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

[LDM #DBK-582040]: 6.10.1 how to diagnose pqact memory leak?



Daryl,

> Thanks for the response, if you could induldge me with one more round of
> questions, please see inline....
> 
> So I am now confused about the need to split pqact processes up for high
> volume feeds, like NEXRAD2 .  I thought the justification was that if you
> have 80 some RADARs producing data and thus writing to files, pqact would
> have to be constantly cycling file descriptors so to only have 32 open at
> once? I thought the split was done to limit the number of RADARs per
> pqact to something less than 32, so that FDs are kept open and recycled
> more slowly...

I don't recall ever hearing that rationale.

> Or is the split done so that there could be multiple write processes
> ongoing at one time?

Reasons for splitting in decreasing order of importance (in my opinion):

    1. To make things easier to understand and maintain (all radar processing 
in one pqact(1) file, etc.);
    2. To take advantage of multiple CPU-s;
    3. To reduce file-descriptor reuse.

> Thanks, I have been attempting to understand the "twisted
> logic" of close_lru() and your statement above helps in that! :)

Glad I could help.

> thanks again,
> daryl
> 
> --
> /**
> * Daryl Herzmann
> * Assistant Scientist -- Iowa Environmental Mesonet
> * http://mesonet.agron.iastate.edu

Regards,
Steve Emmerson

Ticket Details
===================
Ticket ID: DBK-582040
Department: Support LDM
Priority: Normal
Status: Closed


NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.