[McIDAS #JPC-196190]: mcidas problems

Hi Jennie,

Sorry for the tardy reply...
 
> Well, I suspect you are down in snowy? San Antonio, so I expect
> it might be hard to get any quick replies.

Sorry.  I opted to not pay for Internet service in my hotel room, and
the booth was always busy enough that doing support was not possible.

> I have some problems
> with windfall, I am not getting any grid files.  see below:
> windfall: /home/mcidas/uvaworkdata $ more XCD_START.LOG
> Starting DDS at 07016.224631
> Domestic Data Service
> 016.224630
> no processes are currently acHigh Resolution Data Service

Hmm...

> I probably created this problem myself.  My machine had run out
> of space on the main partition (ugly), too many admin messages,
> etc.  The lack of space generated an additional cascade of
> problems.  The long and short is that I started getting core
> files in the uvaworkdata directory, including ugly things like
> FRAME.001, etc.  I deleted a bunch of files with the same time
> stamp, and I am thinking I deleted something crucial, a file with
> decoder info maybe?

Once files like FRAME.001 are created in a user's $MCDATA directory
(~mcidas/workdata for the user 'mcidas'), they will be used in preference
to ones that should be created in a transient, session-specific directory
under ~/.mctmp.  I am sure that your running out of disk space caused
the file(s) to be created in the wrong place, but it would be very
difficult to figure out exactly what happened.

The best course of action at this point is to clean up a bit and then
rerun the XCD setup scripts to insure that the files needed for
decoding are complete/correct and in the appropriate locations.

> I have a product queue, and an HRS.spl file, I believe grids are
> getting passed to my machine, but I am not decoding any.

OK.

> I am VERY happy to report that I finally convinced the department
> to provide me with funds to buy a new linux box to replace
> windfall.

Yahoo!  We are thrilled with some new machines we recently built (purchased
the motherboards, etc.): dual dual-core Xeons with lots of memory.  We
took a couple of these boxes to AMS, and they worked great.

> I may try to get another machine like the Dell box I
> bought on a NASA grant last year, where I was running GEMPAK.
> BUT, in the meantime (which will probably be a month or two?) I
> would really like to maintain my wv scripts, and a few other
> elements of the real-time products on windfall.

OK.

> Don't know if you can still login to user mcidas to investigate
> what havoc I wreaked, but I could use some help re-establishing
> grid decoding.

I just tried to SSH to windfall as 'ldma' and 'root' from both my
home machine (I am still at home at the moment) and from
laraine.unidata.ucar.edu.  The attempt as 'ldm' failed quickly
giving me the message:

Connection closed by 128.143.42.190

The attempt as 'root' failed since 'root's password seems to have
changed.

If you could provide me the current 'ldm', 'mcidas', and 'root' password
(root's would be sufficient, but I want to update my records), I
can login and get things working again.  While doing this, I would like
to upgrade the LDM to a current release (you are running a very old
version, 6.0.14), but I want to check with you first before doing so.

If/when you decide to give me the new passwords, please send email to
my private, Unidata email address and do NOT mention either the
name of the machine that they are for or the account names (for security).
It would also be best to not include the word 'password' (or 'pass', etc.)

> Thanks.  Hope the AMS is great, wish I was there.

The show was interesting as always, but the weather did not cooperate.
Ethan Davis' flight back to Boulder on Wednesday was cancelled, and it
looked like the earliest one he could get back was on Saturday morning
(but that changed to today (Friday) at 2 pm).

Anyway, where we were stayed mostly wet, not icy although metal surfaces
had icicles hanging from them.  Several people slipped and fell badly
when stepping on things like grates, manhole covers, etc.  Luckily,
I managed to stay upright :-)

Cheers,

Tom
****************************************************************************
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: JPC-196190
Department: Support McIDAS
Priority: Normal
Status: Closed