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

Re: 19990801: No McIDAS AREA files (cont.) (fwd)


===============================================================================
Robb Kambic                                Unidata Program Center
Software Engineer III                      Univ. Corp for Atmospheric Research
address@hidden             WWW: http://www.unidata.ucar.edu/
===============================================================================

---------- Forwarded message ----------
Date: Mon, 02 Aug 1999 16:01:41 -0500
From: Chad Johnson <address@hidden>
To: "Alaric S. Haag" <address@hidden>,
    Unidata Support <address@hidden>,
    "Chad W. Johnson" <address@hidden>,
    address@hidden, address@hidden,
    address@hidden, address@hidden
Subject: Re: 19990801: No McIDAS AREA files (cont.)

Chad Johnson wrote:
> 
> "Alaric S. Haag" wrote:
> >
> > On Mon, 2 Aug 1999, Chad Johnson wrote:
> >
> > >Unidata Support wrote:
> > >>
> > >> As you can see, we got two East IRs and one West one.  The strange thing
> > >> that I am seeing is that a notifyme from Unidata to unidata.ssec.wisc.edu
> > >> times out while trying to make a connection.
> > >
> > >Tom,
> > >
> > >In our ldmd logs I am seeing many gethostbyaddr failures and lookup 
> > >timeout failures. The exact
> > >errors look like.
> > >
> > >Aug 02 14:08:24 unidata rpc.ldmd[8929]: gethostbyaddr: failed for 
> > >130.39.129.3
> > >Aug 02 14:08:24 unidata rpc.ldmd[8929]: Denying connection from 
> > >130.39.129.3
> > >
> > >The feeds effected appear to be...
> > >
> > >128.174.80.44
> > >128.95.89.2
> > >141.213.23.92
> > >198.115.158.1
> > >128.118.28.2
> > >130.39.129.3
> > >
> > >Our Technical Computing staff is looking into the problem. Please bear 
> > >with us.
> > >
> > >-Chad
> >
> > Folks,
> >
> > Once again, we (sirocco.lsu, i.e. 130.39.129.3) seem to be part of the
> > problem. I'm still mystified. The logs are showing constant connection
> > time-outs to unidata, but when I restart the feed, it makes the
> > connection OK. I'll dig a bit deeper...
> 
> OK, now _I've_ restarted our LDM here and it looks like connections from feed 
> sites are getting
> through and products are getting broadcasted. However, nslookup's continue to 
> fail to 130.39.129.3
> and other IP addresses. Now I'm really confused. Why would restarting the LDM 
> fix (if only
> temporarily) the gethostbyaddr problem? I will continue to monitor the feed 
> and see if we start
> running into the same problem - similar to yesterday after restarting the LDM.
> 
> We are using LDM 5.0.8 with Solaris 7 on a Sparc platform.

Hi,

I believe the problem we were having was a result of a flaky name server. Our 
primary name server
was up and running, but behaving unusual. The cause of this bahavior was 
unknown, but suspected to
be the result of the power interruptions we have experienced this past week. We 
believe the fact
that the primary name server was up and running, prevented failing over to the 
secondary name
server. Name service has been restored and things appear to be functioning 
nominally the past few
hours.

We apologize for the interruption in service.

-Chad

-- 
Chad W. Johnson                           E-mail: address@hidden
Programmer/Meteorologist                  Voice: (608) 265-5292
Space Science and Engineering Center      Fax: (608) 263-6738
University of Wisconsin -- Madison