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

Re: UCP-475565 CONDUIT Feed



Hi Scott,

Will check into it...

What does the command "hostname" return on:

129.15.40.176

and

129.15.40.177

388 should do the trick, portmapper is not required, let me peek into our
logs.

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 Wed, 8 Mar 2006 address@hidden wrote:

> Thanks Jeff.
>
> Unfortunately something else is preventing me from connecting to
> idd.unidata.ucar.edu from either 129.15.40.176 or 129.15.40.177. Here is
> what I see in my LDM logs:
>
> 128.117.140.3[13222] ERROR: Terminating due to LDM failure; nullproc_6
> failure to 128.117.140.3; RPC: Unable to receive; errno = Connection
> reset by peer
>
> ldmping idd.unidata.ucar.edu gives this: ERROR:  ADDRESSED 0.041143 0
> idd.unidata.ucar.edu RPC: Unable to receive; errno = Connection reset by
> peer
>
> I can connect to 388/tcp. I can not connect to port 111/tcp or 111/udp.
>
> Both 129.15.40.176 and 129.15.40.177 are able to receive LevelII data
> from WRHQ just fine.
>
> 129.15.194.232 has no problem connecting to and is receiving data from
> idd.unidata.ucar.edu.
>
> Is it possible that a firewall near your end is blocking some ports for
> these two machines? I've checked all of ours and everything seems fine.
>
> Thanks again, Scott
>
> On Tue, Mar 07, 2006 at 02:25:16PM -0700, Jeff Weber wrote:
> > Hi Scott,
> >
> > I implemented the allow for:
> >
> >      129.15.40.176
> >      129.15.40.177
> >
> > and
> >
> >      129.15.194.232
> >
> > for the CONDUIT feed
> >
> > from:
> >
> > idd.unidata.ucar.edu
> >
> > after our contact on the 27th...
> >
> > It would be best if only ONE was feeding from idd.unidata.ucar.edu at a
> > time...dual feeds to the same machine only help you if hardware goes bad.
> >
> > I will attempt to find you a geographically unique failover feed for
> > CONDUIT..
> >
> >
> > Please begin feeding your machines at your convenience, and please stop
> > requesting from winkie and munchkin once you start from the other
> > machines.
> >
> > 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 Tue, 7 Mar 2006 address@hidden wrote:
> >
> > > Hi Jeff,
> > > Just checking on the status of this.
> > >
> > > Please don't hesitate to let me know if there is anything I need to do on 
> > > this.
> > >
> > > Thanks,
> > > Scott
> > >
> > > ----- Forwarded message from address@hidden -----
> > >
> > > Date: Mon, 27 Feb 2006 13:46:28 -0600
> > > From: address@hidden
> > > To: Jeff Weber <address@hidden>
> > > Subject: [address@hidden: Re: 20060221: CONDUIT feed]
> > > Lines: 109
> > >
> > > Hi Jeff, Just touching base. Does this plan work for you?  If so, would
> > > it be possible to get an estimate on when it will be implemented?
> > > Sorry to have to ask but my boss will ask me.
> > >
> > > Right now 129.15.194.232 has access to CONDUIT data from
> > > idd.unidata.ucar.edu but 129.15.40.176 and 129.15.40.177 do not.
> > >
> > > Thanks,
> > > Scott
> > >
> > >
> > > ----- Forwarded message from address@hidden -----
> > >
> > > Date: Wed, 22 Feb 2006 15:32:08 -0600
> > > From: address@hidden
> > > To: Jeff Weber <address@hidden>
> > > Cc: address@hidden
> > > Subject: Re: 20060221: CONDUIT feed
> > > Lines: 89
> > >
> > > Hi Jeff,
> > >
> > > On Tue, Feb 21, 2006 at 11:16:35AM -0700, Jeff Weber wrote:
> > > > Hi Scott,
> > > >
> > > > "We would like to
> > > > > ingest the full whack of CONDUIT data to a system here in CAPS for
> > > > > research and distribution to Mark and other OU researchers."
> > > >
> > > > ..any chance of feeding non-OU sites in the future?
> > > Yes. Exactly who we will be feeding, other than OU sites, I can't say for 
> > > sure at this moment. Kelvin Droegemeier and Joel Martin will be the 
> > > people that make this determination.
> > >
> > > > "our
> > > > > two machines winkie.caps.ou.edu and munchkin.caps.ou.edu could be
> > > > > migrated to downstream clients of this new system."
> > > >
> > > > Yes, I doubt you want 3 instances of CONDUIT coming into your domain.
> > > >
> > > > Please give us the FQDN and IP# of your new LDM ingest machine and we 
> > > > can
> > > > allow it access to:
> > > >
> > > > idd.unidata.ucar.edu
> > > There will be two primary machines in a fail-over config so that only one 
> > > machine will be ingesting data at any given time:
> > >     129.15.40.176
> > >     129.15.40.177
> > >
> > > Also, one secondary machine that will provide a transition for some of 
> > > our other systems.
> > >     129.15.194.232
> > >
> > > So in total only one or two out of these three machines will be 
> > > requesting CONDUIT data at any given time.
> > >
> > > We are very interested in fail-over and redundancy. Could you recommend
> > > a second site for CONDUIT data, maybe something geographically distant
> > > from you?
> > >
> > > >
> > > > Please do not continue to feed CONDUIT to winkie or munchkin from
> > > > idd.unidata.ucar.edu once you start up the new ingest machine, but as 
> > > > you
> > > > mentioned, feed winkie and munchkin from your new LDM ingest machine.
> > > No problem.
> > >
> > > >
> > > > We would want to do this for all feeds going to winkie and munchkin from
> > > > idd.unidata.ucar.edu...looks primarily to be CONDUIT.
> > > I think this is possible but it may need to wait until we complete our 
> > > move to the new building in July. I'm the sysadmin for these machines but 
> > > the person controller data ingest on winkie and munchkin is our RT 
> > > Forecast guy. I'll discus this with him and come up with a plan to get 
> > > this done.
> > >
> > > >
> > > > Once your new machine is up and running we will remove the "allow's" for
> > > > winkie and munchkin.
> > > Makes sense. Could we coordinate this to eliminate any chance of holes in 
> > > our data stream?
> > >
> > > Thanks Jeff,
> > >
> > > Scott
> > >
> > > >
> > > > 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 Tue, 21 Feb 2006 address@hidden wrote:
> > > >
> > > > > Hi Jeff, I was told by Mark Laufersweiler (OU Meteorology/SOM) that 
> > > > > you
> > > > > were the person to contact for a IDD CONDUIT data feed. We would like 
> > > > > to
> > > > > ingest the full whack of CONDUIT data to a system here in CAPS for
> > > > > research and distribution to Mark and other OU researchers. We have a
> > > > > small cluster of machines dedicated to moving LevelII, LevelIII, and
> > > > > other data via LDM so horse power on our end should not be a problem.
> > > > > Mark and others here at OU have expressed an interest in using our
> > > > > system's resources to reduce their overhead in data ingestion. Also,
> > > > > although we're not really using CONDUIT on these machines right now, 
> > > > > our
> > > > > two machines winkie.caps.ou.edu and munchkin.caps.ou.edu could be
> > > > > migrated to downstream clients of this new system.
> > > > >
> > > > > Is this possible, and if so, what do you need from me?
> > > > >
> > > > > Thanks much for any help,
> > > > > Scott Hill
> > > > > Systems Engineer
> > > > > CAPS/RCCC, University of Oklahoma
> > > > > Office: 405-325-7268
> > > > > Mobile: 405-630-0910
> > > > > Email:  address@hidden
> > > > >
> > > > >
> > >
> > > ----- End forwarded message -----
> > >
> > > ----- End forwarded message -----
> > >
>