bbgeek17's latest activity

  • bbgeek17
    bbgeek17 replied to the thread Cluster + shared storage.
    Hello @bitlan, welcome to the forum. Starting with PVE 9, you now have the ability to preview snapshot technology that’s compatible with iSCSI-type storage: https://kb.blockbridge.com/technote/proxmox-qcow-snapshots-on-lvm/ Even if you use this...
  • bbgeek17
    Hello, @pipsqueak . Welcome to the forum. What is the IP of the PVE server? What is the IP of your workstation? What is the IP of your gateway? Can you ping the PVE server? Can you ssh into PVE server? What is the output of "ip a" on PVE? What...
  • bbgeek17
    You could also try to run "sync" inside the VM prior to cloning. Cheers Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
  • bbgeek17
    bbgeek17 replied to the thread VirtIO-SCSI but for NVME.
    To answer your initial question directly: in current versions of QEMU/Proxmox, emulating an NVMe device is unlikely to deliver a dramatic performance improvement. The reason lies in understanding what "virtio" actually is. Virtio is a...
  • bbgeek17
    bbgeek17 reacted to BobhWasatch's post in the thread VirtIO-SCSI but for NVME with Like Like.
    To be clear, the studies @bbgeek17 refers to are about using nvme-over-tcp vs iscsi on the host. It doesn't speak to whether nvme emulation in qemu would perform better or worse than virtio-scsi.
  • bbgeek17
    bbgeek17 replied to the thread VirtIO-SCSI but for NVME.
    There are always trade-offs. In general, NVMe creates multiple queues for I/O processing compared to iSCSI/SCSI. However, each queue consumes CPU. So you may get better I/O performance, but in a shared virtual environment, you may adversely...
  • bbgeek17
    Maybe you have it set at host's level firewall on PVE? Check on webUI or in /etc/pve/nodes/<NODE>/host.fw
  • bbgeek17
    One other thought, so you don't experiment on your production systems - install a PVE as VM and get the process worked out there. Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
  • bbgeek17
    Curious. Perhaps try: echo "options nf_conntrack nf_conntrack_max=524288" > /etc/modprobe.d/nf_conntrack.conf You'll need to reboot, or unload/reload the module which will cause connection reset. Not sure, of the top my head, what could be...
  • bbgeek17
    Don't forget to ensure that you make the change persistent, or you will be in for a surprise sometime after next reboot. I'd even recommend testing it by rebooting the host and checking that it was set again. Blockbridge : Ultra low latency...
  • bbgeek17
    Hey there @bbgeek17 we are up to 13 hosts and about 864 powered on VMs. currently I have 262144 for conntrack_max. I have a TB of ram in each host so I imagine I can try bumping up that max.
  • bbgeek17
    You should ask the authors of those scripts. The information for proper repository context is here: https://pve.proxmox.com/wiki/Package_Repositories#sysadmin_no_subscription_repo You'd need to use an editor, I recommend Emacs. Blockbridge ...
  • bbgeek17
    Hi @dhenrickson , welcome to the forum. Disk naming (indexing) can and does vary across the VMs. Whether the boot disk is disk-0 or disk-1 is completely irrelevant. As @leesteken quickly pointed while I was typing - the error message is...
  • bbgeek17
    Coincidentally those two files are modified by various helper scripts to disable the reminder that no-sub repo is not suitable for production. A message informing you that a package cannot be reinstalled/downloaded means that you do not have...
  • bbgeek17
    I just noticed that in your qmp_capabilities output, the block related to disk-0 contains: "io-status":"nospace" That’s a important clue. My guess is that the LV is potentially not properly sized. You have a 10 GiB LV with a 10 GiB QCOW image...
  • bbgeek17
    Hi @hatemyway , welcome tot he forum. Sounds like this has worked previously? Can you think of anything that has changed? A common occurrence is people installing community scripts or other tools to "improve" PVE usability. Do you have any...
  • bbgeek17
    HA is a complex topic. If you have a cluster of nodes and one node becomes isolated, the remaining nodes holding the majority quorum can vote the isolated node out and take over services. This is normal PVE HA operation. If a single node loses...
  • bbgeek17
    This articles may be helpful to you as you fine tune your environments: https://kb.blockbridge.com/technote/proxmox-lvm-shared-storage/ https://kb.blockbridge.com/technote/proxmox-tuning-low-latency-storage Blockbridge : Ultra low latency...
  • bbgeek17
    @nwad , to ensure that you don't miss the output from the console you can execute these commands via SSH terminal: script serial.log qm terminal [vmid] and leave it running. Another option that we use in our testing is to add char device...
  • bbgeek17
    Hi @ManFriday , Out of curiosity - how large if your environment? Nodes, VMs, etc? What is your current max? sysctl net.netfilter.nf_conntrack_max Write a basic loop to log number of connections, every 1-10seconds, along with date. That should...