Wow. An amazing powerful new feature from ControlUp was released with ControlUp Real-Time Console 7.4. Automated Actions. This article is going to dive into this exciting new feature and what it is capable of and what value it can offer.
ControlUp Automated Actions has several different categories where it can improve the ability of an organization to deliver desktops, applications or administer servers.
Broadly speaking, the areas automation can cover:
- Improving the User Experience
- Reducing Troubleshooting time by gathering contextual information
- Automatic Remediation.
This document will go over all 3 points with examples for how this can help administrators become more effective.
Improving the User Experience
Application delivery environments, like Citrix, VMWare or Microsoft RDSH, have to share resources. Whether it’s the hardware underneath being shared among desktop VM’s or with users on a server, sharing these limited resources allows one to maximize their return on investment. However, this sharing does come with a performance impact. It’s generally up to an administrator to determine how to best ensure operational efficiency while balancing cost. ControlUp can help by ensuring further efficiency by promoting VIP desktops, users or processes.
ControlUp comes with the ability to measure interactivity of a user’s session or process’s. By attaching these measurements to triggers we can take action when a poor experience is encountered. The most common method of user experience improvement is increasing the priority of a process. This can be done for VIP users, or for VIP processes where experience starts to lag because of resource contention. By increasing the priority of these processes when contention is encountered, we can effectively improve the overall experience of the target application.
If one can increase the density of users with existing hardware, cost savings can be substantial. ControlUp offers a unique ability to provide on this front by having both the high-level and low-level view of the resources in your environment. By configuring triggers to monitor a VM’s activity, CPU shares can be increased for active users and decreased for idle users. By default, virtualization environments treat both of these scenarios identically; with idle users consuming resources at equal footing of active users. Increasing CPU shares for active users and simultaneously decreasing CPU shares of idle users can allow more users on an environment.
Reducing troubleshooting time
One of the largest challenges of a Citrix or VMWare administrator is gathering relevant information. There is so much noise in an environment that finding the signal can be time sensitive. ControlUp can leverage triggers to provide an action immediately when these faults occur. These actions can augment existing troubleshooting processes providing additional and relevant information.
For example, by monitoring logon duration metrics, ControlUp can automatically measure additional information and submit a ticket on the user’s behalf. The additional metric’s can include all things relating to logon duration. Group Policy extension processing time, User Profile sizing and load time, logon script duration, printer mapping duration, etc. All of this additional contextual information can be submitted to ServiceNow or another ticketing system so that this information gathering is completed for an administrator to examine for root cause. This additional contextual, deep dive, information can be time consuming or transient as a user moves from resource to resource. By capturing it at the point of fault administrators will now have the proper context to make a decision on how to best resolve.
Another example is augmenting other tools with contextual information as it occurs. Citrix Session Recording has the ability to annotate events as they occur and with ControlUp, annotation can be tied to any metric within the ControlUp console. By triggering on events like when a user goes idle or active, or the utilization of CPU, Disk, or Network, troubleshooting root causes can be determined in seconds.
Perhaps the most powerful use case of automated actions is for automatic remediation. The detection and correction of a failure state as it occurs can prevent business outages, thus saving substantially in costs to the business. ControlUp offers several different ways to remediate on failures. I’ll go through a few examples but the list can be endless.
On startup, AppV packages are loaded on a Citrix server. A failure occurs during this process and an event is written to the event logs. In the past, this incident would be picked up and an alert sent to the administrator. The process of this failure follows a specific pattern. To resolve the issue a standard set of processes is done.
This particular example can cause numerous issues, from copious tickets being generated from users not able to work to the cost to the business through non-productive employees. These faults occur due to the lag from when an issue is detected or alerted and the action of the administrator to resolve the issue. By automating the issue as it occurs, remediation and solving can occur simultaneously; preventing the business and users from ever experiencing an issue.
Occasionally, Windows or Citrix can detect a fault but does not provide remediation. The Citrix Desktop Service, for instance, can detect a failure of its performance counters to initialize properly. This has the potential to be a major issues as performance counters are used to measure the health of a system. This issue generates an event and it’s up to the administrator to resolve it by acknowledging the alert and execute a single command. This is time consuming for the administrator and an annoyance no doubt, to have to manually execute a minor command.
With ControlUp Automated Actions, this simple command can be tied to this event and automatically resolve the issue, freeing the administrator to continue to focus on other tasks.
Some applications may execute an infinite loop, locking the program and pinning the CPU. Users may be unaware that they are in a fault state as this could be a typical long running process. However, the pinned CPU would give away that the application has frozen or crashed but the user does not have visibility to this metric.
Using metrics and time-based triggers, ControlUp Automation can detect this fault state and automatically execute a creation of a process dump file, log the fault and terminate the application. By tweaking the time in state, administrators can capture debugging information as close to the fault as possible. After this information is captured, ControlUp Automation can terminate the application when it’s in the fault state, and users can relaunch and become productive again as quickly as possible.