor or Ipswitch Whatsup Gold.
The basic server information displayed uses a customizable monitor template to display the data of most interest.
Monitor templates can be modified to change existing threshold values or establish new ones.
Initially, each Windows monitor template watches the disk sub-system, network interface and general system health parameters such as CPU and memory usage plus a few more.
Each of these default categories have additional information that can be displayed if needed.
By default the Stackify Retrace management console does not display any performance counters for a server, but they can easily be added.
For Windows servers the list of available performance counters is huge and includes things like Hyper-V, Server Message Block (SMB), and many more.
Once you add a new performance counter to the template it will show up on all monitored servers in the primary group.
It's also possible to assign new templates on a group basis or by individual server.
The figure below shows the Logs screen from the Stackify sandbox site.
This screen shot is a good example of how the power in log analysis really comes into play with applications.
The screen shown is highly interactive with multiple clickable elements.
To see the information behind an error simply click on the View button next to the right-hand side of the screen.
Similarly, the Trace buttons will launch a separate screen to show the steps taken and code behind that particular action.
Through the use of byte code inspection Stackify provides deep insight into .NET applications.
This includes the ability to see calls to external services and things such as Microsoft SQL database calls that include the query text.
Don't Forget the Hours Used
As mentioned earlier, basic monitoring for applications and servers costs as little as $10 per server per month for pre-production systems.
Additional pricing tiers are $15 for monitoring without code profiling, $25 for small servers.
The regular price is $50 per server per month.
However, Stackify works their pricing on a usage model, so these prices are the maximum you'll pay.
If you don't use Retrace part of the time, then you don't have to pay for the time you're not using.
That represents only the software price, however.
Because Stackify requires that code be added to custom apps for full functionality, be sure to include the man-hours needed to make that happen to your bottom line estimate before purchasing. Even with this caveat, however, Stackify provides the most in-depth information of any of the products tested for Windows and Linux servers.
The same can be said for applications written using the .NET framework or Java.
Graphical elements in the dashboards are informative, and it's easy to drill down with a few clicks to determine the nature of a problem.
Pros
Deep insight into .NET and Java applications.
Lots of information from server-based agent.
Logging integration adds even more data.
Cloud-based management.
View More
The Bottom Line
Stackify Retrace is our Editors' Choice winner when it comes to monitoring application performance, but while it has the ability to manage infrastructure, too, that capability may not work for everyone.
Stackify Retrace (formerly known simply as Stackify) is a solid IT management tool that provides coverage for those burdened with both infrastructure management and application performance management (APM) duties.
However, while Stackify Retrace (which begins at $50 per month per server) can serve as an infrastructure manager, the company touts its product as "built for developers by developers) and it shows.
We find its deep app insight and exceptional data presentation for APM so impressive that it merits our Editors' Choice award in the APM category.
The company also offers a code profiling tool called Prefix for desktop code, but that's not the focus of this review.
(It's also important to note that while Stackify Retrace appears in our infrastructure management roundup, it does not win our Editors' Choice award in that category, a distinction we give instead to MMSoft Pulseway).
Stackify has extensively rewritten this product, and has renamed it Retrace.
It now supports Java on parity with Microsoft .NET.
Retrace now supports monitoring non-web apps.
As mentioned, pricing for Stackify starts as low as $50 per month per server and $10 for pre-production environments, monitoring without code profiling is $15 per month and Retrace APM for small servers is $25 per month per server; Stackify focuses on using logs as a key source of information for the health of a server more than any of the other products tested.
Logs have always been critically important for monitoring and diagnosing app problems.
Stackify Retrace leverages the logs available on most modern server operating systems, including Microsoft Windows and Linux, to provide its insight into the inner workings and interactions between apps and the supporting environment.
In leveraging this information, Stackify Retrace is following the strategy of companies like Splunk, which have built entire platforms on top of log file analysis to better adapt it for real-time operational intelligence.
Stackify's deep app insight can be seen in the figure below, which shows in one screen a graphical and tabular overview of their HelloStackify.Web demo app.
Since many cloud-based apps use some type of database to store information, it's important to have the ability to see actual database transactions such as the list of SQL queries in the lower left-hand corner of the screen.
The figure below also shows web requests in both a graphical form and in a tabular list so you can see the actual data behind the graphics.
You can see these same graphics on the Stackify Sandbox website.
The depth of app information provided by Stackify Retrace totally outshined the other products in the APM category, which is why it's our Editors' Choice for APM along with LogicMonitor.
Installation
Installing a local agent was a breeze using the Stackify installer.
It does require access to the Internet, but with the supplied links it only takes a few minutes.
The downside here is you must install the agent using the supplied key on every machine you wish to monitor.
That's not a big deal for 10 or even 20 machines, but beyond that it could get tedious without some type of automation.
Fortunately, Stackify Retrace now supports installation via command line scripting with all options as arguments, including the key.
This makes larger scale implementation significantly easier.
Other products in this category such as MMSoft Pulseway and Idera Uptime Infrastructure Monitor don't require the use of a key, making them somewhat easier to install.
For those who want to get fancier with installation, Retrace supports Chef Cookbooks, which is an automation standard that can help with configuration and monitoring.
Stackify also includes built-in support for Amazon Web Servides (AWS), Microsoft Azure, and Docker container technology, which provides the possibility of different deployment strategies as well.
Configuring error notifications consists of selecting a specific monitored item and then choosing what conditions will trigger that notification.
Stackify uses a notification group that consists of one or more contacts associated with a specific app, server template, or resource monitor.
Resource monitors include Azure queues, SQL queries, log queries and a website monitor.
Each notification group also has adjustable time limits for sending an email or SMS reminder.
Configuring monitoring for custom apps required adding code to the app at the time we reviewed Retrace.
That's to create logs which can, in turn, be monitored by Stackify.
Stackify Retrace supports Microsoft .NET-based applications as well as Java and non-web apps .
The time it takes to add a few lines of code to an existing application will be well worth it in terms of the insight you will get into the inner workings of data moving between the different pieces of your system.
The key here is to pinpoint bottlenecks and resource issues in order to make the system run better.
Management Interface
The primary operator interface is clean and uncluttered.
The figure below shows the Monitoring screen for the server named SMSERV2012R2.
Stackify makes use of the sparkline graphical element introduced by Edward Tufte to show trends in a small amount of screen real estate.
Any server issues show up on the single status line as either a warning, critical issue or outage.
Opening the details page for any server lets you quickly identify any alert items, and then one click on the issue of concern brings up a more detailed page to help you diagnose the issue.
I liked the way Stackify presented a lot of information in a compact form when compared to either Idera Uptime Infrastructure Monitor or Ipswitch Whatsup Gold.
The basic server information displayed uses a customizable monitor template to display the data of most interest.
Monitor templates can be modified to change existing threshold values or establish new ones.
Initially, each Windows monitor template watches the disk sub-system, network interface and general system health parameters such as CPU and memory usage plus a few more.
Each of these default categories have additional information that can be displayed if needed.
By default the Stackify Retrace management console does not display any performance counters for a server, but they can easily be added.
For Windows servers the list of available performance counters is huge and includes things like Hyper-V, Server Message Block (SMB), and many more.
Once you add a new performance counter to the template it will show up on all monitored servers in the primary group.
It's also possible to assign new templates on a group basis or by individual server.
The figure below shows the Logs screen from the Stackify sandbox site.
This screen shot is a good example of how the power in log analysis really comes into play with applications.
The screen shown is highly interactive with multiple clickable elements.
To see the information behind an error simply click on the View button next to the right-hand side of the screen.
Similarly, the Trace buttons will launch a separate screen to show the steps taken and code behind that particular action.
Through the use of byte code inspection Stackify provides deep insight into .NET applications.
This includes the ability to see calls to external services and things such as Microsoft SQL database calls that include the query text.
Don't Forget the Hours Used
As mentioned earlier, basic monitoring for applications and servers costs as little as $10 per server per month for pre-production systems.
Additional pricing tiers are $15 for monitoring without code profiling, $25 for small servers.
The regular price is $50 per server per month.
However, Stackify works their pricing on a usage model, so these prices are the maximum you'll pay.
If you don't use Retrace part of the time, then you don't have to pay for the time you're not using.
That represents only the software price, however.
Because Stackify requires that code be added to custom apps for full functionality, be sure to include the man-hours needed to make that happen to your bottom line estimate before purchasing. Even with this caveat, however, Stackify provides the most in-depth information of any of the products tested for Windows and Linux servers.
The same can be said for applications written using the .NET framework or Java.
Graphical elements in the dashboards are informative, and it's easy to drill down with a few clicks to determine the nature of a problem.
Pros
Deep insight into .NET and Java applications.
Lots of information from server-based agent.
Logging integration adds even more data.
Cloud-based management.
View More
The Bottom Line
Stackify Retrace is our Editors' Choice winner when it comes to monitoring application performance, but while it has the ability to manage infrastructure, too, that capability may not work for everyone.
