ehouse
cb4db25f27
call shutdown on cref rather than deleting it; wrap mutex actions with
...
debugging logfs like those in mlock.h.
2005-10-30 05:16:35 +00:00
ehouse
bdace2be94
shutdown in response to SIGINT
2005-10-23 15:49:48 +00:00
ehouse
be6769a937
cleanup ctrl printing; repeat most recent command; cleanup internal names.
2005-10-16 01:19:25 +00:00
ehouse
2a0bc3c2de
fix leak; move heartbeat timer into crefmgr, and only run it when
...
there are active crefs.
2005-10-15 16:30:10 +00:00
ehouse
bf1264b2ee
rename states so can format to 80 cols; add log level param. No
...
change to functionality in this checkin.
2005-10-02 16:08:42 +00:00
ehouse
2b58da4cdf
New scheme for accepting and limiting reconnections. Now device must
...
send player counts, local and expected. Based on these the relay
accepts connections, declares the game full and ready for message
forwarding, and decides whether to accept a reconnect.
2005-10-02 15:39:38 +00:00
ehouse
e098e6f7b2
make transmitted vars smaller where possible; new relay identifying
...
scheme where cookie is used only to connect, and is replaced for
reconnects by a relay-generated name that's supposed to be unique
across all games on all relays and includes a hostname read in from
config file; relay assign non-servers' hostIDs.
2005-10-01 16:33:45 +00:00
ehouse
d9826ae275
Add and handle disconnect message so a device can reuse its socket for
...
a new game.
2005-09-05 15:50:49 +00:00
ehouse
38764b3f73
fix deadlock by collecting crefs into local storage before starting to
...
act on them.
2005-09-04 20:36:53 +00:00
ehouse
175f4eb87d
Send disconnect events with error codes on heartbeat and connect timer
...
failures; change how state machine handles heartbeats: only put events
into the machine when there's a failure.
2005-09-03 06:55:08 +00:00
ehouse
3cd8397085
fix failure to record bytes forwarded
2005-09-02 07:18:39 +00:00
ehouse
048c2ffc6d
first checked in. works.
2005-09-02 06:35:25 +00:00
