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

20040909: Possible pqact issue in LDM?

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: 20040909: Possible pqact issue in LDM?
  • Date: Thu, 09 Sep 2004 13:56:00 -0600
Steven,

>Date: Thu, 09 Sep 2004 13:02:24 -0500
>From: "Steven Danz" <address@hidden>
>Organization: Aviation Weather Center
>To: address@hidden
>Subject: Possible pqact issue in LDM?
>Keywords: 200409091803.i89I3pnJ023109

The above message contained the following:

>      Here's the scoop.  We have a 6.0.14 install running on a NOAAPort
> receiver that is running RH 7.2. The software on the NRS brings in the
> data an inserts it into the queue via calls to ldm.a.  The pqact.conf
> on the system has an entry that looks like:
> WMO ^([A-Z].*) (KKCI) (.*)
>      EXEC    insert_message OTHER "received \1 \2 \3"
> 
> which is attempting to catch any product issued by the AWC and run the
> listed script with its header.  The script uses pqinsert to insert a
> note into the queue with the time that the product was seen.
> 
> The issue we are seeing is that at times when we get a 'chunk' of KKCI
> products (maybe 90 within a few seconds) not all of them cause the EXEC
> to happen.  Sometimes 1 or 2 are missed, sometimes a dozen, sometimes 0.
> The script creates a log entry at the start so we can see if it is being
> called and it doesn't appear to be.  I'm trying to duplicate the situation,
> however all I have right now is a script that uses pqinsert, which is
> probably not as 'fast' as the program that inserts the NOAAPort datastream
> directly.  So far, no luck.  I've attached the script for reference.
> 
> One thing we have noticed is that on the backup server, which has no
> downstream clients, it appears that it doesn't happen or doesn't happen
> as often.

First things first.  Try adding the "-wait" option to the EXEC entry 
(without the quotes and bracketed by tabs.

Along those lines, what is the limit on the number of child process a 
user may have? (Try the command "ulimit -a")

Is there anything relevant in the LDM logfile from the pqact(1) process?

> Thanks
> 
> -- 
> Steven Danz
> Senior Software Development Engineer
> Aviation Weather Center (NOAA/NWS/NCEP)
> 7220 NW 101st Terrace, Room 101
> Kansas City, MO 64153-2371
> 
> Email: address@hidden
> Phone: 816.584.7251
> Fax:   816.880.0650
> URL:   http://aviationweather.gov/
> 
> The opinions expressed in this message do not necessarily reflect those
> of the National Weather Service, or the Aviation Weather Center.

Regards,
Steve Emmerson