REPLY: 33.fdn_argus@3:633/509 220e227f
MSGID: 1:3634/12.73 5da8f593
PID: GED+LNX 1.1.5-b20180707
CHRS: CP437 2
TZUTC: -0400
TID: hpt/lnx 1.9.0-cur 17-02-17

 On 2019 Oct 18 09:34:06, you wrote to me:

 ML>> paul, has a connection with this system worked in the past or is this
 ML>> the first attempt at connecting between your systems?

 DG> So it has in the past. (I'm the connection Paul is refering to).

yup :)

 DG> This is Argus -> Argus - whats recently changed is the tosser - and
 DG> thus the format of the ?LO files. (The packets are still going to the
 DG> same path they were previously).

the FLO files' format should not have changed... my binkd and sbbs' binkit
both work just fine with sbbsecho's ?lo files...

the path HAS to be in the ?LO files so the mailer knows where to look for the
file...

you might consider to tcpdump a transaction from sbbs -> binkd and one from
sbbs -> argus and seeing if the path in the ?LO file is actually transmitted...

 DG> I dont recall what the format was previously (it was Ezycom), but now
 DG> that SBBS is tossing the mail, the interpretation of the ?LO files is
 DG> incorrect by Argus.

perhaps you have or can grab the argus source code and take a look?

 DG> (My) Argus is quite happily sending to an upstream SBBS, but fails
 DG> when sending to an upstream Argus.

what does the argus ?LO files look like? you should tcpdump capture that,
too...

 DG> SBBS is configured to toss mail in the outbound C:/mailer/outbound
 DG> (I've also tried C:\mailer\outbound - but the format of the ?LO file
 DG> is ^C:/mailer/outbound...)

because the direction of the slash doesn't matter to modern software... IIRC,
argus/radius/taurus don't care... i've run them all in the past when i had a
working winwhatever setup... somewhere around here i also have the source code
to each of them... it is delphi and i don't do delphi so i need to port it to
freepascal/lazarus which, when i tried, needed some more work on the
conversion translator... that was at least 5+ years ago when the rough idea
was to port them to linux as well as to freepascal/lazarus...

 DG> Argus is configured for the outbound has \mailer\outbound (and its
 DG> installed on C:)

that's fine...

 DG> If I drop the "C:" from the SBBS config - argus interprets the
 DG> outgoing packet as C:\mailer\outbound/mailer/outbound.015/xxxx.pkt
 DG> (and thus is invalid and doesnt send). (The format of the ?LO is
 DG> ^/mailer/outbound.015/xxxx.pkt)

yeah... that's a slight bug, IMHO... but we know how it wants things so we do
it that way...

 DG> If I leave the "C:" in the SBBS config - argus happily attempts to
 DG> transfer the files (it knows about it, reports the size, etc) - and
 DG> sends happily to another SBBS - but to another Argus, the remote
 DG> refuses. (In this case the remote is Paul). (And in this case the
 DG> format of the ?LO file is ^C:/mailer/outbound.015/xxx.pkt)

yeah... the problem is that that local path should not be being transferred...
the remote side has no need to know anything about your local paths...

 DG> My work around is to get SBBS to toss the mail into an outbox, and
 DG> have that outbox configured in Argus.

that'll work, too... painful but works...

 DG> I'm not sure who is "at fault"? I would suggest Argus - it shouldnt be
 DG> sending to the remote the name of the file as /mailer/outbound.015/xxx.pkt
 DG> right? it should just say here is packet xxx.pkt and the remote saves it
 DG> with whatever name it wants, in whatever path it is configured?

right... so something (argus) is not stripping out the path from the
transmitted information...

did you see my other post about using a different protocol than hydra? the
fault could be in there... tell your argus to use something other than hydra
and see what happens...

FWIW: argus was the first in the "ART" family... then came radius... it is a
drop-in replacement and fixes numerous bugs while adding a few features...
taurus is the latest one... it is a drop-in replacement for argus and
radius... it also fixes bugs and brings even more features...

you don't say what version of argus you are running but a newer one might have
a fix for this bug... perhaps it would be an idea to try radius and/or taurus,
too...

)\/(ark

Once men turned their thinking over to machines in the hope that this would
set them free. But that only permitted other men with machines to enslave them.
... Habs. Doan Touch! Too hut! - Neekha
---
 * Origin:  (1:3634/12.73)
SEEN-BY: 1/19 120 123 14/6 15/0 2 16/0 18/0 19/36 34/999 90/1 104/57
SEEN-BY: 106/201 116/18 116 120/302 123/0 25 50 130 131 140 150 755
SEEN-BY: 132/174 135/300 153/7715 203/0 218/700 221/0 1 6 360 222/2
SEEN-BY: 227/114 229/354 426 452 728 1014 230/0 150 152 240/5832 249/206
SEEN-BY: 249/317 250/1 261/38 100 1466 266/512 267/155 275/100 280/464
SEEN-BY: 280/5003 282/1031 1056 291/1 111 292/854 300/4 317/3 320/119
SEEN-BY: 320/219 322/0 757 340/400 342/13 200 396/45 423/120 633/280
SEEN-BY: 640/1321 1384 712/848 770/1 801/161 189 2320/105 2452/250
SEEN-BY: 3634/0 12 15 27 50 119 5020/1042
PATH: 3634/12 261/38 320/219 221/1 280/464 229/426