common tobuild,1.345,1.346

seth vidal skvidal at phy.duke.edu
Tue May 10 20:30:21 UTC 2005


On Tue, 2005-05-10 at 15:25 -0500, Quentin Spencer wrote:
> seth vidal wrote:
> 
> >On Tue, 2005-05-10 at 21:51 +0200, Michael Schwendt wrote:
> >  
> >
> >>On Tue, 10 May 2005 14:27:11 -0500, Quentin Spencer wrote:
> >>    
> >>
> >>>I just got a notice that the build of octave failed. The only thing the 
> >>>build log says is: "Build process terminated by timeout".
> >>>      
> >>>
> >>It's a large package and builds for a long time. Most likely exceeds
> >>the time-out set in the build system, which perhaps is still to short.
> >>    
> >>
> >it's an hour and a half, iirc.
> >probably there is another error that's causing it. I've seen a case
> >where the popen calls can hang up b/c of too much error output. I think
> >I know a fix but I've not found a way to replicate it yet :(
> >  
> >
> FWIW, octave is a big package that takes a long time to build (close to 
> an hour on a relatively modern system is common), and it generates a lot 
> of output. The spec uses the --enable-picky-flags option on configure (I 
> don't know why--it was that way in core) which generates lots of 
> warnings and could potentially double the size of the log output. If 
> this can potentially help it compile, I'd be happy to change it.
> 

does octave time out EVERY time?

if so then I'd like to use it to test some things.

-sv





More information about the fedora-extras-commits mailing list