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

[LDM #LTV-934218]: [ldm-users] Pqact running behind

Dale,

[We've driven your inquiry into our support tracking system.]

> We are having an unusual problem with LDM.  We are running the AWIPS-2
> version of LDM that is redistributed by Raytheon and the NWS (the versions
> are based on ldm-6.8.1).
> 
> We have an upstream LDM box that feeds two downstream LDMs with similar
> pqact.conf and ldmd.conf files.  One box simply writes files out for an
> archive, while the other box writes data in the same structure, but also
> is connected to AWIPS-2 (using edexBridge) sending messages for  ingest
> and processing.

Are the data-products being written using the FILE action of pqact(1)? If so, 
is the "-close" or "-flush" option specified?

> When running a pqutil command (e.g., pqutil -w -f NIMAGE
> ldm.pq) on both downstream boxes we verify that each product queue is
> updated very nearly simultaneously.  However, we see one box write out
> its data instantaneously while the other may be delayed by 20 or more
> minutes.

How do you know the delay is occurring? Are the pqact(1)s running in verbose 
logging mode?

> Yesterday the ldm that fed AWIPS-2 was behind, and today the
> archive box is behind.  We made a hardware configuration change on the
> AWIPS-2 box today but no change whatsoever was made to the archive box.

What kind of file-system are the pqact(1) processes writing the data-products 
to? A local disk? An NFS-mounted disk? A Network Addressable Storage (NAS) 
device?

> Does anyone have any ideas what could cause this behavior?
> 
> Thanks,
> 
> Dale Morris
> CIMMS/OU
> NWS/Warning Decision Training Branch
> _______________________________________________
> ldm-users mailing list
> address@hidden
> For list information or to unsubscribe,  visit: 
> http://www.unidata.ucar.edu/mailing_lists/

Regards,
Steve Emmerson

Ticket Details
===================
Ticket ID: LTV-934218
Department: Support LDM
Priority: Normal
Status: Closed