Created attachment 1875358 [details] engine log dump xmls Description of problem: when migration of a VM which is part of hard positive affinity rule that VDSM fails with error VDSErrorException: Failed to MigrateBrokerVDS, error = Invalid parameter: {'reason': 'length of cpusets (0) must match number of CPUs (12)'}, code = 91 Version-Release number of selected component (if applicable): ovirt-engine-4.5.0.4-0.1.el8ev.noarch How reproducible: 100% Steps to Reproduce: configuration: host1 (ibm-p8-rhevm-04.lab2.eng.bos.redhat.com) CPU(s): 128 On-line CPU(s) list: 0,8,16,24,32,40,48,56,64,72,80,88,96,104,112,120 Off-line CPU(s) list: 1-7,9-15,17-23,25-31,33-39,41-47,49-55,57-63,65-71,73-79,81-87,89-95,97-103,105-111,113-119,121-127 Thread(s) per core: 1 Core(s) per socket: 4 Socket(s): 4 NUMA node(s): 4 host2 - the same topology for VM1 <topology> <cores>3</cores> <sockets>4</sockets> <threads>1</threads> </topology> topology for VM2 <topology> <cores>1</cores> <sockets>2</sockets> <threads>1</threads> </topology> 1. Create on cluster affinity VM enforcing positive rule including VM1, VM2 2. Run VM1 and VM2 on host1. 3. select VM1 and migrate with the option 'Migrate all VMs in positive enforcing affinity with selected VMs.' Actual results: 2022-04-27 12:57:04,530+03 ERROR [org.ovirt.engine.core.bll.MigrateVmToServerCommand] (default task-68) [4cfb7805] Command 'org.ovirt.engine.core.bll.MigrateVmToServerCommand' failed: EngineException: org.ovirt.engine.core.vdsbroker.vdsbroker.VDSErrorException: VDSGenericException: VDSErrorException: Failed to MigrateBrokerVDS, error = Invalid parameter: {'reason': 'length of cpusets (0) must match number of CPUs (12)'}, code = 91 (Failed with error InvalidParameter and code 91) Expected results: Additional info:
a correction for https://bugzilla.redhat.com/show_bug.cgi?id=2079351#c0 - both VMs are configured with dedicated policy
I tested locally on master, works for me. Polina, can you please try again on other environment maybe? If it works, i would suspect something went wrong when generating or allocating the new physical CPUs to use.
the problem still happens in ovirt-engine-4.5.1.2-0.11.el8ev.noarch. Re-assigning after discussion with Liran
We manage to solve one problem (letting the engine to think he is possible to migrate and sending the migration command with wrong parameters) - which showed us the error from VDSM in the initial bug. Now the engine doesn't succeed to schedule the VM group as desired.
verified on ovirt-engine-4.5.1.3-0.36.el8ev.noarch according to the description
This bugzilla is included in oVirt 4.5.2 release, published on August 10th 2022. Since the problem described in this bug report should be resolved in oVirt 4.5.2 release, it has been closed with a resolution of CURRENT RELEASE. If the solution does not work for you, please open a new bug report.