A user reported seeing a resource leak in both the cuagent and sipagent of three machines. The screenshot shows one machine with a 17-day uptime and a sustained 25% CPU for the sip agent. The user asked if there were any known resource leak issues in the 9.0.5.423 and 2.15 agents. Another user responded, saying they were not aware of any issues with 2.15 for Desktops and asked if the machines were Windows desktop OS and single user machines. They also suggested checking for AV exclusions, providing a link to documentation for setting them up. The original user confirmed that the machines were single user persistent VDI and said they would check the AV exclusions. URL mentions: https://support.controlup.com/docs/en/av-exclusions-for-controlup?highlight=exclusi
Read the entire article here...
CPU Related Training & Support Archives
CPU training and support-related archives from inside the ControlUp Community on Slack.
Controlling High CPU Processes with ControlUp: Tips and Suggestions from ControlUp Team
A discussion took place about the most efficient way to throttle high CPU-consuming processes, including recommendations for using Citrix WEM or ControlUp for automation. A member suggested checking out built-in ControlUp actions that can be automated through triggers using cmdlets. WEM can also be activated for this customer in the coming weeks. It was also noted that a member named @member discussed this topic recently and may be able to provide more insights. The conversation ended with a mention of "fairly new" features in ControlUp.
Read the entire article here...
Read the entire article here...
Using ListOfColumns Template Variables in ControlUp Triggers: A Helpful Tip and Exciting New Feature in the Works
Users discussed using ListOfColumns template variables for triggers and the impact of using them. One user shared an example of encapsulating the template with ListOfColumns to use multiple metrics in a trigger. Another mentioned that ListOfColumns can be replaced multiple times and added a warning about this. Another expressed excitement about the potential uses of this feature. The discussion also touched on the possibility of seeing the top 50 biggest files that were written and a suggestion for running a follow-up script to enrich the data. One user mentioned that the developer responsible was not available at the moment, but that more information would be shared soon and a real-time disk monitor was already available. It was suggested that the user hang on until Monday for a cool invite.
Read the entire article here...
Read the entire article here...
Improving App Identification in ControlUp VDI with Expanded Streaming API
A user was having difficulty identifying a CPU-heavy app in their VDI. They realized that the App Trends report was showing an average, which was being skewed by multiple instances of the same app with only one of them using significant CPU. To address this, ControlUp is expanding their VDI Streaming REST API to allow for pulling process data with filters. Users can contact their CSM/TSM or open a support ticket for assistance.
Read the entire article here...
Read the entire article here...
Troubleshooting Device Location Updates in Edge DX
A user reported that their device location does not always update in Edge DX, giving an example of their device showing work location after working from home for 18 hours. A suggestion was made to check if device's public IP is changing when switching locations and checking the status of location services on the device's OS. A link to documentation on device geolocation was also shared.
Read the entire article here...
Read the entire article here...
High CPU Usage on macOS Sequoia from CoreAgent – ControlUp Community Discussion and Support
Multiple users reported experiencing high CPU usage on macOS Sequoia from /usr/local/com.controlup.edgedx.agent/Bin/CoreAgent. The issue has been observed on a variety of macOS versions, and restarting the device only temporarily resolves it. A ticket has been opened with CU support, and users are discussing the issue in the community. Some have seen similar issues with earlier agent versions, but it has since been fixed.
Read the entire article here...
Read the entire article here...
Understanding CPU Metrics in ControlUp
The group discussed the discrepancies between the CPU percentage on a single host and the "Host CPU Usage" column for all VMs on that host. The data for the host level comes from the hypervisor platform, while per-VM CPU comes from the ControlUp agent on the Windows OS. It was mentioned that the console is not effective for comparing data due to potential performance issues. The behavior of the host CPU reaching 100% and the VMs not going higher was deemed expected. It was clarified that the host CPU in the "Machine view host CPU" column refers to the CPU utilization by the VM as reported by the hypervisor API. The monitoring intervals for the agent and API may also result in different reported numbers.
Read the entire article here...
Read the entire article here...
Troubleshooting a VM not reporting CPU in ControlUp
A user asked about a possible cause for only one virtual machine (VM) not reporting CPU. Another user suggested checking the value in the console and another user suggested a corrupt WMI may require a rebuild. The user thanked the group for their input.
Read the entire article here...
Read the entire article here...
Inquiring About Browser Performance Monitoring for Real Time DX
A user asked for more details about Browser Performance Monitoring for Real Time DX, specifically asking if metrics for CPU and Memory would be displayed. The only article available provided limited information and listed metrics such as Browser URL Domain, Browser URL, and Browser Tab Title. Another user briefly looked at it and observed that Chrome would appear multiple times, with the resources consumed by each tab listed.
Read the entire article here...
Read the entire article here...
Setting Up a Trigger and Alert in ControlUp for a Stolen Device
A user needed help setting up a trigger and alert to detect a stolen device that may come back online. It was suggested to use the device_status, CPU, and device name to limit alerts and test it with another machine first. Similar guidance was also provided by the team.
Read the entire article here...
Read the entire article here...