Subj : dhcp update To : Mike Luther From : Bob Jones Date : Wed Jul 04 2001 04:51 am * Original Area: OS2 * Original To : Mike Luther (1:343/41) ML> Sort post answer here Bob BJ> How are you configuring Binkley to handle outbound BJ> telnet fido calls? I never have taken the time to BJ> work through what I need for special "phone" number BJ> subsitutions to get that to work. [I also have BJ> Binkley handling my POTS line....] ML> for convenience .. in the QNODE.CFG file .. sorta ML> exmples. If this is going to go ballistic to Bink, ML> let's move it there. In that the same concept could be ML> handled by other than BINK, though, it's not quite ML> inappropriate to pass it here so .. ML> You can hard code dedicated IP numbers as follows .. ML> PHONE 1:3##/#### 1-555.55.555.55 ML> PHONE 2:2##/### 1-444.444.44.444 I was trying to avoid the hard coding and use some way of compiling from the nodelist. :( I guess I could also use: PHONE 1:343/41 tophat.darktech.dns And have it telnet back to my system, or PHONE 1:343/41 #tophat.darktech.dns to have it use VMODEM protocol. Ok.... Do you know of any OS/2 based programs that can build this phone number overide table from the nodelist, using the various fields and flags currently available? (IEM flag, bbs name field, phone number starting 000, etc.) ML> That's no worse than any other hard coded address. As ML> well, you can bind all the IP address stuff into a .PVT ML> file tied to your net segment if needed. Either method ML> will work with the below. Guess I'm going to have to start doing this.... Ok.... When you say .PVT file, are you talking for nodelist compilation or for inclusion in Binkley's configuration file? If you are talking nodelist compilation, I assume you are assigning point numbers to those nodes, and having your system dial the point number, and then actually have bink see the real number when it connects.... ML> Some of the NodeList stuff carries a convention of ML> 000.# to designate an IP address situation. It can be ML> translated like this: Duh..... Ok, I've know about this, but I didn't think about it right. Thanks for the reminder. I was already using cost values to keep different binks from dialing wrong nodes.... Hmmmm... I need to read the docs again.... I'm not getting the dialout limitations I want. :( ML> ; The following set up local calls from College Station ML> 000.0 0 50 ML> 000.1 1 50 ML> 000.2 2 50 ML> 000.3 3 50 ML> 000.4 4 50 ML> 000.5 1 50 ML> 000.6 6 50 ML> 000.7 7 50 ML> 000.8 8 50 ML> 000.9 9 50 Snipped.... Except that should be like: 000-0 0 000-1 1 .... ML> ' The dedicated numbers from up above can then be auto ML> handled by, as well as what is by 'convention' in the ML> NodeList by translation below: ML> 1-194.164.53.196 194.164.53.196 50 ML> 1-207.43.172.54 207.43.172.54 50 ML> 1-208-180-150-187 208.180.150.187 50 ML> ML> Note the 'cost' figure. What you do next is to ML> differentiate whether a given task can or cannot handle ML> 'LD' calls according to cost in the .EVT file. If it ML> is above a certain task, only that tast, dedicated to ML> IP use is used. Thus that task ONLY handles IP ML> connections. All costs below it are POTS calls, ML> reserved for another task. In that case, 'LD' calls ML> are blocked to the IP task. Your call as to the ML> figmentary 'cost' that directs the task switch. The cost table trick I've been using for a while.... That's how I keep the POTS line from dialing my BINKD bound crash mail.... Unfortunately, I'm not getting the limitations I want at the moment.... I think I also need to tweek the dialing based on flags..... I don't want to call a BinkD only node via telnet.... ML> Been working solid without a glitch for many many ML> months now here. If you play this out, you can FREQ me ML> via TELNET for 'FILES' or whatever for tests! How many IP type links do you have? I assume they are all telnet based. ML> If you want more thought, let's move this to BINKLEY or ML> OS2BBS or someplace or netmail, whatever... OS2BBS sounds like a better echo..... I'll forward a copy of this to that echo for further reply. ML> I don't have BINKP or BINKD at work here as of yet. I can help you fix that..... BinkD is easy to get working if you have a functioning Binkley setup.... What may be "fun" is getting the cost table setup for Binkley to handle things in this dual telnet / BinkD setup. ML> Milke @ 1:117/3001 Thanks for the note.... Bob Jones, 1:343/41 --- Maximus/2 3.01 * Origin: Top Hat 2 BBS (1:343/41) .