how Radius appears to say goodbye.

correctly.

2016-01-20 06:06:39 ERROR line 1836 renaming c:\fd\outbound\20060640.pk_ to
c:\fd\outbound\20060640.pkt
I think this is because the file 200606040.pkt already existed when the
.pk_ temp file was created and thus the rename will fail (can't rename over
an existing file) and then the packet is sent to the wrong link.
When generating a packet name, keep incrementing until both the .pkt and
.pk_ variants do not already exist. This should fix this scenario.

inbound dir.

This is now being tested live on my system... watch this space for more
updates.

If all configured remote nodes have this option, encryption is not required.

While we're here, if none of the remote nodes are configured, don't restrict
them in any way... it won't be a secure session regardless.

If this is set, and there is a connection attempt which does not support
encryption, an error will be returned.

I epxect to actually require this by default in the near future.

This is a callout if either client.socket is undefined or if
client.socket.descriptor === -1

command.

Fixes issue with Radius thinking we left because we sent an empty batch.

Fix semaphore list passing.

Make timeouts at different points in the protocol use different log messages.
When a frame is sent, use a single send() call to avoid small packets on the
wire.

Add TODO comment on locking nodes from the address list.

Don't break out of the session until all ACKs are received.
Better handling of timeouts and poll() errors.

it on my system.

If the rx_callback function returns false, send an M_SKIP and continue.
Fix the advertised program name in accept() method.
Don't ACK the file until you have the whole thing.

Set the inbound directory based on the outbound directory (replace "out" with
"in"), if not explicitly set in the .cfg file.
The setting SCFG->Networks->Fido->Inbound File Directory is being deprecated
(and not used by SBBSecho v3+).

iniReadNamedLongInt()
iniGetNamedLongInt()
iniSetNamedLongInt()
iniSetNamedHexInt()

embedding REQ handling into binkit.

Version 1.09: Added a command-line parameter to let the user choose a message sub-board only within their current message group, or file directory only within their current file library.

intervals (in seconds or fractions of a second) with values in .ini files.

(or possibly intervals) in seconds or fractions of a second. Supports
multipliers: (Y)ear, (W)eek, (D)ay, (H)our, and (M)inute.

- Bool option conversion (e.g. nocircularfwd->fwd_circular) didn't work.
- Log flags needs 0x prefix
- Missing ELIST_ONLY -> add_from_echolists_only bool conversion.

Also using "keys" instead of "flags" for echolists and node configurations.

sbbsecho.cfg file is going to be replaced with sbbsecho.ini. This script can
be used to convert an existing sbbsecho.cfg (or other sbbsecho config file with
a different filename) to an sbbsecho.ini (or other filename if desired). Either
just run "jsexec sbbsecho_upgrade.js" or you can pass the cfg filename and the
ini filenames on the command-line (e.g.
jsexec sbbsecho_upgrade.js path/to/my/cfgfile path/to/my/inifile)

Before I commit v3, I wanted to have some SBBSecho sysops give this converter a
test run. The sbbsecho.ini file won't be used by anything today, I just want
you to examine the resulting file or send it to me (with redacted pwds if you
like).

The objective is to retain all your settings (though some of them are going
away, e.g. store_seenby, store_path, store_kludge, as these are just going
be mandatory in future versions) but any comments you may have added by hand
in your sbbsecho.cfg, unfortunately, won't be retained.

On the bright side, the sbbsecho.ini *will* retain your comments when edited
with the future version of echocfg and the modifications made via areafix won't
be so damaging to any manual changes you've made.

I hope you find the new config file easier to read and manage. I've tried to
retain as many key-words as the old sbbsecho.cfg, so it should feel familiar.
Feedback is welcome.