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

Re: sequence numbers, ... (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: Wed, 02 Jun 1999 09:35:58 -0500
From: Chad Johnson <address@hidden>
To: Robb Kambic <address@hidden>
    Steve Chiswell <address@hidden>,
    support-ldm <address@hidden>
Subject: Re: sequence numbers, ...

Robb Kambic wrote:
> 
> Chad,
> 
> I'll update you on the Alden status.
> 
> On the NOAAport ingest machine, I was looking in the /tmp directory and
> noticed many spool files. Are these files created when the data isn't
> being read fast enough from the fifo files?  With all the new radar data
> now being sent over NOAAport, is the pqing/machine quick enough to read
> all the data?  I don't remember that many inge processes running on the
> machine either, currently there are 5.

Robb,

Another addition that is in this version of the ingestor is the ability to feed
two separate copies of the binary and text feeds. The primary data feed 
continues
on /tmp/jmb.fifo.1 and /tmp/jmb.fifo.2. /tmp/jmb.fifo.4 is the second copy of 
the
text stream and /tmp/jmb.fifo.5 is the second copy of the binary stream. Like
fifos 1 and 2, readers must be reading from both fifos 4 and 5. To facilitate
this addition, another copy of inge is forked so five copies of inge is normal.
Because the secondary feed is being fed by another process, readers do not have
be reading from both feeds. Data will continue to flow out the primary feed if
there are no readers on the secondary feed. Also, because two processes are now
reading from the spoolfiles, the spoolfiles are no longer removed. They simply
get reused. That is why you see spool files in the 

> Also, in the inge.log file, I see many messages about:
> 
> NMC 1999.152.212645:bit error
> NMC 1999.152.212733:spoolfiles wrapped, data lost
> NMC 1999.152.213101:hdlc problem
> NMC 1999.152.213201:bit error
> NMC 1999.152.213211:spoolfiles wrapped, data lost
> NMC 1999.152.213644:spoolfiles wrapped, data lost
> NMC 1999.152.214117:spoolfiles wrapped, data lost
> NMC 1999.152.214550:spoolfiles wrapped, data lost

This is an issue and we are looking into it. We are seeing the same thing here
and it appears as though it is an erroneous error message introduced when we
added the second set of fifos. I didn't catch it while debugging because I had
additional debugging messages added which cluttered the output. We believe you
are not losing any data. We are looking into resolving the problem.

I will get back to you about the error messages later today.

-Chad

> On Tue, 1 Jun 1999, Chad Johnson wrote:
> 
> > Mike Schmidt wrote:
> > >
> > > Chad,
> > >
> > > We've made the change here.  Can you update your system and not request
> > > data from Alden until they update their system.  Robb will notify you
> > > when they make the change on their end...
> >
> > OK. I've updated our system. Let me know when I can attach to Alden again.
> >
> > -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
> >
> 
> ===============================================================================
> Robb Kambic                                Unidata Program Center
> Software Engineer III                      Univ. Corp for Atmospheric Research
> address@hidden                   WWW: http://www.unidata.ucar.edu/
> ===============================================================================

-- 
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