It sounds like my issue, i'll have to get the cable provider to change the settings as they won't allow me access into the modem.
This is a Motorola SB6580G in bridge mode.

Adam

On 9/12/2013 3:03 PM, Bao Ha wrote:
We had an issue with a Comcast SMCD3G small business modem earlier with one of our devices. With autodetect on both sides, we only get about 10-15 mbps throughput out of a 50mbps circuit. A PC connected directly to the modem will get the full speed, however.

By forcing the Comcast modem to do 100 mbps Fast Ethernet, Full-Duplex and the pfSense device to auto-detect, we were able to get the full speed. I am not sure it is relevant or not to your case. I think it depends on the PHY, and there might be some incompatibilities at the physical layer. Our device has two lowly RTL8100B Realtek 10/100 Fast Ethernet ports, and that's why we only force 100m/Full-Duplex on the Comcast. Forcing on our device does no good.


On Thu, Sep 12, 2013 at 11:11 AM, Adam Piasecki <[email protected] <mailto:[email protected]>> wrote:

    The switch internally to the LAN Cisco 2960, i've pushed way more
    through it then what i'm doing here, it's been installed for a
    couple of years without a problem. The CPU is only at 15-20%. The
    WAN is a direct connection from the pfsense to the cable modem.
    All my NICs in the pfsense are (em). Running 2.0.1. I'm going to
    upgrade and swap the pfSense to 2.0.3 just to eliminate any faulty
    hardware/software.



    On 9/12/2013 2:03 PM, Christian Borchert wrote:

        What's the CPU and RAM utilization on the switch?  What's its
        spec'd packet forward rate?

        It sounds to me like this is where the problem is.
        Sent via BlackBerry from T-Mobile

        -----Original Message-----
        From: Adam Piasecki <[email protected]
        <mailto:[email protected]>>
        Sender: [email protected]: Thu, 12 Sep 2013
        13:40:33
        To: <[email protected] <mailto:[email protected]>>
        Reply-To: pfSense support and discussion
        <[email protected] <mailto:[email protected]>>
        Subject: Re: [pfSense] pfSense and Cable Modem Throughput

        On 9/12/2013 1:25 PM, Matt Smith wrote:

            On Sep 12, 2013, at 10:28 AM, Adam Piasecki
            <[email protected]
            <mailto:[email protected]>> wrote:

                First I'm almost certain this is a cable
                modem/provider problem. We have a 20mb ethernet
                circuit that works fine with the same pfSense.

                We upgraded to a 100/10mb cable modem, when we put
                this on the WAN of the pfsense, we are getting major
                packet loss during peak times, and speed test sites
                that won't even load. Non-peak times we get no packet
                loss, good speed tests (50+mb)

                The problem I'm having is that when we take the
                pfSense out and plug a PC directly into the cable
                modem, the speedtests look fine and the dropped
                packets go away. Both during peak times and non-peak.

                My thought is the number of packets going over the
                cable modem with the pfSense is a lot greater then
                just one PC doing a speedtest, and the cable modem
                can't handle it. We have about 100 clients behind the
                pfSense trying to access the internet during peak
                times. The traffic graphs on pfSense only indicate we
                are doing  5-10mbs download and 1-5 upload, so we are
                no where near maxing out the cable modem bandwidth wise.

                I've checked wan ethernet settings 1gig full duplex,
                no collisions or errors on the pfSense side. I don't
                see any problems in the log, we are not doing any
                traffic shaping.

            What's the MTU set to on the WAN interface of the pfsense
            box? The cable modem may be encapsulating traffic in PPP
            or doing some other form of tunneling. You might check if
            large packets are being dropped somewhere upstream. You
            can run ping with a large packet size and the do not
            fragment bit set to check this. E.g. 'ping -s 1472 -D
            <some_internet_host>'.

            -Matt Smith

            _______________________________________________
            List mailing list
            [email protected] <mailto:[email protected]>
            http://lists.pfsense.org/mailman/listinfo/list

        Okay, i'll try this, it's 1500 the default.

        _______________________________________________
        List mailing list
        [email protected] <mailto:[email protected]>
        http://lists.pfsense.org/mailman/listinfo/list
        _______________________________________________
        List mailing list
        [email protected] <mailto:[email protected]>
        http://lists.pfsense.org/mailman/listinfo/list



    _______________________________________________
    List mailing list
    [email protected] <mailto:[email protected]>
    http://lists.pfsense.org/mailman/listinfo/list




--
Best Regards.
Bao C. Ha
Hacom - Embedded Systems and Appliances
http://www.hacom.net
voice: (714) 564-9932


_______________________________________________
List mailing list
[email protected]
http://lists.pfsense.org/mailman/listinfo/list

_______________________________________________
List mailing list
[email protected]
http://lists.pfsense.org/mailman/listinfo/list

Reply via email to