Recent content by hanzfritz85

  1. H

    Is it possible to throttle backup and restore disk io?

    why vzdump bwlimit is so dumb? it runs for a few seconds at the full speed than wait's without transmission, so the average will get equal to bwlimit. i've expected a constant transmission rate of the bwlimit value
  2. H

    CPU pinning?

    Anybody, Is there any best practice to share besides script by t.lamprecht ? Tried to use numa pinning myself but it's not the case
  3. H

    TASK ERROR: total memory for NUMA nodes must be equal to vm static memory

    Thank you for clarification. That's exactly what I wanted to achive
  4. H

    TASK ERROR: total memory for NUMA nodes must be equal to vm static memory

    I don't see any effect of pinned cores. Core load being monitored with htop on proxmox migrates between cores. Can anyone clarify to me what is the correct behavior? Maybe in other hypervisors?
  5. H

    Problem re-adding nfs storage

    #BUG PVE 6.1-7 problem still exists mkdir /mnt/nfs: File exists at /usr/share/perl5/PVE/Storage/NFSPlugin.pm line 134. (500) it's not possible to remove and than recreate NFS storage with the same name
  6. H

    TASK ERROR: total memory for NUMA nodes must be equal to vm static memory

    thank you! yes, the idea is to have a specific VM with dedicated cores of maximum possible performance per core. so nothing expected to share it. other vms will be assigned to other cores. and memory hotplug is a default, but i suppose is less important in my case.
  7. H

    TASK ERROR: total memory for NUMA nodes must be equal to vm static memory

    yep. it starts with memory hotplugging disabled and any amount of memory with cpu core pin config is it a bug?
  8. H

    TASK ERROR: total memory for NUMA nodes must be equal to vm static memory

    and why did it start with 1GB memory and сpu pin enabled
  9. H

    TASK ERROR: total memory for NUMA nodes must be equal to vm static memory

    oh it's a single socket server is it possible to pin cpu cores for vm on a non-numa system?
  10. H

    TASK ERROR: total memory for NUMA nodes must be equal to vm static memory

    Started with 1G memory available: 1 nodes (0) node 0 cpus: 0 1 2 3 4 5 6 7 node 0 size: 64288 MB node 0 free: 56078 MB node distances: node 0 0: 10
  11. H

    TASK ERROR: total memory for NUMA nodes must be equal to vm static memory

    faced the same problem. i wonder if it's because PVE uses GiB to alocate memory and KVM uses GB
  12. H

    TASK ERROR: total memory for NUMA nodes must be equal to vm static memory

    VM conf bootdisk: scsi0 cores: 4 cpu: host hotplug: disk,network,usb,memory,cpu ide2: none,media=cdrom memory: 4096 name: debian numa: 1 numa0: cpus=0-3,hostnodes=0,policy=bind,memory=4096 onboot: 1 ostype: l26 scsi0: local:101/vm-101-disk-0.raw,size=25G scsihw: virtio-scsi-pci serial0: socket...
  13. H

    why snapshot works on lvm (not thin)?

    Hm. You mean snapshot on thick LVM is not a point in time logical snapshot, but more a full backup of vm disk + ram?
  14. H

    why snapshot works on lvm (not thin)?

    It's absolutely clear now. And according to https://pve.proxmox.com/wiki/Storage All storage types which have the “Snapshots” feature also support thin provisioning. Thank you for clarification!