http://bugs.winehq.org/show_bug.cgi?id=16674
Summary: Battle for middle earth2: No network multiplayer possible Product: Wine Version: 1.1.11 Platform: PC OS/Version: Linux Status: UNCONFIRMED Severity: minor Priority: P4 Component: -unknown AssignedTo: wine-bugs@winehq.org ReportedBy: mcflow-forum@gmx.net
Multiplayer games with Battle for Middle Earth2 in wine are not possible. According to the information on the game package it uses TCP/IP for network multiplayer. I already tried using the native dplay dlls like it is stated in the wiki, LAN still does not work.
It happens like this: I enter the network mode and the game displays myself two times in the players list. The other player sees me in the list only one time, as it should be. I can read what he writes and vice versa. If I write something I can see it two times in the chat. For the other user it appears only one time. Now, if I open a game, the other player can see it in the game list. If he wants to join he gets a "Connection Timout" error. Same happens if the other player opens a game and I want to join it. I searched the Bugzilla for quite a time now because I do not want to open a duplicate bug. I found a few things that first sounded interesting but when I tested them nothing changed. I found a note at the AppDB page of Garena 3.0 (http://appdb.winehq.org/objectManager.php?sClass=version&iId=14147&i...) Patching with the patch that is linked there did not solve the problem with wine-1.1.11(well one of the hunks failed but it compiled) In Bug 9787 there also was a patch for the AcceptEX for wine-1.1.6. It did not work with wine-1.1.6. It said that wine does not include AcceptEX. I'm not sure if "The battle for middle earth" requires it. I did not find anything about it on the net except that it uses tcp/ip.
http://bugs.winehq.org/show_bug.cgi?id=16674
--- Comment #1 from Florian mcflow-forum@gmx.net 2008-12-30 13:21:04 --- I found out that it's a regression from wine-0.9.60 to wine-0.9.61 I started a regression testing.
http://bugs.winehq.org/show_bug.cgi?id=16674
Florian mcflow-forum@gmx.net changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |mcflow-forum@gmx.net, | |kai.blin@gmail.com
--- Comment #2 from Florian mcflow-forum@gmx.net 2008-12-30 16:29:30 --- I finished the regression test and this is my result:
1bcbe8651acbdf21ceb1cf42bd53e57fa39bdef7 is first bad commit commit 1bcbe8651acbdf21ceb1cf42bd53e57fa39bdef7 Author: Kai Blin kai.blin@gmail.com Date: Thu Apr 24 13:45:10 2008 +0200
ws2_32: Work around the host name resolving to 127.x.x.x when using that for binding.
http://bugs.winehq.org/show_bug.cgi?id=16674
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |regression
http://bugs.winehq.org/show_bug.cgi?id=16674
--- Comment #3 from Florian mcflow-forum@gmx.net 2008-12-31 15:58:15 --- Mhm... Maybe this is no real new bug. If I compile wine without that commit I see everything correctly in the multiplayer mode when I run the game twice with different wineprefix. Now my friend sees me but I do not see him... I don't know if that's really a regression...
http://bugs.winehq.org/show_bug.cgi?id=16674
Arthur H spyro@spyrius.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |NEW Ever Confirmed|0 |1
--- Comment #4 from Arthur H spyro@spyrius.com 2009-01-03 16:36:28 --- *** This bug has been confirmed by popular vote. ***
http://bugs.winehq.org/show_bug.cgi?id=16674
Kai Blin kai.blin@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC|kai.blin@gmail.com |
--- Comment #5 from Kai Blin kai.blin@gmail.com 2009-01-04 01:35:17 --- Can you try to set up your /etc/hosts file in a way that makes your hostname point to your network address instead of loopback and see if that fixes it?
http://bugs.winehq.org/show_bug.cgi?id=16674
Kai Blin kai.blin@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|-unknown |winsock
--- Comment #6 from Kai Blin kai.blin@gmail.com 2009-01-04 01:37:35 --- Changing component to winsock
http://bugs.winehq.org/show_bug.cgi?id=16674
--- Comment #7 from Florian mcflow-forum@gmx.net 2009-01-04 07:20:26 --- (In reply to comment #5)
Can you try to set up your /etc/hosts file in a way that makes your hostname point to your network address instead of loopback and see if that fixes it?
The Problem is now, that I do not have a second machine to test it again. I can only test it starting the application with a second wineprefix and a different CD Key. I don't know if that makes sense... I rather don't think so.
At new Year's Eve I had a friend here. He used Windows, I used wine without the commit that I think to be a regression. First my hostname pointed to loopback. We were not able to see each other in the game list. Then I changed /etc/hosts so that it contained my local IP (192.168.13.90, if that is of any mean... he had *.91). 'hostname -i' returned it as well ;). Then in the multiplayer menu he was able to see me, read my chat messages but not able to join my opened games. I did not see him or his messages or his games in the multiplayermode. To exclude a problem with his firewall he joined a hamachi network and we had luck there were some people and he could play with them. The game seems not to provide a "join by ip" feature. Battlefield 1942 worked fine, although /etc/hosts contained "127.0.0.1" as my IP. We could play it very fine.
Another note: When I set /etc/hosts to '127.0.0.1 mymachinename', I was able to play the game in multiplayer mode with myself, using two different wine prefixes. When I set it to my local IP that did not work. I was not able to see the player using the other wineprefix.
When I use a wine version with that commit the situation is like in the first post of this bugreport. I try to get some terminal output from the winsock component.
I hope you understand the situations I described... If not you can send me a message in IRC. I'm "McFlow" there.
http://bugs.winehq.org/show_bug.cgi?id=16674
--- Comment #8 from Florian mcflow-forum@gmx.net 2009-01-04 09:08:03 --- Created an attachment (id=18470) --> (http://bugs.winehq.org/attachment.cgi?id=18470) WINEDEBUG=-all,+winsock
I attached a log of wine-0.9.60 with +winsock. /etc/hosts contained: '127.0.0.1 rechensklave' With another IP it did not differ much. I can upload it as well if you want.
http://bugs.winehq.org/show_bug.cgi?id=16674
--- Comment #9 from Florian mcflow-forum@gmx.net 2009-01-04 09:09:15 --- Created an attachment (id=18471) --> (http://bugs.winehq.org/attachment.cgi?id=18471) WINDEBUG=-all,+winsock
wine-1.1.11 with same /etc/hosts...
http://bugs.winehq.org/show_bug.cgi?id=16674
Arthur H spyro@spyrius.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |spyro@spyrius.com
--- Comment #10 from Arthur H spyro@spyrius.com 2009-01-04 13:17:59 --- I can confirm that multiplayer does not function for BFME II: ROTWK using Wine 1.1.12. My /etc/hosts file is properly configured. Going to Options -> Settings in the game properly shows my ip addressrather than loopback.
When I start the game on my linux box to play against a Windows computer on the LAN, then I see nothing in the multiplayer screen, but the Windows user can see everything that I type and he can see my games. I could create a game, but the windows user can not join the game that I start.
When I start the game on my linux box to play against another Linux system, both linux users see nothing in the multiplayer area. The /etc/hosts files are fine on both of them (having a unique 192.168.1.xxx ip), but there seems to be a problem with something in the interaction between Wine and BFME.
I may try to experiment with this further, but it does seem like a problem in the winsock component as currently marked.
http://bugs.winehq.org/show_bug.cgi?id=16674
--- Comment #11 from Kai Blin kai.blin@gmail.com 2009-01-04 14:25:58 --- What I'm currently trying to find out is if the patch mentioned as causing the regression is really the culprit or not. If the _same_ problem persists with a non-loopback /etc/hosts entry, that's not the case. If there's a separate problem, that should go into a separate bug. :)
http://bugs.winehq.org/show_bug.cgi?id=16674
--- Comment #12 from Kai Blin kai.blin@gmail.com 2009-01-04 15:01:15 --- Looking at the traces, I don't see a big difference, at least I don't see how it could have worked with 0.9.60.
Can somebody put up a winsock trace with /etc/hosts set to a network IP address?
http://bugs.winehq.org/show_bug.cgi?id=16674
--- Comment #13 from Florian mcflow-forum@gmx.net 2009-01-04 15:27:36 --- Created an attachment (id=18482) --> (http://bugs.winehq.org/attachment.cgi?id=18482) winedebug=-all,+winsock
/etc/hosts configured to "192.168.13.90 rechensklave" wine-1.1.11 So here it is. It does not change the situation as well.
And it does not work with wine-0.9.60... the issue is different. With 0.9.60 I can play a networkgame using two different wine prefixes on my computer. Network chat between the two instances of the game works fine. But this does not work with wine-1.1.11
http://bugs.winehq.org/show_bug.cgi?id=16674
--- Comment #14 from Arthur H spyro@spyrius.com 2009-01-04 17:26:22 --- Created an attachment (id=18486) --> (http://bugs.winehq.org/attachment.cgi?id=18486) Results of winedebug=-all,+winsock with wine 1.1.12
http://bugs.winehq.org/show_bug.cgi?id=16674
--- Comment #15 from Arthur H spyro@spyrius.com 2009-01-04 17:27:49 --- Created an attachment (id=18487) --> (http://bugs.winehq.org/attachment.cgi?id=18487) Results of winedebug=-all,+winsock with wine 0.9.60
I have the same problem using wine 0.9.60 trying to connect to another computer on the LAN. I have not tested using the same computer yet.
http://bugs.winehq.org/show_bug.cgi?id=16674
--- Comment #16 from Kai Blin kai.blin@gmail.com 2009-01-05 01:39:37 --- (In reply to comment #13)
And it does not work with wine-0.9.60... the issue is different. With 0.9.60 I can play a networkgame using two different wine prefixes on my computer. Network chat between the two instances of the game works fine. But this does not work with wine-1.1.11
But that only works if your hostname points to 127.0.0.1, right? That would be expected.
http://bugs.winehq.org/show_bug.cgi?id=16674
--- Comment #17 from Florian mcflow-forum@gmx.net 2009-01-05 06:47:12 --- (In reply to comment #16)
But that only works if your hostname points to 127.0.0.1, right? That would be expected.
Right.
http://bugs.winehq.org/show_bug.cgi?id=16674
b022d b022d@seznam.cz changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |b022d@seznam.cz
--- Comment #18 from b022d b022d@seznam.cz 2009-03-19 16:44:29 --- This bug appears in Battle for middle earth 1 too. Two nicknames, nobody can connect...
Have anyone found something out yet? Does 0.9.60 actually work?
http://bugs.winehq.org/show_bug.cgi?id=16674
--- Comment #19 from Florian mcflow-forum@gmx.net 2009-03-20 06:18:12 --- (In reply to comment #18)
This bug appears in Battle for middle earth 1 too. Two nicknames, nobody can connect...
Have anyone found something out yet? Does 0.9.60 actually work?
No, I was too fast with writing here. At least for me it does not.
http://bugs.winehq.org/show_bug.cgi?id=16674
--- Comment #20 from Austin English austinenglish@gmail.com 2009-09-22 13:59:39 --- Is this still an issue in current (1.1.29 or newer) wine?
http://bugs.winehq.org/show_bug.cgi?id=16674
Matthew Bauer mjbauer95@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |mjbauer95@gmail.com
--- Comment #21 from Matthew Bauer mjbauer95@gmail.com 2010-08-07 16:02:30 --- (In reply to comment #20)
Is this still an issue in current (1.1.29 or newer) wine?
It's still an issue for 1.3.0.
http://bugs.winehq.org/show_bug.cgi?id=16674
Mike Kaplinskiy mike.kaplinskiy@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |mike.kaplinskiy@gmail.com
--- Comment #22 from Mike Kaplinskiy mike.kaplinskiy@gmail.com 2010-10-01 23:54:50 CDT ---
From the logs it appears the game binds to an address and enables SO_BROADCAST.
Sounds like bug 7929. Could you try the hacks from there to see if they help?
http://bugs.winehq.org/show_bug.cgi?id=16674
Henri Verbeet hverbeet@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Regression SHA1| |1bcbe8651acbdf21ceb1cf42bd5 | |3e57fa39bdef7
http://bugs.winehq.org/show_bug.cgi?id=16674
Kai Blin kai.blin@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords|regression | Status|NEW |RESOLVED Resolution| |DUPLICATE Regression SHA1|1bcbe8651acbdf21ceb1cf42bd5 | |3e57fa39bdef7 |
--- Comment #23 from Kai Blin kai.blin@gmail.com 2011-10-13 09:45:10 CDT --- (In reply to comment #2)
I finished the regression test and this is my result:
1bcbe8651acbdf21ceb1cf42bd53e57fa39bdef7 is first bad commit commit 1bcbe8651acbdf21ceb1cf42bd53e57fa39bdef7 Author: Kai Blin kai.blin@gmail.com Date: Thu Apr 24 13:45:10 2008 +0200
ws2_32: Work around the host name resolving to 127.x.x.x when using that
for binding.
After having been pointed at this again, I'm pretty convinced that 36d63105991abef1612433e18207624f2f9e429e actually fixed the regression that was pointed out in your regression test, but possibly your bisect jumped over that.
Looking at the trace, I think this is again one of the cases where windows networking just works differently from Linux networking, as is the case for bug #7929. Please try the fix there, I'm afraid there's nothing much else we can do.
*** This bug has been marked as a duplicate of bug 7929 ***
http://bugs.winehq.org/show_bug.cgi?id=16674
Erich Hoover ehoover@mines.edu changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |ehoover@mines.edu
--- Comment #24 from Erich Hoover ehoover@mines.edu 2011-10-13 15:21:55 CDT --- According to several sources, this game uses the same engine as the C&C games in Bug #7929 - so the fix there should work here.
http://bugs.winehq.org/show_bug.cgi?id=16674
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #25 from Austin English austinenglish@gmail.com 2011-10-17 16:06:36 CDT --- Closing.