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

Re: 20010421: Data feed to UCR

Jimmy Mejia Fernández wrote:
> 
> >
> > Hi Jimmy,
> >
> 
> Hello Anne, Tom:
> 
> ...(...)
> 
> >
> > First, let's not worry about the output of the notifyme command above -
> > that may be a non-issue.  Sometimes there are problems when you do a
> > 'notifyme' to your own host.  In the past you may have used notifyme to
> > see what your upstream host was getting.  I expect you should be able to
> > do that now.  You might try this:
> >     notifyme -vl - -h thelma.unidata.ucar.edu
> > If you start seeing products listed, then you know the LDM on thelma is
> > up and receiving data.  You also know that you are 'allow'ed to connect
> > to thelma for a feed.
> 
> Yes, I know I am allowed to connect to thelma from inti and from titan (a
> machine that finally I coud not make to work, I need to replace the SCSI
> controler)
> 
> >
> > If you want to see information about products as they arrive at your
> > site, do 'ldmadmin watch'.  See the 'ldmadmin' man page for more info on
> > this.  (For example, you can give it a feed type argument.)   'ldmadmin
> > watch' simply calls the 'pqutil' software.  If you want to watch for
> > specific products as they arrive at your site, use pqutil with the
> > pattern argument.  See the pqutil man page for more info.
> >
> 
> Ujum, this 2 days I has been reading the manuals pages they have more info
> than the "unidata LDM5 site managers guide" :)
> 
> > I understand that you are not getting MCIDAS and grib files that you
> > want.  Let's establish whether you're getting the files and not
> > processing them, or whether you're not getting them at all.   You can do
> > this by scanning the produts currently in the queue with pqcat.  Try:
> >     pqcat -vl - -f MCIDAS > /dev/null
> 
> I did it I got a beautiful output, may be is because since 2 days ago we
> changed to a bigger bandwith, a T1
> here is an extraction of the output:
> [ldm@inti data]$  pqcat -vl - -f MCIDAS > /dev/null
> Apr 28 02:10:50 pqcat: Starting Up (13868)
> Apr 28 02:10:50 pqcat:    62080 20010427134912.146  MCIDAS 000  pnga2area
> Q0 CE 1140 GOES-10_SND UNKBAND 14km 20010427 1200
> Apr 28 02:10:50 pqcat:     3756 20010427140529.360  MCIDAS 000  pnga2area
> Q1 U3 201 GRAPHICS UNKBAND 5km 20010427 1359
> Apr 28 02:10:50 pqcat:   144826 20010427161627.260  MCIDAS 000  pnga2area
> Q1 UB 171 GOES-10_IMG 6.8um 8km 20010427 1600
> Apr 28 02:10:50 pqcat:   303931 20010427173151.463  MCIDAS 000  pnga2area
> Q1 UA 161 GOES-8_IMG 0.65um 1km 20010427 1715
> Apr 28 02:10:50 pqcat:    71415 20010427174808.087  MCIDAS 000  pnga2area
> Q0 CD 1130 GOES-10_SND UNKBAND 14km 20010427 1600
> Apr 28 02:10:50 pqcat:   164050 20010427181624.962  MCIDAS 000  pnga2area
> Q1 UB 172 GOES-10_IMG 6.8um 8km 20010427 1800
> ........
> 
> > This will give you a bit of information about each MCIDAS product in your
> > queue.  (It will also dump the actual product to /dev/null.)  If this
> > returns without reporting anything, then there are none in your queue.
> >
> > You can also give pqcat a product pattern argument for it to be more
> > selective.  To do a similar test for the grib products try something
> > like:
> >     pqcat -vl - -f HRS -p "^[YZ].[QRUT].*/mETA" > /dev/null
> > (Substitute a pattern appropriate for you inside the double  quotes.)
> >
> > Please try these last two pqcat commands and let me know what happens.
> > Also, please resend a copy of your pqact.conf - the attachment got
> > munged.
> 
> Again here is a short output of the above command:
> 
> [ldm@inti data]$ pqcat -vl - -f HRS -p "^[YZ].[QRUT].*/mETA" > /dev/null
> Apr 28 02:12:08 pqcat: Starting Up (13869)
> Apr 28 02:12:08 pqcat:      764 20010427140147.169     HDS 778  YOTA10
> KWBE 271200 /mETA_84
> Apr 28 02:12:08 pqcat:      978 20010427140147.189     HDS 779  YOTA15
> KWBE 271200 /mETA_84
> Apr 28 02:12:08 pqcat:      978 20010427140147.568     HDS 816  YOTA20
> KWBE 271200 /mETA_84
> Apr 28 02:12:08 pqcat:     1192 20010427140147.571     HDS 817  YOTA30
> KWBE 271200 /mETA_84
> Apr 28 02:12:08 pqcat:     1192 20010427140147.575     HDS 818  YOTA25
> KWBE 271200 /mETA_84
> Apr 28 02:12:08 pqcat:     1192 20010427140147.580     HDS 819  YOTA35
> KWBE 271200 /mETA_84
> Apr 28 02:12:08 pqcat:     1406 20010427140147.583     HDS 820  YOTA40
> KWBE 271200 /mETA_84
> Apr 28 02:12:08 pqcat:     1406 20010427140147.587     HDS 821  YOTA80
> KWBE 271200 /mETA_84
> Apr 28 02:12:08 pqcat:     1406 20010427140147.592     HDS 822  YOTA45
> KWBE 271200 /mETA_84
> 
> >
> > Anne
> 
> I also  followed the changes that Mr. Yoksas pointed in his last letter,
> posible that also helped (Spliting the requests), I see that the
> /home/ldm/data/mcidas/area directory is being filled with area files, the
> same is hapening with the /home/ldm/data/GRIB directory with the *.avn
> files.
> 
> A single question, how much hard disk will I need for this two products?
> I commented out the lines:
> 
> WMO
> ^([^H][A-Z])([A-Z][A-Z])([0-9][0-9]) (....) ([0-3][0-9])([0-2][0-9])
>        FILE    data/\2/\4/\6/\1\3.wmo
> 
> form the pqact.conf file, that is because the *.wmo files are filling the
> system, so, for now I do not want this files. Next week I will install
> another hard drive in this machine, how big do you recommend for this?
> 
> Thank you very much for all your help.
> 
> JImmy

Hi Jimmy,

I'm glad to hear things are working for you.

Tom probably knows the sizes of these things better than I do.  But, I
can tell you that on one of our machines in the past 24 hours the MCIDAS
feed amounted to 112Mbytes, HDS was 1641Mbytes, and IDS *and* DDPLUS was
121Mbytes.  (As I recall you're not requesting IDS so yours would be a
little smaller.)

How much space you need depends on how long you want to keep the data. 
Do you know about 'scour'?  That's the LDM utility, run via cron, that
disposes of old files.  See the man page for more information.  And, let
us know if you have any further questions.

Anne 

-- 
***************************************************
Anne Wilson                     UCAR Unidata Program            
address@hidden                  P.O. Box 3000
                                  Boulder, CO  80307
----------------------------------------------------
Unidata WWW server       http://www.unidata.ucar.edu/
****************************************************