>
> what "protocol" should
> we use to determine queue size?
`ldmadmin vetqueuesize` works well for us.
To quote the great Steve Emmerson:
> The objective is to have a queue with sufficient capacity (both in the
> number of bytes and the number of products that it can hold) so that the
> minimum residence time of a product in the queue is greater than the
> "max_latency" parameter in the registry (default: 3600 s) so that duplicate
> product detection will work.
>
On Fri, Oct 1, 2021 at 3:10 PM Patrick L. Francis <wxprofessor@xxxxxxxxx>
wrote:
>
> well now there's 3 good points from karen, gilbert, and ryan...
>
> if we are all 'right' .. and time is a factor, user dependent, then by
> definition our queue sizes should have been changing over the years..
> mine has, but i'm just kind of anal like that... what "protocol" should
> we use to determine queue size?
>
> somebody a long time ago, i have absolutely no idea threw up the idea of
> an hour and it stuck with me... so on my noaaport relays, and I ingest
> everything there is on those, by definition should contain a queue size
> equivalent to roughly one hours of data at average or peak times.. now
> what that was 10 years ago, is not what it is now, nor what it will be
> next year :o :P
>
> good discussion!!! woohoo!!
>
> cheers,
>
> --patrick
>
>
> _______________________________________________
> NOTE: All exchanges posted to Unidata maintained email lists are
> recorded in the Unidata inquiry tracking system and made publicly
> available through the web. Users who post to any of the lists we
> maintain are reminded to remove any personal information that they
> do not want to be made public.
>
>
> ldm-users mailing list
> ldm-users@xxxxxxxxxxxxxxxx
> For list information or to unsubscribe, visit:
> https://www.unidata.ucar.edu/mailing_lists/
>
