[LDM #SVU-966060]: LDM - LDM hangs up.

Scott,

> Institution: NOAA/NWS
> Package Version: 6.3.0
> Operating System: Redhat WS 4
> Hardware Information: Intel based server
> Inquiry: I recently upgraded the OS on my server from Redhat WS 3 to Redhat 
> WS 4.  Since doing so, servers randomly stop being able to retrieve data from 
> my LDM server and are unable to ldmping my server after roughly five days.  
> Attempting to perform an ldmadmin restart or ldmadmin stop results in the 
> message Waiting for LDM to terminate over and over while never actually 
> stopping the ldmd service.  Performing ps -ef | grep ldmd results in between 
> 20 and 100 "rpc.ldmd -v -q /usr/local/ldm/data/ldm.pq 
> /usr/local/ldm/etc/ldmd.conf" entries (just tested, and I have 93 running 
> atm).  Ultimately, the only fix I've found is to kill all ldmd processes, run 
> ldmadmin clean, the pqcat xxx command to clean up the queue, and restart the 
> service.  Unfortunately, this normally requires all my downstream users to 
> restart their LDM servers.  Is this a known problem with WS4?  What am I 
> doing incorrectly, and why are their so many spawned processes? When I ran 
> this on WS3, I normally only!
  h!
> ad 20-40 processes running.  What changed?

93 connections to downstream LDM-s is a lot.  I take it that the hardware in 
question is pretty good.

Changing the operating-system *shouldn't* affect the LDM, but apparently it has 
in your case.  Could you upgrade your LDM to the latest version that was just 
released (v 6.4.6)?  There have been more that a couple of bug-fixes since 
version 6.3.0 and it's possible that one of them has fixed your problem.

Regards,
Steve Emmerson

Ticket Details
===================
Ticket ID: SVU-966060
Department: Support LDM
Priority: Normal
Status: On Hold