With the burstable feature, we set VM cores to 0 and then set the actual values during provisioning. So, it's 0 for runner allocated logs if it's provisioned on a VM host with slices enabled. Therefore, we should use vcpus instead of cores in the logs. I'm keeping the cores since they have actual values after provisioning, and I don't want to break all the views in our monitoring tool.
6.7 KiB
6.7 KiB