Hi, after updating the testing server, I can't play online any more. I get instant desync or even segfault. Tried also with a bit older versions (but not sure about the exact commit), and it is also bad. I even prepared a fresh clean game and it disconnects as well. Here is the debug output on client:
Warning: karte_t::load: Fileversion: 120008
Warning: karte_t::load(): loaded savegame from 0/1900, next month=4194304, ticks=114458 (per month=1<<22)
Message: network_command_t::rdwr: read packet_id=7, client_id=0
Warning: network_check_activity(): received cmd id=7 nwc_ready_t from socket[8]
Warning: nwc_ready_t::execute: set sync_step=1563 where map_counter=54118
Warning: karte_t::network_game_set_pause: steps=195 sync_steps=1563 pause=0
Message: network_command_t::rdwr: read packet_id=12, client_id=0
Warning: network_check_activity(): received cmd id=12 nwc_auth_player_t from socket[8]
Message: nwc_auth_player_t::execute: plnr = 255 unlock = 32767 our_client_id = 0
Message: network_command_t::rdwr: read packet_id=8, client_id=0
Warning: nwc_tool_t::rdwr: rdwr id=8 client=0 plnr=255 pos=koord3d invalid tool_id=8224 defpar=Welcome, Vladki! init=1 flags=0
Warning: network_check_activity(): received cmd id=8 nwc_tool_t from socket[8]
Message: network_command_t::rdwr: read packet_id=9, client_id=0
Warning: network_check_activity(): received cmd id=9 nwc_check_t from socket[8]
Warning: NWC_CHECK: time difference to server 33
Message: network_command_t::rdwr: read packet_id=8, client_id=0
Warning: nwc_tool_t::rdwr: rdwr id=8 client=0 plnr=255 pos=koord3d invalid tool_id=8224 defpar=Now 1 clients connected. init=1 flags=0
Warning: network_check_activity(): received cmd id=8 nwc_tool_t from socket[8]
Warning: karte_t:::do_network_world_command: sync_step=1568 server=[ss=1568 st=196 nfc=0 rand=3767704857 halt=1 line=1 cnvy=1 ssr=3767704857,3767704857,0,0,0,0,0,0 str=3767704857,3767704857,3767704857,3767704857,3767704857,3767704857,3767704857,3767704857,3767704857,3767704857,3767704857,3767704857,3767704857,53565,1882,3767704857 exr=0,0,0,0,0,0,0,0 client=[ss=1568 st=196 nfc=0 rand=1380863887 halt=1 line=1 cnvy=1 ssr=1380863887,1380863887,0,0,0,0,0,0 str=1380863887,1380863887,1380863887,1380863887,1380863887,1380863887,1380863887,1380863887,1380863887,1380863887,1380863887,1380863887,1380863887,53565,1882,1380863887 exr=0,0,0,0,0,0,0,0
Warning: karte_t:::do_network_world_command: disconnecting due to checklist mismatch
Warning: karte_t::network_disconnect(): Lost synchronisation with server. Random flags: 0
Warning: nwc_routesearch_t::reset: all static variables are reset
World finished ...
Show banner ...
However the swedish server on port 13355, which has not been updated for a while, is fine. And I can connect even with new client version. Unfortunately I do not know which revision is it running. Probably compiled on 2016-11-27.
I assume that you are running Linux to Linux? Can I ask you to test to determine the latest version that does not desync?
Well, I have a version of server and client from Nov 29th most probably: commit 4cf6848ff04f2b5af7e702dd0a27c4d94dfd052b
But, I can connect and play just fine with both versions to pak128.Sweden, but not to pak128.Britain.
Do I understand you to mean that the 29th of November version with commit no. 4cf6848ff04f2b5af7e702dd0a27c4d94dfd052b is the last version to which you can connect and maintain a stable connexion with Pak128.Britain-Ex, with a Linux client and a Linux server, both running the same executable and pakset version?
No, I have the same problems with pak128.Britain-ex with either version of simutrans-experimental (4cf6848ff04f2b5af7e702dd0a27c4d94dfd052b = 29th Nov or ffcc58ef9e740addbeb6b3912b35bb840b3498ae = 3rd Dec), while there are no problems with both versions running pak128.Sweden. Just as if there is something in the pakset that causes problems
Now I noticed that both versions do not update the game info when I click on query server, but again only if I query the pak.britain server. Pak.sweden returns the proper game info
What is the earliest version that does not desync with Pak128.Britain-Ex? Also, are you sure that you have the same pakset version as the server?
Edit: I have just pushed an update that should fix a problem that might have caused a desync. I cannot reproduce a desync with the current version of devel-new-2.
Quote from: jamespetts on December 03, 2016, 06:57:20 PM
What is the earliest version that does not desync with Pak128.Britain-Ex? Also, are you sure that you have the same pakset version as the server?
I have cleaned and recompiled again, and found that there is a file: .routemaster-rescaled.pak. I have uncomented the obj=vehicle in bus/routemaster.dat and voila.. I can connect
So I have tested the latests commit, it fixes the problem with private cars getting stuck.
The problem with desyncs was imho in the routemaster-rescaled.pak.
Splendid, thank you for confirming that. I had been using the rescaled routemaster for testing graphical alignments prior to getting started on my Christmas project for the year, being road vehicles this year. I had not realised that an enabled version of that .dat file had made it into the pakset.
Well it was probably disabled, but not completely. Only the obj=vehicle was commented out. Either leave it there, or comment the whole definition.
That is how it was before - that object should not appear with obj=vehicle commented out. It would certainly not write anything to the pak file like this, and therefore not be capable of causing a desync.
Well, at least for me it created a pak file named: ".routemaster-rescaled.dat"
Hmm - odd.