Open GoogleCodeExporter opened 9 years ago
What does the failure look like?
Original comment by jeff.br...@gmail.com
on 2 Nov 2009 at 10:28
Unfortunately I don't see anything in the log that would help to explain the
cause.
Try capturing a log with Debug verbosity. This will be very big but we can
throw out
everything before/after the Gallio test run.
You should see some messages from NCover get written out. I wouldn't be too
surprised if the problem were related to NCover's IPC mechanism. IIRC it
creates a
new IPC channel dynamically on each invocation but the naming scheme is
time-sensitive. I'd need to reverse-engineer that code again to find out. I
had
similar problems before when I used to run NCover v1.5.8 in-process rather than
spinning off a new process for each run because that version of NCover forgets
to
close its IPC channel...
Original comment by jeff.br...@gmail.com
on 3 Nov 2009 at 10:29
Any progress on this?
Original comment by uldis.ka...@gmail.com
on 26 Nov 2009 at 9:27
I can see a couple of problems here.
First there's this:
D:\work\projects\PRCHaT_Main\Nightly_Build\576432\SCM\build\msbuild\Scripts\Comm
on\Main\codecoverage.targets
: warning : Forcibly killing the host process!
D:\work\projects\PRCHaT_Main\Nightly_Build\576432\SCM\build\msbuild\Scripts\Comm
on\Main\codecoverage.targets(29,3):
error : Host process exited with code: -1
I have no idea why Gallio is killing the process. Usually it only does that
sort of
thing if the process hangs. In that case it would print some additional debug
output.
The other error is this:
Invalid NCover coverage xml file:
C:\Documents and Settings\depanthill\Local Settings\Temp\Gallio\NCover\b2ip3iki.tux
That's a little easier to understand. Because Gallio killed the previous
process, it
probably didn't write a complete coverage file to disk for that process. Later
when
Gallio tries to merge in that coverage fail, it fails. We should probably add a
check to ensure that the coverage file looks sane before merging it in.
Have you tried NCover 3?
Original comment by jeff.br...@gmail.com
on 5 Jan 2010 at 12:38
No, I haven`t because I have license only for NCover2.
Original comment by uldis.ka...@gmail.com
on 5 Jan 2010 at 7:23
I have the same problem in 3.1 build 397. Running through TeamCity and
MSBuildExtensions. I have switched from using the NCover(1) runner to
IsolatedProcess,
and I still get the issue intermittently. I have attached an excerpt from the
log
showing the exception that I am encountering.
Original comment by steve.cz...@gmail.com
on 15 Mar 2010 at 5:39
Attachments:
Original comment by Yann.Tre...@gmail.com
on 19 Jul 2010 at 9:16
Original comment by Yann.Tre...@gmail.com
on 21 Sep 2010 at 1:45
We are seeing the same thing running in a x64 environment (but single process).
This same process was working fine in our 32 bit build agent. We are running
Gallio under NCover 3 with TeamCity. We did take NCover and TeamCity out of
the picture and ran directly under Icarus on the agent and it raised a
threading exception.
Original comment by robertre...@gmail.com
on 11 Mar 2011 at 3:45
What's the exception? I really wouldn't recommend using Icarus for unattended
builds.
Original comment by grahamr...@gmail.com
on 11 Mar 2011 at 4:05
Original comment by Yann.Tre...@gmail.com
on 14 Jun 2011 at 5:53
Original issue reported on code.google.com by
uldis.ka...@gmail.com
on 29 Oct 2009 at 9:17