[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

[Updated] PPP Rx errors...



Well... I have continued to play with this...switching between suse and
debian...and have been able to partially fix the problem.

By setting my wvdial baud rate down to 57600 from 115200 i get only
1 error in 1000 packets... which is respectable...

However, I would still like to know why 115200 works without a single
error on the SuSE setup...but with the same settings debian gives me
errors...

I am using an external USR v.everything modem on a 16550A serial port...

The SuSE box was 6.4 with a 2.2.14 kernel..and the debian box is potato
with the 2.2.15 kernel... all the versions seem to be the same (wvdial,
pppd, etc...)

any ideas? anyone?

thanks again


On Sat, Jul 29, 2000 at 12:12:50AM -0500, Sean Richardson wrote:
> Hello,
> 
> I recently migrated from SuSE 6.4 to Debian potato...and I am having some 
> problems with my PPP connection.
> 
> Under SuSE my connection worked flawlessly and I was able to use  SuSE 
> firewall to share it among my home network with little effort. However, 
> under Debian I have been unable to get a quality ppp connection. With both 
> setups I am using wvdial...and the se
> 
> ttings are identical(well almost)...
> 
> The problem is that on the debian setup I can barely pull up most complex 
> web pages, they time out and I have to keep hitting reload to get all of 
> the page. When I check /proc/net/dev I am getting close to 20% Rx/frame 
> errors. With large single file downl
> 
> oads the Rx errors drop to <5%...but are still there and still occassionaly 
> cause problems.
> 
> On the SuSE setup (which is the same box, same modem, same cables, just 
> different distrib (i am dual booting between them while i migrate), but I 
> have no Rx errors...and pages/dls pull up fine.
> 
> I have compared the configurations in /etc/ppp, version numbers of ppp and 
> wvdial, and looked through countless other possibilities but have been 
> unable to find a possible cause.... or difference between the two 
> distributions.



Reply to: