https://bugs.winehq.org/show_bug.cgi?id=45627
Bug ID: 45627 Summary: mdac28 fails to install (SetupDefaultQueueCallbackW copy error 32 L"C:\users\austin\Temp\IXP000.TMP\msdaorar.dll") Product: Wine-staging Version: 3.13 Hardware: x86 URL: https://github.com/Winetricks/winetricks/issues/1061 OS: Linux Status: NEW Keywords: download, Installer Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs@winehq.org Reporter: austinenglish@gmail.com CC: leslie_alistair@hotmail.com, z.figura12@gmail.com Distribution: Gentoo
This was reported to winetricks upstream here: https://github.com/Winetricks/winetricks/issues/1061
I was unable to reproduce it with plain wine. When I ran with wine-staging-3.13, however, I see it: austin@austin2:~$ wineserver -k ; rm -rf ~/.wine ; WINE=wine-staging-3.13 WINEARCH=win32 winetricks -q mdac28 Using winetricks 20180603-next - sha256sum: b65cef4e1f530ecf7cbd067b6849222503fa994818ca154c2f52fdaf7d976a56 with wine-3.13 (Staging) and WINEARCH=win32 Executing w_do_call mdac28 Executing load_mdac28 Using native,builtin override for following DLLs: mtxdm odbc32 odbccp32 oledb32 Executing wine-staging-3.13 regedit /S C:\windows\Temp_mdac28\override-dll.reg Setting Windows version to nt40 Executing wine-staging-3.13 regedit /S C:\windows\Temp_mdac28\set-winver.reg Executing cd /home/austin/.cache/winetricks/mdac28 Executing wine-staging-3.13 mdac_typ.exe /q /C:setup /qnt 0066:err:richedit:ReadStyleSheet skipping optional destination 0066:err:richedit:ReadStyleSheet skipping optional destination 0066:err:richedit:ReadStyleSheet skipping optional destination 0066:err:richedit:ReadStyleSheet skipping optional destination 0066:err:richedit:ReadStyleSheet skipping optional destination 0066:fixme:setupapi:extract_cabinet_file awful hack: extracting cabinet "C:\users\austin\Temp\IXP000.TMP\msvcrt.CAB" 0066:fixme:setupapi:extract_cabinet_file awful hack: extracting cabinet "C:\users\austin\Temp\IXP000.TMP\mtxfiles.CAB" 0066:fixme:setupapi:extract_cabinet_file awful hack: extracting cabinet "C:\users\austin\Temp\IXP000.TMP\MDACxpak.CAB" 0066:fixme:setupapi:SetupDefaultQueueCallbackW notification 262144 params 32eebc,0 0066:fixme:setupapi:extract_cabinet_file awful hack: extracting cabinet "C:\users\austin\Temp\IXP000.TMP\MDACxpak.CAB" 0066:fixme:setupapi:extract_cabinet_file awful hack: extracting cabinet "C:\users\austin\Temp\IXP000.TMP\MDACxpak.CAB" 0066:err:setupapi:SetupDefaultQueueCallbackW copy error 32 L"C:\users\austin\Temp\IXP000.TMP\msdaorar.dll" -> L"C:\Program Files\Common Files\System\OLE DB\msdaorar.dll" ------------------------------------------------------ Note: command wine-staging-3.13 mdac_typ.exe /q /C:setup /qnt returned status 5. Aborting.
https://bugs.winehq.org/show_bug.cgi?id=45627
--- Comment #1 from Austin English austinenglish@gmail.com --- Maybe not a staging bug. I manually applied staging patches to bisect (on a different machine), and the bug doesn't occur. Perhaps it's compiler related?
Good machine has gcc-7.3.0. I'll have to check the other machine tomorrow (and see if I can maybe bisect the patch as well).
https://bugs.winehq.org/show_bug.cgi?id=45627
bforce.forum@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |bforce.forum@gmail.com
--- Comment #2 from bforce.forum@gmail.com --- I can confirm this bug.
Tested with the latest Wine 3.15 as well as Wine 3.15 staging. Also tested with 64 bit as well as 32 bit.
Running Manjaro, Kernel 4.18.9-1
It's a bit difficult, since there are quite some programs that actually need mdac28 to run or even install properly.
https://bugs.winehq.org/show_bug.cgi?id=45627
PaulStogov paulstogov@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |paulstogov@gmail.com
--- Comment #3 from PaulStogov paulstogov@gmail.com --- Same here.
Kernel 4.18.16
Using winetricks 20180815 - sha256sum: 9343f9174146c0fa4d34e18e05269ae772a45ba78e3d8786829fa21895f6f7f4 with wine-3.20 and WINEARCH=win64
0050:err:setupapi:SetupDefaultQueueCallbackW copy error 32 L"C:\users\ps\Temp\IXP000.TMP\msdaorar.dll" -> L"C:\Program Files (x86)\Common Files\System\OLE DB\msdaorar.dll"
https://bugs.winehq.org/show_bug.cgi?id=45627
--- Comment #4 from PaulStogov paulstogov@gmail.com --- Created attachment 62778 --> https://bugs.winehq.org/attachment.cgi?id=62778 winetricks --force mdac28 log
https://bugs.winehq.org/show_bug.cgi?id=45627
--- Comment #5 from PaulStogov paulstogov@gmail.com --- Last working version is wine 3.7, pls see attachment
https://bugs.winehq.org/show_bug.cgi?id=45627
Rick rickvip@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |rickvip@gmail.com
--- Comment #6 from Rick rickvip@gmail.com --- Hi guys, I'm also confirming that Fedora 28, wine 3.19. is getting the same issue so, that is really bad cuz I've to downgrade wine 3.19 on Fedora 28 to wine 3.5.1 in order to have the application working.
Is there any way that I could be helping with logs?
https://bugs.winehq.org/show_bug.cgi?id=45627
--- Comment #7 from Zebediah Figura z.figura12@gmail.com --- Austin, did you ever check what GCC version the "bad" machine had? Or has anyone else, for that matter? This could be another hot-patching bug.
https://bugs.winehq.org/show_bug.cgi?id=45627
--- Comment #8 from Rick rickvip@gmail.com --- (In reply to Zebediah Figura from comment #7)
Austin, did you ever check what GCC version the "bad" machine had? Or has anyone else, for that matter? This could be another hot-patching bug.
Hi Figura,
is that what are you looking for?
gcc --version gcc (GCC) 8.2.1 20181105 (Red Hat 8.2.1-5) Copyright (C) 2018 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
https://bugs.winehq.org/show_bug.cgi?id=45627
--- Comment #9 from Zebediah Figura z.figura12@gmail.com --- (In reply to Rick from comment #8)
(In reply to Zebediah Figura from comment #7)
Austin, did you ever check what GCC version the "bad" machine had? Or has anyone else, for that matter? This could be another hot-patching bug.
Hi Figura,
is that what are you looking for?
gcc --version gcc (GCC) 8.2.1 20181105 (Red Hat 8.2.1-5) Copyright (C) 2018 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
Yes, that would certainly seem to suggest that hot-patching is the culprit. As the situation stands I guess we need to determine which functions the mdac28 installer tries to hot-patch; I'm not sure how this has been done?
https://bugs.winehq.org/show_bug.cgi?id=45627
Anastasius Focht focht@gmx.net changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|-unknown |-unknown CC| |focht@gmx.net Product|Wine-staging |Wine
--- Comment #10 from Anastasius Focht focht@gmx.net --- Hello folks,
--- quote --- As the situation stands I guess we need to determine which functions the mdac28 installer tries to hot-patch; I'm not sure how this has been done? --- quote ---
one part is tracked in bug 46135 since it was easier to debug due to a crash. The config tool is run as part of install process to register providers.
The problem is reproducibility. Relay trace will automatically work around due to the nature of relay thunks. Without relay, it's not 100% repro for me, but I've definitely seen it a few times too (existing destination file is reopened with write flags combination that causes sharing violation).
Also this is not a Wine-Staging only bug, changing to mainline Wine here.
Regards
https://bugs.winehq.org/show_bug.cgi?id=45627
--- Comment #11 from Rick rickvip@gmail.com --- Hi Focht,
I performed "winetricks mdac28" where it generated follow exception:
0055:err:setupapi:SetupDefaultQueueCallbackW copy error 32 L"C:\users\rick\Temp\IXP000.TMP\msdaorar.dll" -> L"C:\Program Files\Common Files\System\OLE DB\msdaorar.dll"
A note is that I'm using wine profile as 32bits "WINEARCH=win32"
Thank you in advance
https://bugs.winehq.org/show_bug.cgi?id=45627
--- Comment #12 from bforce bforce.forum@gmail.com --- I just tested this again in the new Wine 3.20 version and it still has the same problem. I can also confirm it worked fine in version 3.7
It doesn't matter when using Staging or 32bit/64bit.
The odd thing is that it returns a copy error of msdaorar.dll, but if you look into the right directory it actually did copy that specific file.
https://bugs.winehq.org/show_bug.cgi?id=45627
--- Comment #13 from Anastasius Focht focht@gmx.net --- Hello folks,
I can't reproduce this from current git.
Using some endless loop the exits only on 'winetricks' recipe failure:
--- snip --- $ i=0 ; while true ; let i++ ; echo "*** TRY: $i ***" ; do wineserver -k ; rm -rf ~/.wine ; winetricks -q mdac28 || break ; done --- snip ---
It didn't fail after 30 times. With Wine 3.19 and 3.20 I can see failures with the loop.
It seems after bug 46135 fixed (https://source.winehq.org/git/wine.git/commitdiff/cb8353b6e8d54ddab8ba0a54e9...) there was no further hotpatching needed or it doesn't result in explicit failures.
@bforce
--- quote --- The odd thing is that it returns a copy error of msdaorar.dll, but if you look into the right directory it actually did copy that specific file. --- quote ---
No, that's actually the second attempt to copy and it results in a sharing violation. The file was already copied earlier.
$ wine --version wine-3.20-149-g04fe35a0a3
Regards
https://bugs.winehq.org/show_bug.cgi?id=45627
--- Comment #14 from bforce bforce.forum@gmail.com --- (In reply to Anastasius Focht from comment #13)
Hello folks,
I can't reproduce this from current git.
Using some endless loop the exits only on 'winetricks' recipe failure:
--- snip --- $ i=0 ; while true ; let i++ ; echo "*** TRY: $i ***" ; do wineserver -k ; rm -rf ~/.wine ; winetricks -q mdac28 || break ; done --- snip ---
It didn't fail after 30 times. With Wine 3.19 and 3.20 I can see failures with the loop.
It seems after bug 46135 fixed (https://source.winehq.org/git/wine.git/commitdiff/ cb8353b6e8d54ddab8ba0a54e9c9d411b7f2d09c) there was no further hotpatching needed or it doesn't result in explicit failures.
@bforce
--- quote --- The odd thing is that it returns a copy error of msdaorar.dll, but if you look into the right directory it actually did copy that specific file. --- quote ---
No, that's actually the second attempt to copy and it results in a sharing violation. The file was already copied earlier.
$ wine --version wine-3.20-149-g04fe35a0a3
Regards
What does this mean in practical terms? That it's going to be fixed in the next Wine release?
https://bugs.winehq.org/show_bug.cgi?id=45627
--- Comment #15 from PaulStogov paulstogov@gmail.com --- Created attachment 62876 --> https://bugs.winehq.org/attachment.cgi?id=62876 winetricks --force mdac28 @ WINE 3.21
With WINE 3.21 works like a charm now. Thank youall!
https://bugs.winehq.org/show_bug.cgi?id=45627
--- Comment #16 from bforce bforce.forum@gmail.com --- (In reply to PaulStogov from comment #15)
Created attachment 62876 [details] winetricks --force mdac28 @ WINE 3.21
With WINE 3.21 works like a charm now. Thank youall!
That sounds hopeful!
I am using Arch/Manjaro, so have to wait until they roll out the Wine update. But I will let you guys know.
https://bugs.winehq.org/show_bug.cgi?id=45627
Anastasius Focht focht@gmx.net changed:
What |Removed |Added ---------------------------------------------------------------------------- Fixed by SHA1| |cb8353b6e8d54ddab8ba0a54e9c | |9d411b7f2d09c Component|-unknown |advapi32 Resolution|--- |FIXED Status|NEW |RESOLVED
--- Comment #17 from Anastasius Focht focht@gmx.net --- Hello folks,
resolving 'fixed' here, as already indicated in my comment #13
Fixed by commit https://source.winehq.org/git/wine.git/commitdiff/cb8353b6e8d54ddab8ba0a54e9...
Regards
https://bugs.winehq.org/show_bug.cgi?id=45627
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #18 from Alexandre Julliard julliard@winehq.org --- Closing bugs fixed in 4.0-rc1.