On 11/22/12 2:15 PM, Peter Fennell wrote:
> Alexander,
> 
> Thanks for your reply
> 
>>Inline text actually works better for us, because it's easier for
>>searches of the mailing list archives.  Here's the contents:
> 
> Apologies. I'm not familiar with the etiquette.

No problem.  That's why I told you. :-)

> 
>>Your snippet doesn't indicate that, or anything that
>>looks obviously like an error.  Is this where the console output
>>stopped, too?  Both times?
> 
> The installation didn't naturally stop, I stopped it. At that stage it
> was running for 48 hours and I was advised that it shouldn't take that
> long. Through reading the output it didn't look like there were any
> errors, it just seemed to be verifying and running lots of different things.
> 

The time really to concerned is if you get to a point where your get no
additional output for an extended period of time, several minutes
perhaps, because that indicates that something is hanging.  If you're
still getting new output to the terminal that usually means things are OK.

> 
>>3)  Try again using the "-l" option to get a full log file (in /tmp),
>>e.g. via "fink -l install atlas".  You can send me that log file
>>off-list.  That will make it clearer what exactly is going on.
> 
> I tried to locate the log file in /tmp but it didn't exist; is a log
> file still generated if the installation is prematurely stopped? I'll
> rerun it with the -l option and send it to you

The log file only gets generated if you use the -l option.  It gets
continually appended to as the build proceeds, so if you stop the build
the log will still be present, and will indicate that the build was
manually stopped.

> 
>>4)  If you're in a hurry, I can suggest the other variants of Octave:
>>just 'octave', which uses a built-in system framework instead of ATLAS;
>>or 'octave-ref', which uses a non-optimized, and therefore quicker to
>>build, implementation of the same libraries that go into ATLAS.
> Thanks for the advice. I would do that, but I read that octave-atlas
> caters for some bugs/features of the X86_64 architecture. Do you know
> otherwise?
> 
> Thanks again
> 
> Peter
> 
> 
> 

I'm the maintainer for Octave, so yes, because I set it up.  :-)

The issue I know about is that on 10.6 Octave (new versions, anyway)
can't be built on x86_64 without using either ATLAS or the non-optimized
BLAS/LAPACK in lapack342.  On 10.7 and 10.8 this doesn't apply, and it
can use the system's Accelerate framework.

This gets handled automatically, anyway.  There wouldn't be an "octave"
package for your configuration if it didn't work (barring screwups on my
part :-) ).
-- 
Alexander Hansen, Ph.D.
Fink User Liaison
My package updates: http://finkakh.wordpress.com/

------------------------------------------------------------------------------
Monitor your physical, virtual and cloud infrastructure from a single
web console. Get in-depth insight into apps, servers, databases, vmware,
SAP, cloud infrastructure, etc. Download 30-day Free Trial.
Pricing starts from $795 for 25 servers or applications!
http://p.sf.net/sfu/zoho_dev2dev_nov
_______________________________________________
Fink-users mailing list
[email protected]
List archive:
http://news.gmane.org/gmane.os.macosx.fink.user
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-users

Reply via email to