fixing the case where we use every single page in 1 vm

Bug #1831652 reported by Satish Patel
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
OpenStack Compute (nova)
Triaged
Undecided
Unassigned

Bug Description

Got this error: qemu-kvm: -object memory-backend-ram,id=ram-node0,size=12884901888,host-nodes=0,policy=bind: cannot set up guest memory 'ram-node0': Cannot allocate memory.

My Hardware: 32CPU / 32G memory

Flavor: 30 vCPU / 24000 MB (24G) memory

Grub hugemem setting: hugepagesz=2M hugepages=12288

[root@ostack-compute-sriov-196 ~]# cat /sys/devices/system/node/node*/meminfo | grep -i hugepage
Node 0 AnonHugePages: 0 kB
Node 0 HugePages_Total: 6144
Node 0 HugePages_Free: 394
Node 0 HugePages_Surp: 0
Node 1 AnonHugePages: 0 kB
Node 1 HugePages_Total: 6144
Node 1 HugePages_Free: 394
Node 1 HugePages_Surp: 0

Solution was set 23000 MB (23G) in flavor to make it work.

IRC: sean-k-mooney suggested to open bug to use every single page in 1 vm.

Tags: libvirt numa
Revision history for this message
sean mooney (sean-k-mooney) wrote :

we used to have a bug in the libvirt driver where we effectly had an off by 1 error and could not use all pages. in this case i belive this might actully be a qemu issue but i asked for this to be filed so we can figure out if it is a qemu issue or not.

tags: added: libvirt numa
Changed in nova:
status: New → Triaged
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.