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

    Traffic Shaper Drops qOthersDownH

    Scheduled Pinned Locked Moved Traffic Shaping
    31 Posts 6 Posters 17.7k Views
    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
      danswartz
      last edited by

      But this makes no sense - you have no control over what gets put into the pipe at the ISP's end, only when you are uploading data.  The only thing you can really control download-wise is the total bandwidth.

      1 Reply Last reply Reply Quote 0
      • J
        jlepthien
        last edited by

        Okay. Then maybe I entered to small Mbit/s in my shaper config? How do you test your actual download speed and what are the numbers you enter in the wizard? 10% less of your real bandwidth?

        | apple fanboy | music lover | network and security specialist | in love with cisco systems |

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

          I usually just put the supposed download speed, since it doesn't really matter that much (to me at least).  90% of usable speed seems reasonable.

          1 Reply Last reply Reply Quote 0
          • J
            jlepthien
            last edited by

            No I tested again. I put in the supposed speed and raised qwanacks to 65% and qothersdownh to 70% started a torrent and a http download. Now the queues look like on the new screenshot. I get only some drops in my http traffic which is good, but still a lot in my p2p queue. Isn't this bad? But as you told me this is the only method to throttle the download speed?

            Thanks guys…

            ![Bildschirmfoto 2009-12-21 um 21.13.05.png](/public/imported_attachments/1/Bildschirmfoto 2009-12-21 um 21.13.05.png)
            ![Bildschirmfoto 2009-12-21 um 21.13.05.png_thumb](/public/imported_attachments/1/Bildschirmfoto 2009-12-21 um 21.13.05.png_thumb)

            | apple fanboy | music lover | network and security specialist | in love with cisco systems |

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

              dropping packets is one of the two ways a shaper can slow down inbound traffic.  dropped packets are not inherently bad for TCP traffic.  unless you have an actual performance issue here, you are (IMO) spending a lot of time to try (futilely) to solve an intractable problem.

              1 Reply Last reply Reply Quote 0
              • S
                SlickNetAaron
                last edited by

                @danswartz:

                dropping packets is one of the two ways a shaper can slow down inbound traffic.  dropped packets are not inherently bad for TCP traffic.  unless you have an actual performance issue here, you are (IMO) spending a lot of time to try (futilely) to solve an intractable problem.

                He's having the same problem I'm having.  It's dropping packets when there is no congestion and well below the max pipe bandwidth.  In order for my ~30mb connection to be fully utilized I have to tell the shaper I have a 40mb connection.  If I set it to 35mb it drops packets and maxes me out at 25mb.  If I set to 30mb I max out at about 17/18mb.

                It's dropping packets when I'm using a paltry 64kb bandwidth on the "high priority" queue, but if traffic gets routed to the "others" low priority queue no packets get dropped.  It's functioning in reverse, and shaping (dropping) far too early.

                Also, if I try to set the upper limit on the root queue, it is 100% ineffective.  I could set upper limit to be 1mb and it will still allow 30mb to cross on the wire.  (yes, I've tested for both directions.)  The shaper is pretty much not functioning correctly.

                lastly, there are other ways to shape download - by delaying packet delivery which will cause the app to back off.  see http://www.netequalizer.com/tsfaq.htm

                1 Reply Last reply Reply Quote 0
                • J
                  jlepthien
                  last edited by

                  Probably the shaper will be better in 2.0 which will hopefully be released next year…

                  | apple fanboy | music lover | network and security specialist | in love with cisco systems |

                  1 Reply Last reply Reply Quote 0
                  • J
                    jlepthien
                    last edited by

                    I have got another question concerning the rules of the shaper. The wizards creates rules based on LAN net. Can I simply change LAN net to a single IP? I want some rules to apply only for one IP…

                    | apple fanboy | music lover | network and security specialist | in love with cisco systems |

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

                      @SlickNetAaron:

                      @danswartz:

                      dropping packets is one of the two ways a shaper can slow down inbound traffic.  dropped packets are not inherently bad for TCP traffic.  unless you have an actual performance issue here, you are (IMO) spending a lot of time to try (futilely) to solve an intractable problem.

                      He's having the same problem I'm having.  It's dropping packets when there is no congestion and well below the max pipe bandwidth.  In order for my ~30mb connection to be fully utilized I have to tell the shaper I have a 40mb connection.  If I set it to 35mb it drops packets and maxes me out at 25mb.  If I set to 30mb I max out at about 17/18mb.

                      It's dropping packets when I'm using a paltry 64kb bandwidth on the "high priority" queue, but if traffic gets routed to the "others" low priority queue no packets get dropped.  It's functioning in reverse, and shaping (dropping) far too early.

                      Also, if I try to set the upper limit on the root queue, it is 100% ineffective.  I could set upper limit to be 1mb and it will still allow 30mb to cross on the wire.  (yes, I've tested for both directions.)  The shaper is pretty much not functioning correctly.

                      lastly, there are other ways to shape download - by delaying packet delivery which will cause the app to back off.  see http://www.netequalizer.com/tsfaq.htm

                      Yeah, I'm aware of that technique.  You have provided a better (more convincing) description of what is wrong.  I would open a ticket on redmine.pfsense.org.

                      1 Reply Last reply Reply Quote 0
                      • R
                        Rick164
                        last edited by

                        Having the same problem, the strange part is that it works fine for about a day or so but after that it starts dropping qOthersdownH and thus giving qP2p higher priority(internet pages either timeout or load slow).
                        The only thing that helps is restarting pfsense or the traffic shaper, can't find the problem as there are no errors in the logs and nothing has changed in the configs in the meantime.
                        This is with pfsense 1.2.3 final(4GB nanobsd) and it worked ok in 1.2.2 and in some of the early 1.2.3 builds, tried alot of different traffic shaper settings(lowering/raising bandwith for instance) but with no luck so far.

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

                          i too suffer from this issue, and i thought its just me… with aggressive apps, drops could sometimes reach 5 digits.. it doesnt matter if qlimit is specified or not, but it happens either way.

                          would be good to hear feedback from a dev

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