Try the ez shaper wizard and do not over commit your real bandwidth
available.  Over commiting the bandwidth values will have huge
consequences.

Scott


On 7/25/05, Xtian <[EMAIL PROTECTED]> wrote:
> 
> Bill and Scott:
> 
> Many thanks for the info and the field descriptions. Right, I was doing about
> 105KBps down (on my 1Mbps down, 384Kbps up DSL) which is everything, and then
> initiated an SSH session and latency was as high as ever. Then I looked in
> the rules and saw nothing for SSH. So I assumed it didn't know about SSH. That
> ACKs in general are prioritized makes sense. I tried to make a queue
> specifically for port 22 traffic, and wanted to elevate that above the
> default queue, and thats where I was at a loss as to what I should put in
> those schedule fields. I assumed that what Monowall handles with pipes is
> what got put into scheduler options, but I was just not groking the logic
> behind it.
> 
> I'm a sysadmin by trade, not a netadmin, but I try to learn, you know? ;)
> 
> -Christian
> 
> 
> On Mon, 25 Jul 2005, Bill Marquette wrote:
> 
> > On 7/25/05, Christian Rohrmeier <[EMAIL PROTECTED]> wrote:
> >> I haven't found that to be true. It doesn't create any rules for SSH.
> >> pfSense has a wide selection of games and P2P software that it will make
> >> rules and queues for, but not SSH, unless I overlooked something.
> >> Certainly trying to SSH whilst FTPing a large suffered from the same
> >> massive lag as always.
> >
> > SSH sets the TOS lowdelay bit on all it's ACKs, so non-bulk SSH should
> > by default go into the ACK queue.  Any chance you were saturating your
> > downstream with ACKs, which would force SSH and FTP to then compete
> > within the same queue?
> >
> >> I would still like to know what the 6 fields in the traffic shaper
> >> scheduler are for though!
> >
> > I'll update the code with comments, in the meantime, from the pf.conf man 
> > page:
> >     The hfsc scheduler supports some additional options:
> >
> >     realtime _sc_
> >                 The minimum required bandwidth for the queue.
> >
> >     upperlimit _sc_
> >                 The maximum allowed bandwidth for the queue.
> >
> >     linkshare _sc_
> >                 The bandwidth share of a backlogged queue.
> >
> >     <sc> is an acronym for service curve.
> >
> >     The format for service curve specifications is (m1, d, m2).  m2 controls
> >     the bandwidth assigned to the queue.  m1 and d are optional and can be
> >     used to control the initial bandwidth assignment.  For the first d mil-
> >     liseconds the queue gets the bandwidth given as m1, afterwards the value
> >     given in m2.
> >
> > The boxes correspond to m1, d, m2 in that order (except m1 and d are
> > not optional with pfsense).
> > --Bill
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: [EMAIL PROTECTED]
> > For additional commands, e-mail: [EMAIL PROTECTED]
> >
> 
> --
> devo dot com - "Where the future is only a memory."
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
> 
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to