[Support #VOC-356419]: latency

Michael,

> Thanks much for the analysis and advice.  I appreciate it greatly.  One more 
> question if you don't mind.  I ran netstat this morning and found several of 
> our field offices with upwards of 20 processes.  Our western region office 
> alone has 83 processes active or waiting.  I wrote to my contact there and 
> described the problem.

The problem is that you think there are too many processes?

> He told me that in their ldmd.conf file, they make three requests for each of 
> their 24 field offices accounting for 72 requests, and then they have a few 
> individual requests totaling 83.  If I remember our training correctly, each 
> of those requests results in a daemon on my LDM server.

Each REQUEST entry in an LDM configuration-file results in one downstream LDM 
on the local host and one upstream LDM on the remote host.

> Western Region's LDM person seems to think that more requests like ".*" 
> outperforms those like "EXP|FSL2|FSL3".  He is out until Monday at which time 
> I plan to place a call.  Do I understand this correctly?

I'm not sure I understand this last part.  ".*" is a pattern for matching 
against a product's identifier and "EWXP|FSL2|FSL3" is a feedtype expression 
for matching against a product's feedtype.  They're apples and oranges.

> Thanks again,
> Michael


Regards,
Steve Emmerson

Ticket Details
===================
Ticket ID: VOC-356419
Department: Support LDM
Priority: Normal
Status: Closed