Created attachment 1310157 [details] vdsm log Description of problem: Increase of HE VM memory raises tracebacks under VDSM log 2017-08-07 17:55:15,246+0300 ERROR (jsonrpc/5) [virt.vm] (vmId='b58fdeda-45bb-43d2-b336-ef9953171347') hotplugMemory failed (vm:2971) Traceback (most recent call last): File "/usr/lib/python2.7/site-packages/vdsm/virt/vm.py", line 2969, in hotplugMemory self._dom.attachDevice(deviceXml) File "/usr/lib/python2.7/site-packages/vdsm/virt/virdomain.py", line 95, in f ret = attr(*args, **kwargs) File "/usr/lib/python2.7/site-packages/vdsm/libvirtconnection.py", line 125, in wrapper ret = f(*args, **kwargs) File "/usr/lib/python2.7/site-packages/vdsm/utils.py", line 586, in wrapper return func(inst, *args, **kwargs) File "/usr/lib64/python2.7/site-packages/libvirt.py", line 540, in attachDevice if ret == -1: raise libvirtError ('virDomainAttachDevice() failed', dom=self) libvirtError: unsupported configuration: Attaching memory device with size '1966080' would exceed domain's maxMemory config Version-Release number of selected component (if applicable): vdsm-4.20.2-25.git7499b81.el7.centos.x86_64 libvirt-client-3.2.0-14.el7_4.2.x86_64 ovirt-engine-4.2.0-0.0.master.20170803140556.git1e7d0dd.el7.centos.noarch How reproducible: Always Steps to Reproduce: 1. Deploy hosted-engine 2. Add master storage domain 3. Wait for auto-import operation 4. Increase amount of HE VM memory via UI Actual results: Action succeeds under the engine, but I can see Traceback under vdsm log Expected results: Action succeeds under the engine and vdsm log does not have any new tracebacks Additional info:
what's the maximum memory value in engine's HE VM dialog?
16Gb, I installed HE VM with 4Gb of memory, so max value is 4 * memory.
I believe the problem that the vdsm receive command for memory hotplug, when we still do not support it under the HE VM.
(In reply to Artyom from comment #2) > 16Gb, I installed HE VM with 4Gb of memory, so max value is 4 * memory. nope it's started with 4GB as well, hence you're unable to hot plug anything more. Indeed it may not be supported
Memory hotplug for hosted engine is still not supported and there used to be a condition in the engine code that skipped the call for hosted engine. I know Michal was not fond of it, but it should have prevented this error. Btw: Did the apply now / later dialog show up?
No, when I update the memory, it does not show "Apply Later" dialog.
If memory hotplug for HE VM is not supported, why is the bug targeted for 4.2.0?
Moving forward since we're done with 4.2.
I deployed Hosted Engine on a new Host as well as on a VM. The current versions tested are here as per yum: vdsm.x86_64 4.30.0-372.git422d0a2.el7 libvirt-client.x86_64 3.9.0-14.el7_5.5 @updates ovirt-engine.noarch 4.3.0-0.0.master.20180522090138.git7375cf9.el7 I used the FQDN as the URL for running the Ovirt Engine GUI and doubled the Memory Size of the VM multiple times. There were no errors related to the memory increase either in the vdsm.log nor mom.log files. For one increase the memory status went from: "memoryStats": { "swap_out": 0, "majflt": 0, "minflt": 170, "mem_cached": "465220", "mem_free": "2325176", "mem_buffers": "2104", "swap_in": 0, "pageflt": 170, "mem_total": "5128712", "mem_unused": "2325176" }, To: "memoryStats": { "swap_out": 0, "majflt": 0, "minflt": 236, "mem_cached": "465740", "mem_free": "2321268", "mem_buffers": "2104", "swap_in": 0, "pageflt": 236, "mem_total": "5128712", "mem_unused": "2321268" }, So it does appear that the memory hot-plugging is working and that the trace back no longer occurs.
This is likely working properly since 4.2.z started using the libvirtxml to start the hosted engine VM.