Text 23162, 88 rader
Skriven 2015-04-10 09:33:34 av mark lewis (1:3634/12.0)
  Kommentar till text 23136 av Joe Delahaye (49391.fidonews)
Ärende: Fidonews.us
===================
 On Thu, 09 Apr 2015, Joe Delahaye wrote to mark lewis:

 ML> you don't need to attach anything to a poll FLO file... if one 
 ML> already exists and has files listed in it, you only need to 
 ML> "touch" it to update the time and date stamp... if it still 
 ML> doesn't go anywhere, the binkd logs should tell you why...

 JD> I sometimes end up with orphaned mail packets.  I dont know why,
 JD> but when I have a MO* or similar packet sitting there from 5 days
 JD> ago, and all I have to do is open it with InspectA to see where its
 JD> destinations is, I would like to have some ways of actually trying
 JD> to send that again.  The log window shows Binkd as idle, and what
 JD> mail is waiting or on hold, but that orphan does not show.  I'm
 JD> guessing that a .FLO file is required for that.

yes, a ?LO file would be needed for binkd to know about that orphan again...
what i would do would be to go back through the binkd logs to find where that
file was originally listed and see if binkd had a problem unlinking it after it
was sent... if it did, that could make it an orphan because the rest of the
files and the ?LO associated with it could have been unlinked without
problems... it may also be possible that your tosser was adding more mail to it
at that time... either way, you should get it sent to its destination and let
them deal with duplicates if there are any... otherwise there is the
possibility that mail going to that destination is lost if you just delete the
bundle...

to create a FLO file containing that bundle, add its name to an existing FLO
file for that destination system or create a new FLO file with its name...

eg: echo #X:/full/path/to/bundlename.mo0 >> x:\binkd\out\xxxxyyyy.flo

the leading '#' tells binkd to truncate the bundlename.mo0 file to zero bytes
after it is sent... that is the signal to the tosser to create a new mo1 bundle
if it is the same day... one can use '^' instead of '#' to tell binkd to delete
the file after it is sent... no '#' or '^' means to leave the file alone after
it is sent...

xxxx and yyyy are the four character hex digits for the destination system's
net and node as described previously... make sure you create the flo in the
proper outbound directory... if your default zone is Z1, then your default
outbound directory does not have an extension... if the bundle is destined for
a Z2 system, your outbound would have .002 for the extension...

 JD> Somebody mentioned a .CUT file.  So I am a little confused.

?UT files are PKTs... you don't need to worry about them in regard to
bundles... they are named in the same manner, though...

 JD> For 20 years I have done FD style of mail transfer, so all this is 
 JD> new to me and a little hard to grasp  

yes, it does take some time... FD held out hands and did all of this type of
stuff for us... even going so far as to rescan the outbound to find orphan
files and ensure that they get sent to the destination... the binkleyterm
family is not that sophisticated and never was...

 ML> aside from that, the notation is simple hex... the directory is 
 ML> the destination zone... the file name is broken into two parts of 
 ML> four hex characters each... each portion is padded with leading 
 ML> zeros if the number is not four characters... the first portion is 
 ML> the destination net in hex... the second portion is the 
 ML> destination node in hex...

 ML> eg: 3634/12
 ML> 3634 == 0e32
 ML> 12 == 000c

 ML> 3634/12 == 0e32000c

 ML> since this is a zone 1 address, it would be placed in your zone 1 
 ML> outbound directory...

 ML> the windows calculator has hex, octal, binary and decimal 
 ML> converstion built in... open the calculator and change the view 
 ML> format to scientific (IIRC)... there's a radio selection in 
 ML> decimal by default... enter your decimal number and then click in 
 ML> the hex, octal or binary radio selector to change the number to 
 ML> the chosen base...

 JD> Nick mentioned that as well.  It also has a programming mode.

yeah, use whichever mode is necessary to see the hex, oct, bin modes... i
generally leave mine in that mode or the scientific mode... all of them perform
the math functions in the same way ;) 

)\/(ark

 * Origin:  (1:3634/12)