[Bioclusters] Wrong system and load values from GE on MacOS X

[Ron Chen]

The port available on bioteam.net was the first
implementation. (btw thanks bioteam.net for the port!)

Later on, other people added more fixes -- detecting
the number of processors was one of them (added by
myself if I remember correctly).

Sun is planning to provide binaries on
gridengine.sunsource.net, but they are waiting for a
few more fixes before do that.

If bioteam.net can update the binary to the current
version -- SGE 5.3p4, then we don't need to wait for
Sun's binary.

Lastly, you can get further help from the SGE dev
mailing list if you want to compile from source:

http://gridengine.sunsource.net/project/gridengine/maillist.html

 -Ron


--- "Beltrami, Riccardo" <Riccardo_Beltrami@Chiron.it>
wrote:
> Dear all,
> 
> I'm trying to add few MacOS X execution nodes in our
> experimental Grid 
> Engine cluster.
> 
> The GE port I'm using is the Van Etten's 5.3p2
> version available as a 
> package on the BioTeam web site.
> 
> It's working ok however when I run qhost as user
> sgeadmin on an 
> administration host I get:
> 
> HOSTNAME             ARCH       NPROC  LOAD   MEMTOT
>   MEMUSE   SWAPTO   SWAPUS
>
-------------------------------------------------------------------------------
> ...
> stark                darwin         1     -      0.0
>        -      0.0        -
> 
> where stark is my MacOS X execution host.
> 
> The problem is that stark is a dual processor PPC G4
> 1.25GHz with 1.5GB of 
> RAM so the NPROC, MEMTOT etc. values are wrong
> reported.
> 
> In fact, if I run the command loadcheck found in
> $SGE_ROOT/utilbin/darwin 
> directory as user root this is what I get:
> 
> # ./loadcheck
> arch            darwin
> num_proc        1
> load_short      0.00
> load_medium     0.00
> load_long       0.00
> mem_free        0.000000M
> swap_free       0.000000M
> virtual_free    0.000000M
> mem_total       0.000000M
> swap_total      0.000000M
> virtual_total   0.000000M
> mem_used        0.000000M
> swap_used       0.000000M
> virtual_used    0.000000M
> cpu             0.0%
> 
> Those values are clearly wrong, and if I run the
> same command on other 
> execution host (for example Solaris/Sparc machine) I
> get correct values.
> 
> Stark runs MacOS X 10.2.6 with all the latest
> patches available from Apple.
> 
> Does anybody know how to fix/solve this problem?
> 
> Tia, Riccardo.
> 
> 
> 
> 
>                           O
>                          / \
>  
> ,---------------------+---+-----------------------.
>   | Riccardo Beltrami                              
> |-.
>   |                                                
> | |
>   | Bioinformatics                                 
> | |
>   |                                                
> | |
>   | riccardo_beltrami@chiron.it                    
> | |
>   |                                                
> | |
>  
> `-------------------------------------------------'
> |
>   
> `--------------------------------------------------'
> _______________________________________________
> Bioclusters maillist  - 
> Bioclusters@bioinformatics.org
>
https://bioinformatics.org/mailman/listinfo/bioclusters


__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
http://sitebuilder.yahoo.com