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

Re: 20020207: missing products

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: Re: 20020207: missing products
  • Date: Thu, 07 Feb 2002 14:35:48 -0700
Harry Edmon wrote:
> 
> It was the feed that includes MCIDAS and WMO - it was a McIDAS product that
> timed out.
> 
> Unidata Support <address@hidden> wrote:
> > >From: Harry Edmon <address@hidden>
> > >Organization: UCAR/Unidata
> > >Keywords: 200202071958.g17Jwtx15748
> >
> > >I am having an internal problem sending products between two machines.  For
> > so
> > > me
> > >reason the RPC call on the sending machine is timing out.  This causes the
> > >process to die and a new feed process to start.  However, it appears that 
> > >the
> > >new process does not take over exactly where the old one left off - the
> > produc
> > > t
> > >being sent at the time is not resent - and it appears that a few products
> > just
> > >after that product may be missing.  Any ideas?
> > >
> > >
> > >--
> > >Dr. Harry Edmon                      E-MAIL: address@hidden
> > >206-543-0547                         address@hidden
> > >Dept of Atmospheric Sciences FAX:    206-543-0308
> > >University of Washington, Box 351640, Seattle, WA 98195-1640
> > >
> >
> > Harry,
> >
> > Anne may have some ideas on why the RPC is timing out. You can
> > increase the timeout with the -t flag to rpc.ldmd....but
> > it sounds bad that two local machines are having a timeout.
> >
> > Are the missing products "CONDUIT" products, or another feed?
> > If the missing prods are CONDUIT, then the point of discontinuity
> > might result from the order which the products were received from the split
> > requests. When the LDM reconnects, it would request "from" the timestamp
> > of the latest product received of that feedtype - which may be more
> > recent for one part of the split request than another. If not CONDUIT,
> > then I don't have a quick guess, unless it was fore something like
> > the craft feed where products might be injected at different hosts
> > in the same feed- again creating a time continuity between
> > the same feedtype products with different sequencing.
> >
> > Steve Chiswell
> 
> --
> Dr. Harry Edmon                 E-MAIL: address@hidden
> 206-543-0547                            address@hidden
> Dept of Atmospheric Sciences    FAX:    206-543-0308
> University of Washington, Box 351640, Seattle, WA 98195-1640

Hi Harry,

Offhand, without much information, the RPC time out sounds somewhat like
a resource limitation.  What does netstat show you?  Perhaps you've run
out of connections?  Maybe the OS is configured to keep connections open
for too long of a time.

Is this intermittent?  What's the history? And, what OS are you using? 

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