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

    SG-1000 performance

    Official Netgate® Hardware
    6
    15
    1.9k
    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.
    • D
      d2910rf
      last edited by

      ethernet connection on WAN and LAN. Have experimented with different cables and different duplex/speed settings on the WAN interface. The speeds do seem to be peak a little faster the first speed test following a reboot, but by the second speed test the peaks are closer to the average.

      Havent been able to find any errors associated with the interfaces, though not totally certain I am looking in the right places, will continue to dig around.

      One thing that did seem curious is that while I perform a speed test from a client connected to the LAN, I can see the following process is always the top CPU consumer, however never peaks above 5% CPU

      PID USERNAME  PRI NICE  SIZE    RES STATE    TIME    WCPU COMMAND
        10 root      155 ki31    0K    8K RUN    13:54  88.33% [idle]
        11 root      -92    -    0K  176K WAIT    0:08  1.86% [intr{aintc0,41: cpswss0}]

      Still researching, posting this in case it makes others think of something.

      Thanks for the help here, any other tips greatly appreciated.

      1 Reply Last reply Reply Quote 0
      • stephenw10S
        stephenw10 Netgate Administrator
        last edited by

        Mmm, if it's on;y passing 10Mbps, 86% idle seems about right. That process looks normal.

        Does it show a link at 1Gbps on both NICs? What exactly is it connected to on the WAN? What does your ISP provide there?

        Steve

        1 Reply Last reply Reply Quote 0
        • D
          d2910rf
          last edited by

          I think the attachment shows that both interfaces are negotiating 1Gbps? Is there a better way to check for both NICs showing a link at 1Gbps?

          I have tried 3 different connections on the WAN, all showing the same throughput. directly into a Netgear CM600 - https://www.netgear.com/home/products/networking/cable-modems-routers/CM600.aspx. Then behind that CM600 both a TP Link TL-SG108 - https://www.tp-link.com/us/products/details/cat-42_TL-SG108.html, and an ASUS RT-88U - https://www.asus.com/us/Networking/RT-AC88U/ connecting into the SG-1000.

          My ISP is comcast and the package is advertised as 150Mb/s download speed and 5Mb/s upload speed. with the CM600 as the modem and the ASUS router I have seen speeds slightly greater than these advertised speeds, even over wireless.

          Interfaces.png
          Interfaces.png_thumb

          1 Reply Last reply Reply Quote 0
          • stephenw10S
            stephenw10 Netgate Administrator
            last edited by

            Have you ever seen a faster speed than that? Is this a new deployment? In a previous location perhaps? It looks like that device is ~6m old.

            Steve

            1 Reply Last reply Reply Quote 0
            • SammyWooS
              SammyWoo
              last edited by

              Quick glance suggests Interface(s) somehow negotiated down to 10BT. Interfaces statistics under Dashboard should tell you exactly what's what, but I suspect undisclosed configurations on your part doing it. Did it EVER worked properly when you newly installed pfsense BEFORE any configuration changes?

              1 Reply Last reply Reply Quote 0
              • D
                d2910rf
                last edited by

                I regularly see faster speeds through my Asus router on the same cables. I had this device on a different connection prior to its current location and saw speeds better than they are now. When I moved the device I set it back to factory defaults and upgraded to the 2.4 release. I went ahead and flashed back to the factory image, but no change in speeds. The interface statistics in the dashboard show both interfaces at 1000T.

                At this point I am thinking there must be some optimization within pfsense configuration for this specific device…any ideas?

                1 Reply Last reply Reply Quote 0
                • B
                  bcruze
                  last edited by

                  i would disable ip6 on the sg 1k and see if your speeds change

                  1 Reply Last reply Reply Quote 0
                  • D
                    d2910rf
                    last edited by

                    I have gone into Interfaces and set the IPv6 Configuration Type to none for both interfaces. I have also disabled the DHCPv6 Server & RA service. No change in speeds with any of these configuration changes. Are there other settings relevant to disabling IPv6?

                    1 Reply Last reply Reply Quote 0
                    • B
                      bcruze
                      last edited by

                      i've always followed this: https://airvpn.org/topic/17444-how-to-set-up-pfsense-23-for-airvpn/#entry40137

                      1 Reply Last reply Reply Quote 0
                      • stephenw10S
                        stephenw10 Netgate Administrator
                        last edited by

                        Just to confirm these are the speeds you're seeing direct right? Not via VPN?

                        Steve

                        1 Reply Last reply Reply Quote 0
                        • D
                          d2910rf
                          last edited by

                          Have gone through those steps to disable IPv6, no change in speeds.

                          And correct, this is just a direct connection to my ISP, no VPN connections.

                          1 Reply Last reply Reply Quote 0
                          • First post
                            Last post
                          Copyright 2025 Rubicon Communications LLC (Netgate). All rights reserved.