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

20001002: upgrading to the ldm-mcidas pnga2area decoder (cont.)

>From: Erick Lorenz <address@hidden>
>Organization: UC Davis
>Keywords: 200009220039.e8M0dBb07890 McIDAS-X 7.60 Linux

Erick,

>The ldm can be stopped and restarted quickly again

Good.  The problem was me adding FSL2 to the request line.  We are looking
into this here.

re: request for FSL2 data
>I don't think there is a demand here for wind profiler data currently.

OK.

>Also I am in no hurry to get NLDN data at present.

The reason I was adding this is that you used to get the data.

re: increase of queue from 100 MB to 200 MB
>After your last ministrations the LDM stopped saving files again so I
>increased the product queue to 400Mb.  Since then it has been behaving.

I am perplexed about this one.  Things should have worked fine with a
200 MB queue.  The "fix" may have been in deleting and recreating the
queue.

>The only problem is that we still are not getting GRID files.  I studied
>all the documentation on GRIB and XCD and retraced the steps to starting
>those processes up.  I did all this on Sunday afternoon and two new GRID
>files showed up in /var/data/xcd but when I came in today they were gone.
>I noticed in the troubleshooting chapter that there should only be one copy
>each of ingetext.k and ingebin.k running.  When I looked, sure enough there
>were two each.
>------------------
>[mcidas@ATM20 ~/data]$ ps -aux|grep inge
>ldm        902  0.0  0.3  1904   968  ?  S   23:53   0:00 ingetext.k DDS
>ldm        919  0.1  0.2  1952   732  ?  S   23:53   1:24 ingetext.k DDS
>ldm       8823  0.0  0.3  1788   956  ?  S   12:15   0:00 ingebin.k HRS
>ldm       8831  0.6  0.4  1832  1084  ?  S   12:15   2:25 ingebin.k HRS
>------------------
>I stopped the LDM and confirmed that all inge... processes were gone and
>restarted it only to find that there were two each again.  The docs suggest
>that this may be due to an error in pqact.conf but I can only find the one
>entry there which is just like the manual says to do
>------------------
># Entries for XCD decoders    (created on 22 Sept 2000)
>DDPLUS|IDS      ^.*     PIPE
>        xcd_run DDS
>HRS     ^.*     PIPE
>        xcd_run HRS
>------------------
>so I don't know how to edit to prevent this doubling from happening.  These
>are the only places where xcd_run appears in the file.

Your troubleshooting was correct up to the point of enabling the GRID decoding
on the McIDAS side:

<login as 'mcidas'>
cd workdata
decinfo.k LIST

When I just logged on, GRID decoding was inactive.  I turned it back on with

decinfo.k SET DMGRID ACTIVE

It must have gotten turned off when I was redoing the XCD setup while we
were on the phone.

>I am also curious about LOCAL.NAM. First there is:
>-----------------
>" Realtime GRID files (the location should be in the directory
>"   in which the ldm-mcidas decoders are producing data files)
>"
>GRID000*     /var/data/mcidas
>GRID010*     /var/data/mcidas
>----------------
>Does ldm-mcidas still have any grid files anymore?

The Unidata-Wisconsin datastream no longer has any products other than
images.  The McIDAS distribution is now over a year old, so the comments
in LOCAL.NAM are now dated.

>And then then there is:
>--------------------------
>" Additional GRID files: ETA, MRF, NGM, MAPS (RUC), ECMWF created by McIDAS
>XCD
>"
>GRID5*      /var/data/xcd
>--------------------------
>which seems right.

This is correct.  The problem was that DMGRID was inactivate.  We send out
XCD with GRID decoding turned off because a lot of sites don't have enough
disk space to turn on the firehose of GRID decoding.

>Thanks
>Erick Lorenz, UCDavis

Tom