Follow-up Comment #4, bug #17249 (project freeciv):
I tried to reproduce this from the attached savegame, with 2.2.4 built with
--enable-debug=yes. (I couldn't easily make it a game with only AI players, as
minplayers=1; instead I started the game controlling the Peruvians and did
"/aitoggle agust" before the first timeout, but I suspect that's not
I didn't get exactly the errors Daniel did, but I'm not that surprised given
the above and that we haven't had deterministic autogames for some time.
However, I did get some, from the client:
1: game_remove_unit() at (48,32) unit 1350, Peruvian Riflemen home 1316
1: game_remove_unit() at (48,32) unit 1359, Peruvian Transport home 1582
1: game_remove_unit() at (48,32) unit 4307, Peruvian Alpine Troops home 1094
1: game_remove_unit() at (48,32) unit 1740, Peruvian Riflemen home 1582
1: game_remove_unit() at (48,32) unit 4300, Peruvian Alpine Troops home 1053
1: game_remove_unit() at (51,20) unit 5588, Peruvian Cruise Missile home 1200
1: game_remove_unit() at (30,30) unit 3672, Peruvian Engineers home 1269
These were not accompanied by any diagnostics from the server (which I didn't
run with any "-d" option). (Daniel, were your messages also from the client? I
guess so since they were on the terminal.)
This occurred at about turn 549, 2102AD (I don't know exactly), the Peruvians
and the Mexicans having declared war around 2075AD, and the game ending in
2148AD. I saved games every turn, but replaying from around that time, I
couldn't reproduce it.
I guess this error comes from common/game.c:game_remove_unit(), indicating
that the homecity of a dying unit on the client had somehow ended up invalid.
My client was failing to keep up with the server, so I had some "1: connection
(jtn from localhost (player Agustín Gamarra)) [hack-level] ping timeout
ignored". Could the client get out of sync?
Reply to this item at:
Message sent via/by Gna!
Freeciv-dev mailing list