Hi Nicholas,

On 2022-04-17 09:00:40, you wrote to me:

 NB>>> Not sure what to tell you. Anything I write should leave here to
 NB>>> all links at roughly the same time. Does the order of callouts
 NB>>> have anything to do with how you list nodes in binkd.conf? If so,
 NB>>> my list starts with 1: and 2: is further down the list in
 NB>>> numerical order. Otherwise, who knows.

 WV>> It of course depends on in what order your tosser presents the files
 WV>> to binkd. But besides that it should handle most of the outgoing mail
 WV>> in parallel, unless you have your maxclients set to a low value...

 WV>> But your log should be able to tell you what happend (if you're
 WV>> interested enough ;-))...

 NB> Besides what is already out in the open, the other things I can think of is
 NB> (1) an issue that I'm in the process of resolving is that 301/1 connects
 NB> here about 1500 times per day, and roughly 1450 of them are failed
 NB> connections.. This I can't really do anything about until I get a response,
 NB> and (2) with over 100 links, if 1 or more of them are down and maxclients
 NB> is set to a low number, it delays as long as it takes for the connection to
 NB> timeout - which can also hold things up quite a bit. Every once in awhile I
 NB> look into this and remove dead or unresponsive links. What else more can I
 NB> say?

This sounds like a very plausible cause for what I've seen happening.

 NB> I don't remember why I originally set my maxclients to a lower value,
 NB> but something not happening the way I wanted it to in the past made me
 NB> do it. That said, I've raised it for now to see if it makes things
 NB> faster. However, if I start having other issues because of it.. 4
 NB> minutes in transit is still better than a week or more via snail mail.
 NB> ;)

Indeed... I just found it odd. In the end it doesn't really matter if mail comes in via another longer route, as long as it comes in... ;-)

Bye, Wilfred.

--- FMail-lnx64 2.1.0.18-B20170815