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

[LDM #EIE-998144]: LDM 6.12.6: is this expected behavior?

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: [LDM #EIE-998144]: LDM 6.12.6: is this expected behavior?
  • Date: Fri, 19 Sep 2014 11:08:39 -0600
Gilbert,

No. That's not the expected behavior at all. The attempt by the ldmadmin(1) 
script to create a new queue of larger size should have failed and resulted in 
the error message "vetQueueSize(): Couldn't create new queue: $newQueuePath" 
(the pathname of the new queue would be the same as the old queue with a ".new" 
suffix).

Did you get that error-message?

After that failure, you should periodically get the same error-message.

> Hello Steve,
> 
> A troublemaker opening trouble tickets. Did you expect anything less?
> Anyway, happy Friday. I have a question for you...
> 
> Recently, the College of DuPage upgraded their LDM to 6.12.6, and switched
> to the new NOAAport feed on their satellite ingest machine. On that
> ingest machine, they put their ldm.pq in /run/shm, or shared memory.
> 
> Lasst week, ldm.pq reached 6.5 GB...which was how big their shared
> memory was. The LDM was instructed to increase queue size to maintain an
> hours worth of data, and it finally hit the machine memory size limit last
> week.
> 
> So this happened: suddenly, only small files started coming through when
> /dev/shm hit 100%, and only IDS|DDPLUS stuff.
> 
> To its credit, the LDM didn't crash. But, it did pretty much stop the
> flow of data, minus a few small IDS|DDPLUS products. Once the ldm.pq file
> was deleted and remade to hard disk, all was well.
> 
> So, just a curiosity question: expected behavior? I would more or less
> think so, but I'm just curious.
> 
> Gilbert
> 
> *******************************************************************************
> Gilbert Sebenste                                                    ********
> (My opinions only!)                                                  ******
> Staff Meteorologist, Northern Illinois University                      ****
> E-mail: address@hidden                                  ***
> web: http://weather.admin.niu.edu                                      **
> Twitter: http://www.twitter.com/NIU_Weather                            **
> Facebook: http://www.facebook.com/niu.weather                           *
> *******************************************************************************


Regards,
Steve Emmerson

Ticket Details
===================
Ticket ID: EIE-998144
Department: Support LDM
Priority: Normal
Status: Closed