[asterisk-users] Zaptel timing on TE405P

[Tzafrir Cohen]

On Fri, Jan 18, 2008 at 05:44:12PM +0200, Atis Lezdins wrote:
> On 1/17/08, Atis Lezdins <atis at iq-labs.net> wrote:
> > On 1/17/08, Tzafrir Cohen <tzafrir.cohen at xorcom.com> wrote:
> > > On Thu, Jan 17, 2008 at 03:09:59PM +0200, Atis Lezdins wrote:
> > > > Hi,
> > > >
> > > > I'm wondering why zttest shows
> > > > Best: 99.976 -- Worst: 99.967 -- Average: 99.971469, Difference: 99.971469
> > > >
> > > > Shouldn't it be 100% as timing is hardware and comes from PRI? Am I
> > > > missing some kernel config?
> > >
> > > It may be slightly different. Your system clock may be slightly off. But
> > > more importantly, zttest doesn't start and stop messuring time at
> > > exactly the right spot.
> >
> > Anything i can improve?
> >
> > I think - zttest should do it correctly, as manpage says - definite
> > pass is  100% or 99.99%
> >
> > I'm just having some issues with faxing, so i thought this could be a problem.
> 
> Ping.
> 
> Any ideas what i could do to improve timing accuracy? Some kernel
> options? Newer kernel? Currently I have kernel from RPM:

The question is: "How to improve the meassurment of timing".

Also, some report that Steve Underwood's sliptest is a useful tool for
that. If you find it useful, I have a small patch that makes it slightly
more usable.

  http://www.soft-switch.org/downloads/sliptest.c

-- 
               Tzafrir Cohen
icq#16849755              jabber:tzafrir.cohen at xorcom.com
+972-50-7952406           mailto:tzafrir.cohen at xorcom.com
http://www.xorcom.com  iax:guest at local.xorcom.com/tzafrir