On Wed, Oct 2, 2013 at 7:36 AM, Tom Kent
On Mon, Sep 30, 2013 at 6:10 PM, Tom Kent
wrote: On Mon, Sep 30, 2013 at 8:41 AM, Andrey Semashev < andrey.semashev@gmail.com> wrote:
On Monday 30 September 2013 07:46:31 Tom Kent wrote:
On Mon, Sep 30, 2013 at 3:34 AM, Andrey Semashev
wrote: On Wednesday 11 September 2013 09:34:55 Tom Kent wrote:
> I tried building the library and tests on my local machine many
and
> I've never seen linking to take 5 minutes. I have another setup > though. > Did > you notice any other abnormal behavior, such as memory growth or
swapping
> or compiler/linker crashes maybe? Does linking ever complete?
It doesn't seem to always be failing, and I don't have a good way
to
watch
it when the automated tests are run. Whenever I have manually
built it
on this machine the linking never failed.
Tom, the tests are continuing to fail and I don't know what can be done on Boost.Log side. Could you try to increase the timeout (to 10 minutes, for example)?
I'm not sure how, I think it needs to be added to the regression
times scripts or
test modules.
I think you just need to add --timeout=10 option to your run.py command line.
I've added this to all the builds running on teeks99-03, so if this works we should see it start getting better over the next couple days.
So, apparently it is something other than the timeout (or the timeout is longer than 10) because they are still failing with --timeout=10:
http://www.boost.org/development/tests/trunk/developer/output/teeks99-03d-wi...
Hang on, the timeout flag didn't showup in the run.py command line at the bottom, I just tried moving it further forward in the options, we'll see if it shows up now.