[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

20000428: update on 17 April McIDAS bugfix and AVN 12Z run (cont.)



>From: weather <address@hidden>
>Organization: NMSU/NSBF
>Keywords: 199909242206.QAA23500 McIDAS-XCD DMGRID AVN grid missing SC5.0

Robert,

Strike two :-(

I just compared the number of grids in the AVN GRID files for the
April 28 run 0 Z run on four different machines.  What I found gave me
that old sinking feeling:

machine        dataset         #MERC  #PS    Total  OSystem  Compilers
--------------+---------------+------+------+------+--------+---------
adde.unidata   RTGRIDS/AVN.9   368    1212   1580   x86      SC5.0
wxmcidas.nsbf  RTGRIDS/AVN.9   368    1212   1580   x86      SC5.0
windfall.evsc  RTGRIDS/AVN.9   368    1212   1580   SPARC    SC4.2
psnldm.nsbf    RTGRIDS/AVN.9   368    1212   1580   SPARC    SC5.0

adde.unidata   RTGRIDS/AVN.19  296     713   1009   x86      SC5.0
wxmcidas.nsbf  RTGRIDS/AVN.19  296     713   1009   x86      SC5.0
windfall.evsc  RTGRIDS/AVN.19  296     713   1009   SPARC    SC4.2
psnldm.nsbf    RTGRIDS/AVN.19  254     565    819   SPARC    SC5.0

adde.unidata   RTGRIDS/AVN.29  148     612    760   x86      SC5.0
wxmcidas.nsbf  RTGRIDS/AVN.29  148     612    760   x86      SC5.0
windfall.evsc  RTGRIDS/AVN.29  148     612    760   SPARC    SC4.2
psnldm.nsbf    RTGRIDS/AVN.29  133     212    345   SPARC    SC5.0

This means that one of three things is possible:

o a change was made to one of the XCD configuration files after the AVNs
  started coming in (you would have had to have done this; I didn't).
  I don't have any hope that this is what happened, but...

o there is something wrong with the dmgrid.k executable on your SPARC.
  This could be a Sun SC5.0 thing that we were facing before without
  realizing it.  I offer this as a possibility since 'windfall' (machine
  at UVa) is a Sun SPARC, but they use the Sun SC4.2 compilers.

o there is some sort of systematic problem on psnldm.  What this could
  be, I have no idea.

As soon as you assure me that the first option above is not a possibility,
we can two different options:

o FTP and rebuild the distribution from scratch on psnldm.  This will
  eliminate the possibility that some other routine that should have been
  recompiled wasn't.

o FTP the UVa version of dmgrid.k to your machine and let it run there.  This
  would eliminate the possibility that psnldm has a some sort of systematic
  problem.

Tom