Baslinux mailing list

[Sindi Keesan <keesan AT iamjlamb.com>]

On Fri, 30 Apr 2004, Matrix Mole wrote:

> On Fri, 30 Apr 2004, Sindi Keesan wrote:
> 
> > After dialing with kermit I still cannot dial on the first try with eznet.
> > What script is doing the dialing?  All I can find is the eznet executable
> > 'eznet', which is binary.  It came with the basiclinuxes so you probably
> > have it.  I am using BL2.  pppd is also binary.
> 
> If you look in /var/eznet you'll see eznet.conf (which is usually where
> the eznet writes the info needed to contact your ISP). In that file you
> may have to put a special init string. Also, look at transcript.0 (in the
> same directory)  that appears to be a log of exactly what is done when
> eznet/pppd runs.  There are a few other files in that directory as well
> that may help track down the problem.
> 
 If an init string doesn't work in
> eznet.conf (have you tried that yet?

transcript.0 for this computer indicated that the modem was being sent atz 
and then at&d3.  On another computer which dials properly the second 
string was at&f, which is what I usually give to unknown modems.  (USR 
likes AT&F1 for hardware flow control).

SO I put 0 0 init1 at&f
at the end of eznet just after where Steven gives the instruction to add 
a modem init string (he suggests a slightly longer one and says to look in 
the modem manual, which people never give us with their used computers).

Now it tells me 
registered device ppp0
and then it also DIALS ON THE FIRST TRY!!!!

I don't know why the program was using at&d3 for this modem and at&f for 
the other one, but this has fixed it.  Would at&f work for all modems if 
Steven were to include it in eznet.conf, or might it cause problems?

Thanks for the information which led to the fix.

By the way, you overestimate my ability to understand source code.  All I 
can do so far is run 'make' on it.  

Off topic - we did a low level format on the very slow (190 K/sec transfer 
rate) hard drive and out of 545MB it found over 300 bad cylinders, or 
about 35MB bad.  We will recycle this drive as we have many others.
The conclusion is that you should not use drives that test very slow.

And when I zipped fs.img and transferred the zip file via laplink pro 
(DOS) it produced fs.zip with the same md5sum at both ends of the transfer 
(which did not happen when transferring fs.img) but when I unzipped the 
file at the end it produced fs.img with a different md5sum than the 
starting fs.img file.  Odd.  I will try plip next and after that Kermit 
and a nullmodem cable.