https://bugs.winehq.org/show_bug.cgi?id=41577
Bug ID: 41577 Summary: Ragnarok Online (iRO): Crashes when attempting to run the patcher. Product: Wine Version: 1.9.21 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs@winehq.org Reporter: proveit@tutamail.com Distribution: ---
Created attachment 55912 --> https://bugs.winehq.org/attachment.cgi?id=55912 This is the entire output that Wine spit out after erroring.
I'm attempting to run Ragnarok Online on Linux Mint 18, Mate, 32-bit. Here's the app's page: https://appdb.winehq.org/objectManager.php?sClass=version&iId=928
I've already exhausted everything in the page, so it's most likely not an existing bug or a problem others have run into. I got the game to work on the same system using Wine 1.6, which was in my official repository, but I hesitate to call this a regression because a tester got the game to work with Wine version 1.9.18, which is very close to mine.
If it IS a regression, doing regression testing is beyond my abilities, but based on the latest test results, we can narrow the problem down to somewhere between the tester's version of Wine (1.9.18), which the tester reported as platinum, and my version of Wine, which is 1.9.21.
I'm using Wine HQ's official development branch, version 1.9.21.
To reproduce the problem, just: 1. Run the RO installer 2. Run "Ragnarok Online" shortcut on desktop
Here's the exact installer I used:
File name: Ragnarok_Installer-09272013.exe MD5: 81D59BEE0621A858428E1CF76843BCF7 SHA1: A2795AA047F29A2547365AC7D4246F7B8F133899
However, I don't think the installer you use matters too much because the bug I'm reporting happens after the game itself is installed.
If you do have my exact installer, however, the installer will hang at "Status: Moving files". To work around the error, follow Richard Graham's comment here: https://appdb.winehq.org/objectManager.php?sClass=version&iId=928
Specifically, first extract the real installer from the .exe file:
$ wine Ragnarok_Installer-09272013.exe /extract "$(winepath -w "$(pwd)")"
Then run the installer itself:
$ wine msiexec /i Ragnarok_Installer.msi
After the installer finishes, run the "Ragnarok Online" shortcut from the desktop. The patcher will open and it will start downloading updates, but after less than 30 seconds, you'll get this error:
------------------------------ Window title: Program Error
The program Ragnarok.exe has encountered a serious problem and needs to close. We are very sorry for the inconvenience.
This can be caused by a problem in the program or a deficiency in Wine. You may want to check the Application Database for tips about running this application. ------------------------------
I saved and attached the program error details, including the backtrace. Please forgive any mistakes/omissions; I'm new to Wine HQ bug reporting. Let me know if there's anything else I can do to help.
https://bugs.winehq.org/show_bug.cgi?id=41577
proveit@tutamail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Distribution|--- |Mint
https://bugs.winehq.org/show_bug.cgi?id=41577
Aaron ironblood@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |ironblood@gmail.com
https://bugs.winehq.org/show_bug.cgi?id=41577
Squik xsquik@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |xsquik@gmail.com
https://bugs.winehq.org/show_bug.cgi?id=41577
winetest@luukku.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |winetest@luukku.com
--- Comment #1 from winetest@luukku.com --- There is also bug 33981. Have you tried that trick? Appdb has 2 links and it seems that they go totally different places and the other was reported some attack page...also the language was something that I couldnt understand.
I am already doing this diffrently the download I have here is .exe and it's somekind of download manager for the game. Also the download manager seems to need winetricks -q ie8 to get started.
There is something wrong with the installer or wine after downloading files. There is a launch button but I can't launch the game pressing the button. After one or two clicks it asks me if I want to remove/modfiy etc the game. Launching the game from install directory gives a gui which has a "Play classic" button, pressing that kills the launcher and it's not clear at all why.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #2 from winetest@luukku.com --- I found your installer
http://ftp.warpportal.com/RagnarokOnline/
At least the name matches I haven't downloaded it yet so I can't tell if the sha1sum matches but why it wouldnt? It feels like that it would be easier to get it working with steam.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #3 from winetest@luukku.com --- Ok I got the same crash location as you. I am not sure if I used winetricks -q ie8 for the launcher. I didnt notice before that the gui was updating files. The font isnt so visible. It seems that it once succees and wants to restart the client and it crashes later. vcrun2008 didnt seem to help as suggested in the other bug report.
I made another run. Now with vcrun2008 installed before hand and I didnt use winetricks -q ie8 and used the steps you mentioned. I get the same crash.
I think this bug is actually a dupe of bug 33981. The issue seems to be the same, but the workaround just don't work anymore.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #4 from winetest@luukku.com --- I got good news.
You need to use the other installer. From here
http://www.playragnarok.com/downloads/clientdownload.aspx
Use fresh wineprefix.
winetricks -q vcrun2008 ie8
Start the installer. Complete all steps launching the game from the isntaller doesn't work. Start the game from the installed directory and it starts to updated itself. I was able to get into game.
wine 2.0rc3.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #5 from winetest@luukku.com --- Since you are extracting the .msi file from exe this bug could be even invalid. The not working exe should have it's own bug. Can you try the steps I got it working?
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #6 from Aaron ironblood@gmail.com --- I was having similar crashes, but usually at the end of the self update, for both the classic and renewal clients, even with winetest's instructions. The final fix, for me, was to launch each client's setup utility, and manually select the correct output device and window resolution. I did try selecting the fullscreen option, but that made my system slow down to the point of complete unresponsiveness, with random garbage on the screen. This is in a 32-bit wine prefix, on an HP elitebook 2540p, sporting 64-bit Linux Mint 18.1 Cinnamon Edition, 4 GB ram, a 250GB hdd, and wine 2.0rc3 from the winehq repo.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #7 from Aaron ironblood@gmail.com --- (In reply to Aaron from comment #6)
I was having similar crashes, but usually at the end of the self update, for both the classic and renewal clients, even with winetest's instructions. The final fix, for me, was to launch each client's setup utility, and manually select the correct output device and window resolution. I did try selecting the fullscreen option, but that made my system slow down to the point of complete unresponsiveness, with random garbage on the screen. This is in a 32-bit wine prefix, on an HP elitebook 2540p, sporting 64-bit Linux Mint 18.1 Cinnamon Edition, 4 GB ram, a 250GB hdd, and wine 2.0rc3 from the winehq repo.
Just in case, I thought I'd clarify what I meant by the Setup utilities. After installation, there is a Setup.exe and a Setup_Classic.exe. Launching each of these, and manually selecting the wine d3d7 device and a window resolution that fits on the screen, THEN launching the game clients is what fixed it for me.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #8 from proveit@tutamail.com --- I am so sorry for the delay. I reported this bug so long ago that I thought it had been abandoned. I check back periodically though. I'm going to do the steps winetest mentioned and report back soon.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #9 from proveit@tutamail.com --- Ok, I tested with the new RO installer you linked to. There was a download manager, which led to the download of a new client. The new client downloads as .msi file, so the extraction of .exe is no longer needed.
I installed vcrun2008 with Winetricks: ./winetricks vcrun2008. I could not install ie8 though, because of licensing issues. The crash may not be due to ie8 though. Can you please test without ie8 and see if you can reproduce this crash? I attached the new backtrace.
Thanks for your help.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #10 from proveit@tutamail.com --- Created attachment 56974 --> https://bugs.winehq.org/attachment.cgi?id=56974 Backtrace using new RO installer
This is the new RO installer. It still crashes at the same spot when trying to update itself. I could not get in game because it crashes before the update finishes.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #11 from winetest@luukku.com --- I don't know how you used winetricks but you need to use -q switch.
winetricks -q ie8.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #12 from proveit@tutamail.com --- Created attachment 57007 --> https://bugs.winehq.org/attachment.cgi?id=57007 Backtrace with -q and vcrun2008
Ok, this crash log was with -q. As I said, I cannot install Internet Explorer due to licensing problems, but everything else is exactly like you said:
winetricks -q vcrun2008
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #13 from winetest@luukku.com --- I did the test so many times that if I say it needed ie8 it needs ie8 (some other version could work). I started with zero overrides.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #14 from proveit@tutamail.com --- Would you consider gecko an acceptable alternative to IE8?
Now that you mention it, I noticed that I am NOT prompted by Wine to install mono or gecko. This is strange. I'm using the development branch, which says "WineHQ does not at present package wine-gecko or wine-mono. When creating a new wine prefix, you will be asked if you want to download those components. For best compatibility, it is recommended to click Yes here."
When I last used Wine, I did get these prompts, so why isn't it prompting me to install gecko or mono now? I don't have packages wine-gecko or wine-mono installed, so there can't be a conflict.
According to the Gecko page, to install manually, I just have to copy the wine_gecko-2.47-x86.msi file to /usr/share/wine/gecko. Well, /usr/share/wine does not even exist, so I created the directory, then created gecko directory too and dropped the file in there. No change; it still crashes.
So the next test I can do is to try running the patcher with Gecko, but Gecko has to let me install it first.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #15 from proveit@tutamail.com --- I found someone else with my problem. Gecko in the latest development branch version does NOT ask to install itself when creating a new Wine prefix: https://forum.winehq.org/viewtopic.php?f=8&t=27257
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #16 from proveit@tutamail.com --- Major progress, and I owe it to you.
I got the game running with my official repository version of Wine (1.6.2). Why? Because I don't know how to install wine-gecko (Internet Explorer replacement?) or wine-mono on the development branch. But packages wine-gecko and wine-mono from my repository do work with my official repository version even if I don't get a popup telling me to install them. I also installed vcrun2008 because of your tests, and the game just runs.
But for the sake of Wine, I want to get it working on development branch too. Because there's no crash with the same installer on an older version of Wine, there is a strong possibility that the cause is the lack of gecko (which might be the "other version" you mentioned. If you can tell me how to install gecko on wine 2.0rc6, then I can do a test to confirm that vcrun2006 and gecko will work.
https://bugs.winehq.org/show_bug.cgi?id=41577
Bruno Jesus 00cpxxx@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |NEW Ever confirmed|0 |1
--- Comment #17 from Bruno Jesus 00cpxxx@gmail.com --- I can confirm this issue in a clean prefix using the manually extracted msi from the file in comment 0.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #18 from Bruno Jesus 00cpxxx@gmail.com --- Created attachment 57009 --> https://bugs.winehq.org/attachment.cgi?id=57009 backtrace with symbols
Backtrace with debug symbols.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #19 from Bruno Jesus 00cpxxx@gmail.com --- The problem in this bug is that it happens randomly and once the patch finishes upgrading correctly you won't be able to reproduce it anymore.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #20 from winetest@luukku.com --- (In reply to Bruno Jesus from comment #19)
The problem in this bug is that it happens randomly and once the patch finishes upgrading correctly you won't be able to reproduce it anymore.
I was never able to get the exe working that was extracted from the .msi file.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #21 from Bruno Jesus 00cpxxx@gmail.com --- (In reply to winetest from comment #20)
I was never able to get the exe working that was extracted from the .msi file.
You probably mean the msi extracted from the exe.
In 10 tests I got the crash 3 times so it is really random, seems to depend on internet and cpu usage, but I can't tell for sure.
From my backtrace the problem is that source length is just wrong in a specific
call after some time. The value is 4522168 (0x4500b8) instead of the 44 from previous call with same length text.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #22 from winetest@luukku.com ---
I was never able to get the exe working that was extracted from the .msi file.
Most likely the opposite. Anyway I had to use the other installer that I linked.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #23 from proveit@tutamail.com --- All right, so I tested RO (using the new installer) on a different computer but with the same version of Wine (2.0 rc6 development branch).
It worked. I just did ./winetricks vcrun2008 (didn't even use -q). I did not use IE8.
There are at least 2 possible explanations: 1. The crash was due to the lack of IE8 or an alternative. Gecko was installed on this new computer for sure, and served as the needed alternative to prevent the crash.
2. As Bruno said, "The problem in this bug is that it happens randomly and once the patch finishes upgrading correctly you won't be able to reproduce it anymore."
#1 will be easy to rule out once I check whether the problem computer crashes. If it has gecko on it and still crashes, then gecko isn't the solution. If it doesn't have gecko, I will have to install it manually. I now know that it's supposed to go in ~/.cache/wine instead of /usr/share/wine/gecko as this wiki page incorrectly states: https://wiki.winehq.org/Gecko
#2 is more complicated, but I can narrow it down by doing multiple installations on the problem computer. The problem is that you can't just rerun the patcher to get around the crash. Even if Bruno had the same problem as me (intermittent), once it crashes, it won't resolve itself and I have to wipe the Wine prefix and start over.
I should have more results soon, but I should note that the original bug was with the old installer, and winetest's testing confirmed the bug. We may be able to get around the problem by using a newer installer, or a different version of Wine, but the bug is there for people who have to use the old installer for some reason. I hope they fix it.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #24 from winetest@luukku.com --- Since so much effort has been used to tackle this issue. And there are multiple versions of installers could you post something to the appdb page about your findings? I don't really play this game.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #25 from proveit@tutamail.com --- Yes, I will do that as soon as possible, but I want to make sure I have definitive results to report. All I know right now is that Gecko is a possible cause of the crash on Wine 2.0rc6, and that the game works on Wine 1.6.2 with just ./winetricks vcrun2008 and wine-gecko/wine-mono installed from the official repository.
Once I've finalized testing on version 2.0rc6, I will post a full summary on the game's appdb page.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #26 from proveit@tutamail.com --- It looks like the crash, for me at least, is computer-specific.
This is the procedure I used in a Linux Mint 18 Mate 32-bit virtual machine:
1. Set desired Wine prefix to a new, unused prefix. Example: export WINEPREFIX=$HOME/.wineROtest 2. Download the latest version of Winetricks. Give it execute permission with chmod +x winetricks. 3a. Install vcrun2008: ./winetricks vcrun2008 3b. Gecko and Mono prompts pop up when winetricks first writes files to the new prefix. Install them both. 4. Install RO using the new installer: wine msiexec /i Ragnarok_Installer.msi 5. Run RO from its desktop shortcut and let the patcher update
This procedure successfully got me into the game in the virtual machine. However, the patcher crashes (step 5) when I try it on a physical computer with the EXACT same version of Linux. I did not need to use -q or to install any version of IE in the virtual machine to get the game to run.
The attachments "Backtrace using new RO installer" and "Backtrace with -q and vcrun2008" show crashes on the physical computer. Can anyone please look through it and find out why it's crashing when the test virtual machine won't? It could be hardware specific. I did everything EXACTLY the same, so why is it crashing in one environment but not another? I am totally confused.
https://bugs.winehq.org/show_bug.cgi?id=41577
DX099 temptempor@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |temptempor@gmail.com
--- Comment #27 from DX099 temptempor@gmail.com --- Created attachment 62314 --> https://bugs.winehq.org/attachment.cgi?id=62314 Backtrace with Steam version and vcrun10 + 12
Hi,
Just to report that I'm still experiencing this bug on the Steam version of this game.
Arch Linux 64bits SteamAppID:250740 Wine version: wine-3.15 (Staging) Winetricks: vcrun2010 + vcrun2012, as per game requirements.
Thank you for your hard work and your support.
https://bugs.winehq.org/show_bug.cgi?id=41577
--- Comment #28 from DX099 temptempor@gmail.com --- Created attachment 62315 --> https://bugs.winehq.org/attachment.cgi?id=62315 Backtrace with Steam version and vcrun2008
Likewise, witch VCRUN2008:
Arch Linux 64bits SteamAppID:250740 Wine version: wine-3.15 (Staging) Winetricks: vcrun2008, as per suggestion on this report.
Thank you for your hard work and your support.
I should add, this is a 32 bit Wine prefix.
Additionnaly, the exact same behavior happens on Valve's Proton software.
As a reminder, this happens when the Launcher is updating the game. One cannot launch the game once before the initial update process has completed.
https://bugs.winehq.org/show_bug.cgi?id=41577
zzzzzyzz@hacari.org changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |zzzzzyzz@hacari.org
https://bugs.winehq.org/show_bug.cgi?id=41577
joaopa jeremielapuree@yahoo.fr changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |jeremielapuree@yahoo.fr
--- Comment #29 from joaopa jeremielapuree@yahoo.fr --- Does the bug still occur with vanilla wine-5.9?