http://bugs.winehq.org/show_bug.cgi?id=29997
Bug #: 29997 Summary: StudioTax 2011 crashes during installation Product: Wine Version: 1.4-rc4 Platform: x86-64 URL: http://www.studiotax.com/en/download.htm OS/Version: Mac OS X Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs@winehq.org ReportedBy: mfuhrer@alumni.ucalgary.ca Classification: Unclassified
The StudioTax 2011 installer crashes shortly after clicking the "Install" button. I tried this on a clean wine directory with .NET 2.0 installed (manual installation, without winetricks). I subsequently installed .NET 2.0 sp2 (used winetricks at this point) and tried running the StudioTax 2011 installer again, and got an almost identical crash.
http://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #1 from Martin Fuhrer mfuhrer@alumni.ucalgary.ca 2012-02-25 00:51:43 CST --- Created attachment 39056 --> http://bugs.winehq.org/attachment.cgi?id=39056 Terminal output with .NET 2.0 installed
http://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #2 from Martin Fuhrer mfuhrer@alumni.ucalgary.ca 2012-02-25 00:52:23 CST --- Created attachment 39057 --> http://bugs.winehq.org/attachment.cgi?id=39057 Terminal output with .NET 2.0 sp2 installed
http://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #3 from Martin Fuhrer mfuhrer@alumni.ucalgary.ca 2012-02-25 00:56:04 CST --- Created attachment 39058 --> http://bugs.winehq.org/attachment.cgi?id=39058 Installer crash with .NET 2.0 installed
http://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #4 from Martin Fuhrer mfuhrer@alumni.ucalgary.ca 2012-02-25 00:56:40 CST --- Created attachment 39059 --> http://bugs.winehq.org/attachment.cgi?id=39059 Installer crash with .NET 2.0 sp2 installed
http://bugs.winehq.org/show_bug.cgi?id=29997
Martin Fuhrer mfuhrer@alumni.ucalgary.ca changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |Installer
http://bugs.winehq.org/show_bug.cgi?id=29997
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |download
http://bugs.winehq.org/show_bug.cgi?id=29997
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Attachment #39057|studiotax2011_dotnet20sp2.l |studiotax2011_dotnet20sp2.t filename|og |xt Attachment #39057|application/octet-stream |text/plain mime type| |
http://bugs.winehq.org/show_bug.cgi?id=29997
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Attachment #39056|studiotax2011_dotnet20.log |studiotax2011_dotnet20.txt filename| | Attachment #39056|application/octet-stream |text/plain mime type| |
http://bugs.winehq.org/show_bug.cgi?id=29997
Vitaliy Margolen vitaliy-bugzilla@kievinfo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |dotnet
--- Comment #5 from Vitaliy Margolen vitaliy-bugzilla@kievinfo.com 2012-02-25 11:04:29 CST --- Please try with 32-bit wineprefix: WINEPREFIX=~/.wine-test WINEARCH=win32 wine winecfg
http://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #6 from Martin Fuhrer mfuhrer@alumni.ucalgary.ca 2012-02-25 15:23:29 CST --- I tried with a clean 32-bit wineprefix, but get the same crash. I also realized that .NET does not need to be preinstalled; the crash can be reproduced on a clean wine directory, just by running % wine ./StudioTax2011Install.exe
http://bugs.winehq.org/show_bug.cgi?id=29997
sebastiendevel@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |sebastiendevel@gmail.com
--- Comment #7 from sebastiendevel@gmail.com 2012-03-17 10:15:19 CDT --- I can confirm I get the same error. Maybe we can pass this as confirmed?
http://bugs.winehq.org/show_bug.cgi?id=29997
Vitaliy Margolen vitaliy-bugzilla@kievinfo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |NEW Ever Confirmed|0 |1
--- Comment #8 from Vitaliy Margolen vitaliy-bugzilla@kievinfo.com 2012-03-17 10:56:45 CDT --- Confirming - multiple users having issue.
http://bugs.winehq.org/show_bug.cgi?id=29997
Anastasius Focht focht@gmx.net changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |focht@gmx.net
--- Comment #9 from Anastasius Focht focht@gmx.net 2012-03-17 17:26:40 CDT --- Hello,
I don't get a crash using clean WINEPREFIX but a message box is shown, stating "Invalid command line".
--- snip --- 0024:Call KERNEL32.__wine_kernel_init() ret=7bc534a2 ... 0024:Call KERNEL32.CreateProcessW(0013d9e0 L"Z:\home\focht\Downloads\StudioTax2011Install.exe",00148230 L" /i "C:\users\focht\Application Data\BHOK IT Consulting\StudioTax 2011\install\StudioTaxX64.msi" TRANSFORMS="C:\users\focht\Application Data\BHOK IT Consulting\StudioTax 2011\install\StudioTaxX64.mst" AI_SETUPEXEPATH="Z:\home\focht\Downloads\StudioTax2011Install.exe" SETUPEX"...,00000000,00000000,00000000,00000020,00000000,00000000,0032f684,0032f674) ret=0044ae41 002b:Call KERNEL32.__wine_kernel_init() ret=7bc534a2 0024:Ret KERNEL32.CreateProcessW() retval=00000001 ret=0044ae41 ... 002b:Call KERNEL32.CreateNamedPipeW(00542950 L"\\.\pipe\ToServer42",00000003,00000006,000000ff,00000400,00000400,00001388,00000000) ret=00475fc9 002b:Ret KERNEL32.CreateNamedPipeW() retval=0000008c ret=00475fc9 ... 002b:Call KERNEL32.CreateProcessW(005426a8 L"Z:\home\focht\Downloads\StudioTax2011Install.exe",00544028 L"Z:\home\focht\Downloads\StudioTax2011Install.exe /i "C:\users\focht\Application Data\BHOK IT Consulting\StudioTax 2011\install\StudioTaxX64.msi" EXECUTEACTION="INSTALL" SECONDSEQUENCE="1" CLIENTPROCESSID="42" AI_MORE_CMD_LINE=1",00000000,00000000,00000000,00000000,00000000,00000000,0033ef10,0033ef74) ret=0047c6fb 0034:Call KERNEL32.__wine_kernel_init() ret=7bc534a2 002b:Ret KERNEL32.CreateProcessW() retval=00000001 ret=0047c6fb ... 0034:Call KERNEL32.CreateFileW(0052ef78 L"\\.\pipe\ToServer42",c0000000,00000000,00000000,00000003,00000000,00000000) ret=00475fef 0034:Ret KERNEL32.CreateFileW() retval=00000040 ret=00475fef ... 0034:Call KERNEL32.WriteFile(00000040,0052ef78,00000012,0033f7a4,00000000) ret=004762e2 002b:Ret KERNEL32.ConnectNamedPipe() retval=00000001 ret=004761a9 0034:Ret KERNEL32.WriteFile() retval=00000001 ret=004762e2 0034:Call KERNEL32.ReadFile(00000040,0052e720,00000400,0033f718,00000000) ret=0047620c 002b:Call KERNEL32.ReadFile(0000008c,00543818,00000400,0033ee08,00000000) ret=0047620c 002b:Ret KERNEL32.ReadFile() retval=00000001 ret=0047620c 002b:Call ntdll.RtlAllocateHeap(00528000,00000000,00000020) ret=0048bc1f 002b:Ret ntdll.RtlAllocateHeap() retval=00541860 ret=0048bc1f 002b:Call KERNEL32.PeekNamedPipe(0000008c,00000000,00000000,00000000,00000000,0033ee10) ret=00476257 002b:Ret KERNEL32.PeekNamedPipe() retval=00000001 ret=00476257 ... 0034:Call KERNEL32.lstrcmpiW(0033f7bc L"SET",004c28d4 L"/promptrestart") ret=00437bc0 0034:Ret KERNEL32.lstrcmpiW() retval=00000001 ret=00437bc0 0034:Call KERNEL32.lstrcmpiW(0033f7bc L"SET",004c28b8 L"/forcerestart") ret=00437bec 0034:Ret KERNEL32.lstrcmpiW() retval=00000001 ret=00437bec 0034:Call KERNEL32.lstrcmpiW(0033f7bc L"SET",004c28ac L"/log") ret=00437c18 0034:Ret KERNEL32.lstrcmpiW() retval=00000001 ret=00437c18 0034:Call ntdll.RtlAllocateHeap(00528000,00000000,00000004) ret=0048bc1f 0034:Ret ntdll.RtlAllocateHeap() retval=0052fe40 ret=0048bc1f 0034:Call ntdll.RtlAllocateHeap(00528000,00000000,00000200) ret=0048bc1f 0034:Ret ntdll.RtlAllocateHeap() retval=00530c48 ret=0048bc1f 0034:Call user32.LoadStringW(00400000,00002792,00530c48,00000100) ret=00448dad 0034:Ret user32.LoadStringW() retval=00000014 ret=00448dad ... 0034:Call user32.MessageBoxW(00000000,005309a0 L"Invalid command line",00000000,00000010) ret=0040a9c0 ... <server exits with message box dismissed> ... 002b:Call KERNEL32.WriteFile(0000008c,0033ef94,00000002,0033eea4,00000000) ret=00476339 002b:Ret KERNEL32.WriteFile() retval=00000000 ret=00476339 002b:Call KERNEL32.FlushFileBuffers(0000008c) ret=00476342 002b:Ret KERNEL32.FlushFileBuffers() retval=00000001 ret=00476342 002b:Call KERNEL32.ConnectNamedPipe(0000008c,00000000) ret=004761a9 002b:Ret KERNEL32.ConnectNamedPipe() retval=00000000 ret=004761a9 002b:Call KERNEL32.GetLastError() ret=004761b6 002b:Ret KERNEL32.GetLastError() retval=00000450 ret=004761b6 ... <client continues to churn CPU> --- snip ---
There are three "StudioTax2011Install.exe" processes.
1) bootstrapper 2) client (creates named pipe "ToServer") 3) server (opens the end)
After the server (3) terminated due to command line error the client (2) still tries to reconnect the pipe, churning CPU.
From a quick glance it might be a bug in the app installer itself (bug hidden
in Windows due to different stack usage of APIs). The command line parsing and error handling code doesn't really look robust (doesn't check if FormatMessage() failed due to untranslatable error -> accesses buffer content unconditionally and the like).
$ du -sh StudioTax2011Install.exe 24M StudioTax2011Install.exe
$ sha1sum StudioTax2011Install.exe bc4f4afcf7e3556a29e03ddfbdab2d2fded2aff0 StudioTax2011Install.exe
$ wine --version wine-1.5.0
Regards
http://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #10 from Justin Chevrier jchevrier@gmail.com 2012-04-23 21:08:23 CDT --- Created attachment 39899 --> http://bugs.winehq.org/attachment.cgi?id=39899 Patch adding TRUST_E_NOSIGNATURE kernel32 resource
http://bugs.winehq.org/show_bug.cgi?id=29997
Justin Chevrier jchevrier@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |jchevrier@gmail.com
--- Comment #11 from Justin Chevrier jchevrier@gmail.com 2012-04-23 21:08:54 CDT --- I believe I have made some headway on this issue. I run into the same crash as the original reporter. Doing a '+relay,+seh' I get the following:
--- snip --- 0032:Ret msi.MsiGetFileSignatureInformationW() retval=800b0100 ret=0047bd53 0032:Call KERNEL32.FormatMessageW(00001300,00000000,800b0100,00000400,0033f0b8,00000000,00000000) ret=0047cabe 0032:Ret KERNEL32.FormatMessageW() retval=00000000 ret=0047cabe trace:seh:raise_exception code=c0000005 flags=0 addr=0x48a113 ip=0048a113 tid=0032 trace:seh:raise_exception info[0]=00000000 trace:seh:raise_exception info[1]=00000000 trace:seh:raise_exception eax=00000000 ebx=0033f1a4 ecx=0033f120 edx=00000000 esi=0033f120 edi=0033f120 trace:seh:raise_exception ebp=0033f07c esp=0033f07c cs=0023 ds=002b es=002b fs=0063 gs=006b flags=00010246 --- snip ---
It looks like the installer is calling FormatMessageW with a message ID of: 0x800b0100 (TRUST_E_NOSIGNATURE) and expecting a string to be returned from a resource contained in kernel32. As there is no kernel32 resource for 0x800b0100 FormatMessageW fails and returns a zero. I guess the installer doesn't handle this situation. Anyway, I have attached a patch that adds this resource in, and modifies the wmc tool to accept a messageid larger than 0xffff. No idea if there are consequences for changing the maximum allowed size, or if there is a more correct way to go about this.
With the patch applied I get past the crash and then the installer trips on: --- snip --- err:msi:ITERATE_Actions Execution halted, action L"AppSearch" returned 1615 --- snip ---
http://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #12 from Hans Leidekker hans@meelstraat.net 2012-04-24 02:37:18 CDT --- (In reply to comment #11)
With the patch applied I get past the crash and then the installer trips on: --- snip --- err:msi:ITERATE_Actions Execution halted, action L"AppSearch" returned 1615 --- snip ---
Can you please open a new bug for this and attach a +msi,+msidb trace there?
http://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #13 from Justin Chevrier jchevrier@gmail.com 2012-04-24 18:04:44 CDT --- Bug 30516 has been created for the msi related hang.
http://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #14 from sebastiendevel@gmail.com 2012-09-22 18:53:39 CDT --- Created attachment 41778 --> http://bugs.winehq.org/attachment.cgi?id=41778 Installer path not recognized
This happens with wine 1.5.13 compile on Ubuntu 10.04 LTS and .net sp2 installed with winetricks, winearch 32 bit
http://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #15 from sebastiendevel@gmail.com 2012-09-22 18:56:05 CDT --- Created attachment 41779 --> http://bugs.winehq.org/attachment.cgi?id=41779 log of installer path error
Log file generated with wine 1.5.13 compiled on Ubuntu 10.04 LTS, winearch 32 bits
http://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #16 from sebastiendevel@gmail.com 2012-09-22 19:02:37 CDT --- I tried again since I saw that some msi hang bug had been closed, which I did not experience, but now I get a weird error from the installer showing me a mangled install path. Compiled the latest 1.5.13 since there's no PPA for lucid anymore. I was going to try in a 12.04 vm to see if I get the same result.
http://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #17 from Austin English austinenglish@gmail.com 2013-09-19 22:33:01 CDT --- After (In reply to comment #9)
Hello,
I don't get a crash using clean WINEPREFIX but a message box is shown, stating "Invalid command line".
The same in wine-1.7.2-106-g7f859e5, with either winetricks dotnet20, or wine-mono.
http://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #18 from Justin Chevrier jchevrier@gmail.com 2013-09-20 09:24:00 CDT --- Bug 32988 has a workaround for the 'Invalid command line' error dialog.
http://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #19 from Austin English austinenglish@gmail.com 2013-09-23 14:13:58 CDT --- (In reply to comment #18)
Bug 32988 has a workaround for the 'Invalid command line' error dialog.
Still fails for me if run from ~/.wine/drive_c.
http://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #20 from Justin Chevrier jchevrier@gmail.com 2013-12-08 10:56:07 CST --- (In reply to comment #19)
(In reply to comment #18)
Bug 32988 has a workaround for the 'Invalid command line' error dialog.
Still fails for me if run from ~/.wine/drive_c.
I think the issue here that publisher keeps updating the executable on their site. The crash I identified related to TRUST_E_NOSIGNATURE no longer seems to be an issue with the current version on their site. I'm running a clean prefix (with and without 'dotnet40') on wine 1.7.8 and the latest executable:
MD5Sum: 2930a1beff76babc503b2f0f707feb97 StudioTax2011Install.exe
and don't get the 'Invalid Command Line' dialog; instead the installer just hangs after accepting the license agreement and clicking 'Install'. The other issue is that I have revert to a clean prefix on every install attempt because on the second run the installer crashes much earlier while attempting to 'Computing space requirements' (see attached image). Can you confirm, or are you still recieving the 'Invalid Command Line' dialog?
http://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #21 from Justin Chevrier jchevrier@gmail.com 2013-12-08 10:56:55 CST --- Created attachment 46797 --> http://bugs.winehq.org/attachment.cgi?id=46797 Crash on second run attempt
https://bugs.winehq.org/show_bug.cgi?id=29997
super_man@post.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |super_man@post.com
--- Comment #22 from super_man@post.com --- The installer hangs while trying to install. Last line at console (which is printed few times)
fixme:shell:SHGetKnownFolderPath flags 0x00004400 not supported
wine 1.7.51
https://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #23 from super_man@post.com --- Doesnt crash, but it never seems to finnish.
wine 1.8.rc2
https://bugs.winehq.org/show_bug.cgi?id=29997
--- Comment #24 from super_man@post.com --- (In reply to super_man from comment #23)
Doesnt crash, but it never seems to finnish.
wine 1.8.rc2
The same 1.9.7 and wine-staging 1.9.6
https://bugs.winehq.org/show_bug.cgi?id=29997
winetest@luukku.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |winetest@luukku.com
--- Comment #25 from winetest@luukku.com --- (In reply to super_man from comment #24)
(In reply to super_man from comment #23)
Doesnt crash, but it never seems to finnish.
wine 1.8.rc2
The same 1.9.7 and wine-staging 1.9.6
The isntaller fails really early. I quess it's related to this.
err:msidb:msi_table_load_transform borked.
wine 2.0rc3 and staging same version.
https://bugs.winehq.org/show_bug.cgi?id=29997
Frédéric Delanoy frederic.delanoy@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords|dotnet | Resolution|--- |FIXED CC| |frederic.delanoy@gmail.com Status|NEW |RESOLVED
--- Comment #26 from Frédéric Delanoy frederic.delanoy@gmail.com ---
(In reply to winetest from comment #25)
(In reply to super_man from comment #24)
(In reply to super_man from comment #23)
Doesnt crash, but it never seems to finnish.
wine 1.8.rc2
The same 1.9.7 and wine-staging 1.9.6
The isntaller fails really early. I quess it's related to this.
err:msidb:msi_table_load_transform borked.
wine 2.0rc3 and staging same version.
Got same issue with wine-2.5-52-g8b1737c0fd
Original bug doesn't appear anymore though, so this bug can be closed.
I created bug 42771 for the msi issue.
https://bugs.winehq.org/show_bug.cgi?id=29997
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #27 from Alexandre Julliard julliard@winehq.org --- Closing bugs fixed in 2.6.