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

Re: LDM on cirp

"Bryan G. White" wrote:
> 
> Anne,
> 
> Still nothing but nexrad for the past ~30 minutes from thelma.  I added in a 
> backup site and I'm getting other feeds now.
> 
> I stopped and restarted ldm twice.  Remade the queue.
> 
> Still nothing abnormal in the logs.
> 
> Bryan
> 
> > Anne,
> >
> > The traceroute is normal.  It's being blocked at the router.  I can 
> > traceroute to thelma just fine.
> >
> > There were a gazillion hung jobs related to a MesoWest process gone amok.  
> > I've killed those off.  Looks like cirp was approaching the limit of the 
> > number of jobs it could run.  Maybe that was causing problems.
> >
> > I also just stop and restarted ldm.
> >
> > There is nothing abnormal in my logs.
> >
> > I am getting nexrad data from thelma, but I haven't seen any other data 
> > feeds from thelma come on by the past 5 minutes.
> >
> > Hmmmmm.....
> >
> >
> > Bryan
> >

Hi Bryan,

Thank you for looking into this.  I'm glad the traceroute is good from
your end and explains the bad traceroute from here.  It looks like your
restart was about 0Z, but the disconnect problem persists.  From the
logs:

/local/ldm%  grep cirp ldmd.log | grep Connect 
Apr 24 23:37:23 milton.unidata.ucar.edu cirp[6656]: Connection from
cirp.met.utah.edu
Apr 24 23:38:19 milton.unidata.ucar.edu cirp[6657]: Connection from
cirp.met.utah.edu
Apr 24 23:43:46 milton.unidata.ucar.edu cirp[6682]: Connection from
cirp.met.utah.edu
Apr 24 23:45:36 milton.unidata.ucar.edu cirp[6684]: Connection from
cirp.met.utah.edu
Apr 24 23:48:04 milton.unidata.ucar.edu cirp[6693]: Connection from
cirp.met.utah.edu
Apr 24 23:52:10 milton.unidata.ucar.edu cirp[6698]: Connection from
cirp.met.utah.edu
Apr 24 23:57:59 milton.unidata.ucar.edu cirp[6702]: Connection from
cirp.met.utah.edu
Apr 25 00:01:08 milton.unidata.ucar.edu cirp[6721]: Connection from
cirp.met.utah.edu
Apr 25 00:02:23 milton.unidata.ucar.edu cirp[6723]: Connection from
cirp.met.utah.edu
Apr 25 00:02:29 milton.unidata.ucar.edu cirp[6723]: Connection reset by
peer
Apr 25 00:02:38 milton.unidata.ucar.edu cirp[6724]: Connection from
cirp.met.utah.edu
Apr 25 00:17:11 milton.unidata.ucar.edu cirp[6732]: Connection from
cirp.met.utah.edu
Apr 25 00:19:35 milton.unidata.ucar.edu cirp[6733]: Connection from
cirp.met.utah.edu
Apr 25 00:20:06 milton.unidata.ucar.edu cirp[6735]: Connection from
cirp.met.utah.edu
Apr 25 00:20:37 milton.unidata.ucar.edu cirp[6736]: Connection from
cirp.met.utah.edu
Apr 25 00:21:09 milton.unidata.ucar.edu cirp[6738]: Connection from
cirp.met.utah.edu
Apr 25 00:21:42 milton.unidata.ucar.edu cirp[6739]: Connection from
cirp.met.utah.edu
Apr 25 00:22:42 milton.unidata.ucar.edu cirp[6740]: Connection from
cirp.met.utah.edu
Apr 25 00:23:13 milton.unidata.ucar.edu cirp[6741]: Connection from
cirp.met.utah.edu
Apr 25 00:23:45 milton.unidata.ucar.edu cirp[6742]: Connection from
cirp.met.utah.edu
Apr 25 00:24:25 milton.unidata.ucar.edu cirp[6743]: Connection from
cirp.met.utah.edu
Apr 25 00:25:09 milton.unidata.ucar.edu cirp[6745]: Connection from
cirp.met.utah.edu
Apr 25 00:25:42 milton.unidata.ucar.edu cirp[6746]: Connection from
cirp.met.utah.edu
Apr 25 00:26:50 milton.unidata.ucar.edu cirp[6747]: Connection from
cirp.met.utah.edu
Apr 25 00:36:27 milton.unidata.ucar.edu cirp[6752]: Connection from
cirp.met.utah.edu
Apr 25 00:37:27 milton.unidata.ucar.edu cirp[6753]: Connection from
cirp.met.utah.edu
Apr 25 01:15:20 milton.unidata.ucar.edu cirp[6781]: Connection from
cirp.met.utah.edu
Apr 25 01:16:41 milton.unidata.ucar.edu cirp[6783]: Connection from
cirp.met.utah.edu
Apr 25 01:19:09 milton.unidata.ucar.edu cirp[6784]: Connection from
cirp.met.utah.edu
Apr 25 01:23:22 milton.unidata.ucar.edu cirp[6787]: Connection from
cirp.met.utah.edu
Apr 25 01:28:21 milton.unidata.ucar.edu cirp[6790]: Connection from
cirp.met.utah.edu
Apr 25 01:32:44 milton.unidata.ucar.edu cirp[6793]: Connection from
cirp.met.utah.edu
Apr 25 01:39:40 milton.unidata.ucar.edu cirp[6795]: Connection from
cirp.met.utah.edu
Apr 25 01:44:00 milton.unidata.ucar.edu cirp[6798]: Connection from
cirp.met.utah.edu
Apr 25 01:46:13 milton.unidata.ucar.edu cirp[6800]: Connection from
cirp.met.utah.edu
Apr 25 01:48:46 milton.unidata.ucar.edu cirp[6802]: Connection from
cirp.met.utah.edu
Apr 25 01:52:11 milton.unidata.ucar.edu cirp[6804]: Connection from
cirp.met.utah.edu

One thing you really should do is split up the NMC2 feed.  That feed now
reaches 1.5 gigabytes in some hours.  Without very small RPC turnaround
times (like you would find on a local network) you can't get that much
data across a single connection in one hour.  Once you get behind it may
be impossible to catch up.  We have found that splitting the feed into 5
different feeds produces satisfactory results.  Here's a relevant
message from our archives: 
http://www.unidata.ucar.edu/cgi-bin/mfs/65/4924?14#mfs.  Of course,
you'll have to substitute thelma.ucar.edu for tgsv32.nws.noaa.gov.

Please give this a try and let's see if that improves the situation.

Anne
-- 
***************************************************
Anne Wilson                     UCAR Unidata Program            
address@hidden                  P.O. Box 3000
                                  Boulder, CO  80307
----------------------------------------------------
Unidata WWW server       http://www.unidata.ucar.edu/
****************************************************