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

20030325: rectilinear Global composites (cont.)

>From: Robert Mullenax <address@hidden>

Robert,

>BTW..on another subject, I would like to start making those rectilinear
>Global composites here at NSBF (the bug I sent you from Universal).  At
>Universal I did a temporary workaround by going back to McIDAS 7.806..and
>that works okay.

>At NSBF I don't have an extra machine to do that..  I know you are swamped and
>likely will be a while before you could look at it, but have you been able to
>duplicate the problem, so I know it's not something I am doing wrong?

I put this at the bottom of my high priority list of things to do, so I
hope to be able to look at the problem this coming week.  I will send
you information when I figure out what is going on.

By the way, not that this affects you directly but, Dave Santek of SSEC
found the bug that was causing vertical profile serving in McIDAS under
Linux.  The problem was not the vertical profile server or Linux.  The
problem was a small bug in the client side module that is doing the
vertical profile (i.e., sounding) request, m0vpget.c.  The routine was
sending an extra 4 bytes to the server, and the server code was
apparently reading it into a buffer that was not dimensioned to be able
to handle it (this is my view of what was going on, not SSEC's).  Some
structure was then being stepped on resulting eventually in the
connection being dropped by the OS.  I will be releasing my addendum
now that this bug has been found and fixed ASAP next week.  The thing
that the release is waiting on is a bugfix in NLDNDISP related to lists
of data from OUT=LIST.

Tom
Unidata WWW Service              http://my.unidata.ucar.edu/content/support    
  

>From address@hidden Sat Mar 29 16:36:32 2003

> >BTW..on another subject, I would like to start making those rectilinear
> >Global composites here at NSBF (the bug I sent you from Universal).  At
> >Universal I did a temporary workaround by going back to McIDAS 7.806..and
> >that works okay.
> 
> >At NSBF I don't have an extra machine to do that..  I know you are swamped
> >and likely will be a while before you could look at it, but have you been
> >able to duplicate the problem, so I know it's not something I am doing wrong?
> 
> I put this at the bottom of my high priority list of things to do, so I
> hope to be able to look at the problem this coming week.  I will send
> you information when I figure out what is going on.

Okay, I knew you likely hadn't bene able to look into, I just thought I
check to see if you had duplicated it or not.  I know all you guys are
swamped..of course we all are these days.

> By the way, not that this affects you directly but, Dave Santek of SSEC
> found the bug that was causing vertical profile serving in McIDAS under
> Linux.  The problem was not the vertical profile server or Linux.  The
> problem was a small bug in the client side module that is doing the
> vertical profile (i.e., sounding) request, m0vpget.c.  The routine was
> sending an extra 4 bytes to the server, and the server code was
> apparently reading it into a buffer that was not dimensioned to be able
> to handle it (this is my view of what was going on, not SSEC's).  Some
> structure was then being stepped on resulting eventually in the
> connection being dropped by the OS.  I will be releasing my addendum
> now that this bug has been found and fixed ASAP next week.  The thing
> that the release is waiting on is a bugfix in NLDNDISP related to lists
> of data from OUT=LIST.

Good, I am glad.  With Sun behind Solaris x86 again I doubt I will go
with Linux if/when we get a new box..but you never know what Sun will
do.

I do have a bug with EU in McIDAS2002 that I will report tonight or
tommorrow.  I want to get all the info I can as to when the problem
occurs (basically EU LIST seems not to work properly at times).  Majes
it real hard whe trying to modify an enhancement.

Robert