Netgate Discussion Forum
    • Categories
    • Recent
    • Tags
    • Popular
    • Users
    • Search
    • Register
    • Login

    Proxmox KVM 'issues' with TNSR?

    TNSR
    3
    3
    1.3k
    Loading More Posts
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
    Reply
    • Reply as topic
    Log in to reply
    This topic has been deleted. Only users with topic management privileges can see it.
    • A
      angelo_cloudafrica
      last edited by angelo_cloudafrica

      Hi,

      We're busy evaluating TNSR and have installed a HOME-LAB version inside Proxmox 6.4 for testing.

      As per another post in this forum, we've defined the CPU as Nehalem and that has enabled us to bring up the interfaces (host and dataplane) correctly.

      Inside the Proxmox hypervisor/host, we've increased ring paramaters from 512 (tx/rx) to:

      Ring parameters for enp1s0f0:
      Pre-set maximums:
      RX: 4096
      RX Mini: 0
      RX Jumbo: 0
      TX: 4096
      Current hardware settings:
      RX: 2048
      RX Mini: 0
      RX Jumbo: 0
      TX: 2048

      We've been running this for the past few days, and we're very keen to make this work and replace a number of hardware edge routing devices.

      We've got a very basic static NAT setup, but we've noticed that:

      1. On occasion, the NAT/IP forwarding just stops;
      2. When changing dataplane cpu workers >1, and attempting to restart the service, everything just stops working - permanently. It's not possible to remove the relevant config lines defining this, and even on reboot, no NAT/IP forwarding happens. We've had to re-install after this happens and then everything comes back.

      I'm not sure if there's any issues/special considerations that need to be taken into account when installing a KVM instance on Proxmox?

      And any ideas on the above issues?

      Thanks,

      Angelo.

      DerelictD 1 Reply Last reply Reply Quote 0
      • DerelictD
        Derelict LAYER 8 Netgate @angelo_cloudafrica
        last edited by

        @angelo_cloudafrica Other than the workers issue (I would just use workers 1 for now) you have not stated what you are testing for and what is not passing your testing or what the criteria for success is.

        You really should use PCI pass-through or SR-IOV NICs to benefit from tnsr's dataplane or you are just putting a layer of kernel networking in the middle anyway.

        Also need to know what type of NAT you are using (endpoint-dependent or endpoint-independent) and what the other NAT settings are... (tnsr)# show nat config shows pretty much everything there. Also, how are you testing?

        Chattanooga, Tennessee, USA
        A comprehensive network diagram is worth 10,000 words and 15 conference calls.
        DO NOT set a source address/port in a port forward or firewall rule unless you KNOW you need it!
        Do Not Chat For Help! NO_WAN_EGRESS(TM)

        1 Reply Last reply Reply Quote 0
        • P
          padpn
          last edited by

          Hello!

          I had the same issue when tried to increase cpu workers count, but I found solution. If you would like to have more cpu workers, you have to increase heap-size param.

          https://docs.netgate.com/tnsr/en/latest/advanced/dataplane-monitoring.html#dataplane-statseg

          I set heap-size 500M and 7 cpu workers.

          1 Reply Last reply Reply Quote 1
          • First post
            Last post
          Copyright 2025 Rubicon Communications LLC (Netgate). All rights reserved.