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

[IDD #QBL-783661]: split LDM REQUEST for CONDUIT feed

Hi Rick,

> Thanks.

No worries.

> I did make the change you suggested on npserve2.

OK, I thought as much.

> I think we are still dropping grids.

I just did a visual comparison between the CONDUIT volume on the
real-server backend machine that is feeding npserv2 and npserv2, 
and the volumes look the same after you restarted your LDM.  If
you are, in fact, still dropping grids, it either means that they
were not available from our relay, or they were not processed
correctly there.

> For the most part, npserve2 now has more grids than npserve1, but there
> cases where npserve2 has less.

A visual comparison of CONDUIT volumes reported by npserv1 and npserv2
for the time period after you changed to using a 5-way split REQUEST
on npserv2 shows that npserv2 is consistently receiving more volume
than npserv1.  I do not see any times where of the reverse being true.

> Unfortunately, pollux (my standard) has been down for a couple of days.

Just so you know, pollux was showing the same problem of not receiving all of
the CONDUIT volume due to excessive latencies when I looked while at
the MUG meeting.

> On 9/19/2019 2:38 PM, Unidata IDD Support wrote:
> > Hi Rick,
> >
> > re:
> >> Can you send me the link you were using?
> > One gets to the real-time statistics plots for sites reporting LDM/IDD
> > statistics back to us as follows:
> >
> > Unidata HomePage
> > http://www.unidata.ucar.edu
> >
> >    Data -> IDD Operational Status
> >    http://rtstats.unidata.ucar.edu/rtstats/
> >
> >      Real-Time IDD Statistics -> Statistics by Host
> >      http://rtstats.unidata.ucar.edu/cgi-bin/rtstats/siteindex
> >
> > Scroll down in the last page to find the machine(s) you are interested in.
> >
> > By the way, I see that the latencies for CONDUIT being reported by npserv2
> > had a dramatic drop:
> >
> > http://rtstats.unidata.ucar.edu/cgi-bin/rtstats/iddstats_nc?CONDUIT+npserve2.ssec.wisc.edu
> >
> > Compare this to a randomly chosen real-server backend for the 
> > idd.unidata.ucar.edu
> > cluster:
> >
> > http://rtstats.unidata.ucar.edu/cgi-bin/rtstats/iddstats_nc?CONDUIT+node5.unidata.ucar.edu
> >
> > I am assuming that the drop in latencies being reported by npserv2 are a 
> > result
> > of you making the 5-way REQUEST split active.
> >
> >
> > Cheers,
> >
> > Tom


Unidata User Support                                    UCAR Unidata Program
(303) 497-8642                                                 P.O. Box 3000
address@hidden                                   Boulder, CO 80307
Unidata HomePage                       http://www.unidata.ucar.edu

Ticket Details
Ticket ID: QBL-783661
Department: Support IDD
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.