Transient Windows test failures

[Matt Harbison]

On Sun, 11 Jun 2017 22:38:23 -0400, Sean Farley <sean at farley.io> wrote:

> Matt Harbison <mharbison72 at gmail.com> writes:
>
>> In the last day, I've had a handful of tests fail like so, over several
>> runs:
>>
>> diff --git a/tests/test-push-warn.t b/tests/test-push-warn.t
>> --- a/tests/test-push-warn.t
>> +++ b/tests/test-push-warn.t
>> @@ -589,10 +589,8 @@
>>       $ hg push inner
>>       pushing to inner
>>       searching for changes
>> -  adding changesets
>> -  adding manifests
>> -  adding file changes
>> -  added 1 changesets with 1 changes to 1 files (+1 heads)
>> +  abort: Provider DLL failed to initialize correctly
>> +  [255]
>>
>>       $ cd ..
>>
>> They are different each time, and don't fail twice in a row.  The  
>> previous
>> time I ran a test, was off of hg-committed (though the fact that it
>> doesn't happen every time doesn't preclude a prior change):
>>
>> (562ddb093b50) run-tests: add a way to list tests, with JSON and XUnit
>> support
>>
>>
>> About a week ago I also started noticing occasional errors like these:
>>
>> --- c:/Users/Matt/projects/hg/tests/test-graft.t
>> +++ c:/Users/Matt/projects/hg/tests/test-graft.t.err
>> @@ -324,12 +324,14 @@
>>
>>     Graft again onto another branch should preserve the original source
>>       $ hg up -q 0
>> +  abort: Invalid Signature
>> +  [255]
>>       $ echo 'g'>g
>>       $ hg add g
>>       $ hg ci -m 7
>> -  created new head
>>       $ hg graft 7
>> -  grafting 7:ef0ef43d49e7 "2"
>> +  skipping ancestor revision 7:ef0ef43d49e7
>> +  [255]
>>
>>       $ hg log -r 7 --template '{rev}:{node}\n'
>>       7:ef0ef43d49e79e81ddafdc7997401ba0041efc82
>>
>> I did manage to get a stacktrace with this error message (and it looked
>> like it was in python code), but I can't find it now.
>>
>> No idea what they mean.  I do see google hits on the first one that are
>> python related, but don't seem to be relevant.  I don't see anything  
>> about
>> the second one.  I'm wondering if anybody else doing Windows work sees
>> anything like this...
>
> Hi Matt,
>
> (sorry if you've answered this before) Can you share how you set up this
> test machine so that I can set my own up? If you have answered this,
> then would you mind sharing a link to how this was done?
>
> Much thanks!

Nothing too special.

1) Win7 x64
2) Python 2.7.13 x64
3) MinGW  
(https://sourceforge.net/projects/mingw/files/Installer/mingw-get-setup.exe/download)
4) Visual Studio 2008
5) $PATH is:

$ echo $PATH
.
/usr/local/bin
/mingw/bin
/bin
/c/Python27/
/c/Python27/Scripts
/c/ProgramData/Oracle/Java/javapath
/c/Python27_x86/Lib/site-packages/PyQt4
/c/Program Files/Common Files/Microsoft Shared/Windows Live
/c/Windows/system32
/c/Windows
/c/Windows/System32/Wbem
/c/Windows/System32/WindowsPowerShell/v1.0/
/c/Program Files (x86)/ATI Technologies/ATI.ACE/Core-Static
/c/Program Files (x86)/Common Files/Roxio Shared/DLLShared/
/c/Program Files (x86)/Common Files/Roxio Shared/DLLShared/
/c/Program Files (x86)/Common Files/Roxio Shared/12.0/DLLShared/
/c/Program Files/WIDCOMM/Bluetooth Software/
/c/Program Files/WIDCOMM/Bluetooth Software/syswow64
/c/Program Files (x86)/cvsnt
/c/Program Files (x86)/GTK2-Runtime/bin
/c/Program Files/Git/cmd
/c/Program Files/TortoiseHg/
/c/Program Files/Mercurial/
/c/Program Files (x86)/Palm/PDK/bin
/c/Program Files (x86)/Palm/SDK/bin
/c/Program Files (x86)/Palm/SDK/bin/novacom
/c/Program Files/TortoiseHg
/c/Program Files/Mercurial
/c/Python27_x86
/c/Program Files (x86)/Beyond Compare 3
/c/Program Files/Mercurial-4.2
/c/Program Files/gettext-iconv/bin

Simply open the MSYS prompt, `cd /path/to/hg && make local`, then `cd  
tests && yes | ./run-tests --local -j9 -t9000 -i`.

I set this system up a little over 5 years ago, so the MSYS you get will  
be a little more up to date.  I can't believe that's the issue with the  
sudden appearance of the issue(s) though.  The only thing that's changed  
in the last week was swapping an HDD for an SSD.  I think I saw the  
"Invalid Signature" message before then, but it was a full week of use  
before the DLL issue popped up.  I've seen no indication that the drive is  
flaky.

I've got a similar MinGW install at work on an older 32 bit machine, and a  
current one on an x64 machine (both Win7).  I'll run that overnight some  
day this week on a loop.