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

951114: netcdf-2.4-beta2 installation on cray

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: 951114: netcdf-2.4-beta2 installation on cray
  • Date: Wed, 15 Nov 95 09:55:20 -0700
Christi,

>Date: Wed, 15 Nov 1995 09:37:32 -0700 (MST) 
>From: Christi Forsythe <address@hidden>
>Organization: Sandia National Laboratory
>Subject: Re: 951114: netcdf-2.4-beta2 installation on cray 
>Keywords: 199511142300.AA06524

In the above message you wrote:

> I tried the uname -a command and this is what I got
> 
> sacj001(42)% uname -a
> sacj001 sacj001 8.0.4 roo.4 CRAY J90
> sacj001(43)% 
> 
> What should be the result of uname -a? 

Well, on our Cray Y-MP, the first output field is the serial number of
the machine (even though it's supposed to be the name of the operating
system):

    41-shavano% uname -a
    sn1036 shavano 8.0.4 cbr.5 CRAY Y-MP

The configure script uses the output to determine the operating system.
Apparently, this fails on your J90 because the output from the command
is different (even though they're both running UNICOS 8.0.4!).

I'm beginning to dislike UNICOS.  Not only is it's uname(1) 
non-standard, but it's not even consistent.  ;-(

I suggest explicitly setting the operating system type by setting the
variable OS to `unicos' before running the configure script, e.g.

    % setenv OS unicos
    % rm config.cache
    % ./configure >&! configure.log

Send me the log file if something goes wrong.  Otherwise, please let me
know if this fixes the problem.

--------
Steve Emmerson   <address@hidden>