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

Re: request for data kills LDM

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: request for data kills LDM
  • Date: Fri, 2 Apr 1999 10:58:21 -0700 (MST)
On Thu, 1 Apr 1999, Glenn P. Davis wrote:

> 
> > I wrote to you about this bug a couple weeks ago.  The ldm starts up ok
> > and starts receiving data fine. But if a downstream tries to connect and
> > receive data, then the upstream ldm crashes on a signal 11. This is with
> > the 5.0.7.internal release.  I did a feedme from laraine to weather3 and
> > the ldm died. I was running the weather3 ldm in verbose and debug mode but
> > didn't get much more information. Here's the log messages from weather3.
> >
> > This is on a Linux system at NIU. I also received another report from the
> > NWS having the same problem, previous e-mail messages.
> >
> > Robb...
> 
> 
> Okay.
> 
> Which version of LINUX?
Linux sunshine.unidata.ucar.edu 2.2.5 #1 SMP Tue Mar 30 11:45:14 MST 1999
i686 unknown

> Which version of LDM?

-rw-rw-r--   1 rkambic  ustaff    815721 Mar  2 11:17
/export/dist/ftp/pub/ldm5/ldm-5.0.7internal.tar.Z   

I thought this was the latest code(release) ?

> 
> If the LDM version is the one built here for LINUX, is the version of
> LINUX where it is being run _eariler_ than the one you built one?
> If so, it won't work.
> 
> If it is the same or later than the one we have here, can you duplicate
> the problem on sunshine?
> 
Yep, the ldm is compiled and located at: /tmp/ldm   It's currently running
and receiving DDPLUS from zero.  The ldmd.log.1 file contains a crash
similar to the one I sent yesterday. The crash was created when I did a
feedme on laraine to sunshine.  For some reason, I can't get the ldm to
dump core.  I have in my .cshrc:

limit coredumpsize 100M

still no core file.  I don't understand this. Anyhow, I would suggest
doing a feedme to sunshine and watch the results.


Robb...

> What is signal 11 on the target machine? SIGSEGV? (Usually the answer is in
> /usr/include/sys/signal.h, but on LINUX you'll have to follow the
> includes from /usr/include/signal.h.)
> 
> If you have access to a problem system (or can reproduce the problem here)
> do
> % gdb rpc.ldmd corefile
> 
> and it will say  something like
> 
> Core was generated by `rpc.ldmd'.
> Program terminated with signal 11, Segmentation fault.
> Reading symbols from /lib/libc.so.6...done.
> Reading symbols from /lib/ld-linux.so.2...done.
> #0  ....where error occurred ...
> 
> typing
> (gdb) where
> 
> will yield more info.
> 
> -glenn
> 

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