Description of problem: There is an alert to warn the user when the VM (launcher) is about to exceed it's memory requests. This is an important warning sign, because once the launcher is exceeding it's memory requests, then it is eligible to be killed. Because of the impact on the workload, it is important that the alert is correct in order to give the user visibility into this problem Today, the alert is based on limits. This is also heplful, but less relevant, as requests must have been crossed before the memory usage approaches the limits. Thus a requests based alert is firing earlier (usually, except for gauranteed class), than a limits based on. A limits based on is giving us no insights into if the VM ever approaches or crosses the requests. Version-Release number of selected component (if applicable): 4.10 How reproducible: Difficult - As he launcher needs to grow in it's memory usage to approach the requested limits Steps to Reproduce: 1. TBD 2. 3. Actual results: Expected results: Additional info:
I followed below steps: 1. Created VM (fedora) 2. Installed 'stress' command 3. Applied memory load using 'stress' command couple of times. i. stress --vm 2 --vm-bytes 512M 4. I see the alert "KubevirtVmHighMemoryUsage" having severity:"warning" and state:"Firing" See attached screenshot. @shirly is this sufficient to check/verify this BZ?
After guidance from Shirly, I will verify this BZ.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Moderate: OpenShift Virtualization 4.10.0 Images security and bug fix update), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2022:0947