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

Re: 20040205:WMO Feed - used to be iita



Celia,

Indeed...and if all goes to heck in RAP's room. odds are COMET is not
sitting pretty either ;)

Looks solid, I suspect no problems..

Are you requesting CONDUIT from both machines?

I am seeing some bad time (latency) on the CONDUIT feed (chisel) and am
wondering if you are hammerring your bandwidth with CONDUIT?

You may also want to consider not pushing CONDUIT to level even from
chisel, but perhaps co-mouonting a data disk..CONDUIT will slow you down
if there are multiple instances of it....just an idea.

Cheers,

Jeff
---------------------------------------------------------------------
Jeff Weber                                    address@hidden        :
Unidata Program Center                        PH:303-497-8676        :
University Corp for Atmospheric Research      3300 Mitchell Ln       :
http://www.unidata.ucar.edu/staff/jweber      Boulder,Co 80307-3000  :
---------------------------------------------------------------------

On Thu, 5 Feb 2004, Celia Chen wrote:

> Jeff,
>
> I think Tim will be well covered with level and chisel feeds.
> Here is chisel's feed:
> request WMO     ".*"    thelma.ucar.edu PRIMARY
> request WMO     ".*"    atm.geo.nsf.gov ALTERNATE
>
> and level's:
> request WMO     ".*"    chisel.rap.ucar.edu PRIMARY
> request WMO     ".*"    thelma.ucar.edu ALTERNATE
>
> Celia
>
>
> On Thu, Feb 05, 2004 at 05:07:08PM -0700, Jeff Weber wrote:
> > Hi Celia,
> >
> > If it feeds too many then yes....
> >
> > The question is how many is too many.
> >
> > We feed about 140 sites (feeds) from thelma, anything above that begins to
> > introduce latency.
> >
> > I do not know the specs on chisel so cannot deem what is an overload.
> >
> > Does level get fed from chisel, or does level have its own feed..obviously
> > if level feeds from chisel it would not provide a back-up.
> >
> > We can monitor chisel from our stats:
> >
> > http://my.unidata.ucar.edu/content/software/idd/rtstats/siteindex.php?chisel.rap.ucar.edu
> >
> > If it gets bogged down we can look at other alternatives.
> >
> > Looks as if level gets some from chisel as well as some from thelma...so
> > it may be an ok alternate.
> >
> > Thoughts...
> >
> > Jeff
> > ---------------------------------------------------------------------
> > Jeff Weber                                    address@hidden        :
> > Unidata Program Center                        PH:303-497-8676        :
> > University Corp for Atmospheric Research      3300 Mitchell Ln       :
> > http://www.unidata.ucar.edu/staff/jweber      Boulder,Co 80307-3000  :
> > ---------------------------------------------------------------------
> >
> > On Thu, 5 Feb 2004, Celia Chen wrote:
> >
> > > Hi Jeff,
> > >
> > > In your experience, will chisel be over-loaded if it feeds too many 
> > > machines?
> > > I have noticed chisel being slow sometimes.  It is feeding all RAP's
> > > ldm hosts and a few hosts outside of RAP, which is quite a lot.
> > >
> > > chisel and level are set up to feed Tim's machine now. Perhaps it is 
> > > better for
> > > him to get an alternate feed from someone else in UCAR - just in case
> > > RAP's machine room has a problem...
> > >
> > >
> > > Celia
> > >
> > > On Thu, Feb 05, 2004 at 03:38:01PM -0700, Jeff Weber wrote:
> > > > Hi Tim,
> > > >
> > > > For now let's make:
> > > >
> > > > chisel.rap.ucar.edu
> > > >
> > > > your primary..and for the alternate...well if chisel is not getting it,
> > > > probably nobody is at RAP. I believe RAP has another machine with a feed
> > > > that we could use as an alternate...
> > > >
> > > > Celia,
> > > >
> > > > Do you have any other "feed" machine at rap that Tim could use as an
> > > > alternate?...that does not feed from chisel.
> > > >
> > > >
> > > > If not I will find somebody else in UCAR to use as an alterante.
> > > >
> > > > Cheers,
> > > >
> > > >
> > > > Jeff
> > > > ---------------------------------------------------------------------
> > > > Jeff Weber                                    address@hidden        :
> > > > Unidata Program Center                        PH:303-497-8676        :
> > > > University Corp for Atmospheric Research      3300 Mitchell Ln       :
> > > > http://www.unidata.ucar.edu/staff/jweber      Boulder,Co 80307-3000  :
> > > > ---------------------------------------------------------------------
> > > >
> > > > On Thu, 5 Feb 2004, Unidata Support wrote:
> > > >
> > > > >
> > > > > ------- Forwarded Message
> > > > >
> > > > > >To: address@hidden
> > > > > >From: Tim Alberta <address@hidden>
> > > > > >Subject: WMO Feed - used to be iita
> > > > > >Organization: UCAR/COMET
> > > > > >Keywords: 200402052156.i15LuAp1002462
> > > > >
> > > > > Hi everyone,
> > > > >
> > > > > We are using iita for our primary WMO feed, and chisel.rap.ucar.edu 
> > > > > for
> > > > > our alternate.  Since iita's being retired, Celia suggested I contact
> > > > > you to get a different primary.  Any guidance on how to proceed from
> > > > > here?
> > > > >
> > > > > Thanks,
> > > > >
> > > > > Tim
> > > > >
> > > > > Celia Chen wrote:
> > > > > >
> > > > > > Hi Tim,
> > > > > >
> > > > > > iita has not been an UCAR-wide LDM server for sometime now. I think
> > > > > > it is best for you to talk to the Unidata folks to get your
> > > > > > primary feed from them.  Let me know if you need us to be your
> > > > > > alternate feed source.
> > > > > >
> > > > > > Thanks.
> > > > > >
> > > > > > Celia
> > > > > >
> > > > > > On Fri, Jan 23, 2004 at 11:33:27AM -0700, Tim Alberta wrote:
> > > > > > > Thanks for the heads up Celia,
> > > > > > >
> > > > > > > Does RAP have a replacement that COMET can feed from, or should I 
> > > > > > > talk
> > > > > > > to the Unidata folks for guidance?
> > > > > > >
> > > > > > > Thanks,
> > > > > > >
> > > > > > > Tim
> > > > > > >
> > > > > > >
> > > > > > > Celia Chen wrote:
> > > > > > > >
> > > > > > > > RAP's LDM server "iita" will be retired in about two weeks.  
> > > > > > > > Please
> > > > > > > > switch your LDM data feed to another LDM host if you are getting
> > > > > > > > data from iita currently.
> > > > > > > >
> > > > > > > > Thanks.
> > > > > > > >
> > > > > > > > Celia
> > > > >
> > > > > NOTE: All email exchanges with Unidata User Support are recorded in 
> > > > > the
> > > > > Unidata inquiry tracking system and then made publically 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.
> > > > > --
> > > > >
> > > > > ------- End of Forwarded Message
> > > > >
> > >
>