26 Jul 16 21:02, you wrote to me:

 JL> On 07/26/16, mark lewis said the following...
 ml>> have you turned off your mailer and tried your experiment again so
 ml>> that you can see if there are any flo files (aka ?lo) in your outbound
 ml>> directories?? you have to turn your mailer off to be able to get to
 ml>> them before it sends the files and deletes the ?lo files... that's the
 ml>> only way to see what the tosser is placing in them...
 ml>>

 JL> I did. No luck. Although I did find a .flo file from before I made the
 JL> change to maxservers and maxclients.

 JL> catch22bbs:/home/imp # find . -name "*.flo"
 JL> ./hpt/outbound/944EFE00.flo

cat ~/hpt/outbound/944EFE00.flo

interesting, also, that this one is all caps for the name but lower case for
the extension...

 JL> And the log file entry..

 JL> ? 24 Jul 05:15:12 [18942] start_file_transfer:
 JL> /home/imp/hpt/outbound.02e\944efd00.su0: No such file or directory
 JL> ? 24 Jul 05:15:12 [18942] start_file_transfer:
 JL> /home/imp/hpt/outbound\944efe00.su0: No such file or directory
 JL> + 24 Jul 05:15:12 [18942] done (to 1:154/10@fidonet, OK, S/R: 0/0 (0/0
 JL> bytes)) 24 Jul 05:15:12 [18942] session closed, quitting...

 JL> Not sure why 944efd00.su0 is duplicated for both outgoing directories.

probably because of them all being started at the same time... IIRC hpt uses
unix time for a lot of file names... the 944efd00 certainly doesn't conform to
any BSO style net/node address that i'm aware of...

printf "%d/%d\n" 0x944e 0xfd00
37966/64768

FWIW: 944efd00 doesn't appear to be a valid unix time stamp, either, so i'm not
sure (yet) what's up there...

)\/(ark

Always Mount a Scratch Monkey

... Anything can be funny. All it takes is the right delivery.
---