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

[McIDAS #HIF-234168]: Solaris 10 and McIDAS-X update

Hi Robert,

> Do you have an update on the latest McIDAS release not compiling on Solaris
> x86 due to netCDF issues?

I apologize that I have not had enough time to fully investigate the problem
you are reporting!

> Basically if you guys (GEMPAK seems to have) defacto quit Solaris x86 I need
> to know.

Quitting support of Solaris x86 is about as far from our reality as possible:
our NOAAPort ingest systems here in UCAR run under Solaris x86; our 
servers (motherlode.ucar.edu, atm.cise-nsf.gov, and the soon to be installed
new version of unidata2.ssec.wisc.edu) all run Solaris x86; and more than one
staff member uses it as his primary workstation.

** To be clear: We are NOT contemplating dropping support for Solaris x86 **

As far as GEMPAK on Solaris x86: Michael has been working on updating the code
that composites NEXRAD Level III radar into a national view.  The big job he
has faced is NCEP's _weak_ support for the new Level III products (e.g., NxQ,
the "high resolution" reflectivity products; etc.).  NCEP's approach to 
the new images in their new releases of GEMPAK was to write routines that 
the resolution back to what is in the NxR (reflectivity), etc. products.  We 
that this approach is not acceptable for our community, so a lot of unraveling 
their changes had to be made.  The good news is twofold:

- Michael has successfully updated the compositing code

- he will be implementing that code on one of our server machines (likely 
  to test the production of the Level III composites that we distribute in the 
  feed.  Again, our server machines are running Solaris x86, so the GEMPAK build
  process for Solaris x86 will be vetted in a production environment

One comment in Michael's defense:  NCEP threw us a curve ball with their "final"
GEMPAK release (5.11.4) -- they implemented a 'make' approach to building the 
package that
was completely different than the one that Chiz implemented for Unidata GEMPAK 
(the NCEP
build always used build scripts; you likely never knew this because of the work 
Chiz did).  The job of unraveling what NCEP did and folding in their code 
changes into a build
that Unidata GEMPAK users would be familiar with has been daunting to say the
least.  I dare say that even Chiz (who I view as _the_ GEMPAK authority in the 
would have found the task difficult.  Michael is a junior guy, but he is 
expert in all things GEMPAK at an impressive rate while also working with 
to help make it usable by the community.

> I want to apologize for the tone of my emails. Being stressed out over
> the future of a lot of things is no excuse.

No worries.  I understand being stressed-out about things :-(

Back to the build on Solaris x86:  a quick look at the error you reported
in your original email shows:

- the netCDF component of the distribution got built in 64-bit mode

- the main portion of the McIDAS build (the routines in ~mcidas/mcidas2009/src)
  got built in 32-bit mode

What I now have to figure out is what flag to pass to the netCDF configure to 
it to build in the same mode that the rest of McIDAS is built in.  I will
get together with the netCDF guys today during a break in our workshops 
to learn what I need to do (I am helping in with the LDM workshop today and
am co-presenting the IDV workshop on Wed-Fri).

Again, I apologize for not being able to get back to you on your problem in a
timely manner!


Unidata User Support                                    UCAR Unidata Program
(303) 497-8642                                                 P.O. Box 3000
address@hidden                                   Boulder, CO 80307
Unidata HomePage                       http://www.unidata.ucar.edu

Ticket Details
Ticket ID: HIF-234168
Department: Support McIDAS
Priority: Normal
Status: Closed

NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.