Re: 20020803: several messages from Gilbert re: LDM 5.2 crashing

Gilbert,

I was able to get LDM 5.2 to crash on Linux 7.1 system, didn't crash on a
Solaris 5.9 system though.  It appears that the problem is pqsurf.  So
have you tried running the system w/o pqsurf? and then trying to run 5.2
with 5.1.4 pqsurf?

Let me know the results.

Robb...





On Tue, 6 Aug 2002, Unidata Support wrote:

> 
> These were sitting in the support inbox, so I figured I should send them
> along.
> 
> Tom
> 
> ------- Forwarded Messages
> 
> >From: Gilbert Sebenste <address@hidden>
> >Subject: AHA! Gotcha!
> >Organization: NIU
> >Keywords: 200208030122.g731MA929565 LDM 5.2 pqsurf
> 
> LDM 5.2 crashes continue. But I now have a clue of why:
> 
> Aug 03 00:29:26 weather rpc.ldmd[5249]: child 5253 terminated by signal 11 
> Aug 03 00:29:26 weather rpc.ldmd[5249]: Killing (SIGINT) process group 
> Aug 03 00:29:26 weather rpc.ldmd[5249]: Interrupt 
> Aug 03 00:29:26 weather rpc.ldmd[5249]: Exiting 
> Aug 03 00:29:26 weather pqact[5254]: Interrupt 
> Aug 03 00:29:26 weather pqact[5254]: Exiting 
> Aug 03 00:29:26 weather pqbinstats[5250]: Interrupt 
> Aug 03 00:29:26 weather pqbinstats[5250]: Exiting 
> Aug 03 00:29:26 weather 131.156.8.58[5257]: Interrupt 
> Aug 03 00:29:26 weather 131.156.8.58[5257]: Exiting 
> Aug 03 00:29:26 weather pqact[5252]: Interrupt 
> Aug 03 00:29:26 weather pqact[5252]: Exiting 
> Aug 03 00:29:26 weather striker[5258]: Interrupt 
> Aug 03 00:29:26 weather weather2(feed)[5387]: Interrupt 
> Aug 03 00:29:26 weather weather2(feed)[5388]: Interrupt 
> 
> 
> So what caused this? What is child 5253?
> 
> Aug 02 16:59:08 weather pqsurf[5253]: Starting Up (5249) 
> 
> Now...why?
> 
> >From address@hidden Sat Aug  3 00:05:39 2002
> >Subject: LDM 5.2 crashes again, but again...
> 
> Now that I am saving more than one log, I can now see clearly what is 
> happening. Well, kinda.
> 
> Aug 03 04:06:55 weather nldn2md[30075]: Appending to MD file: 75 
> Aug 03 04:06:55 weather nldn2md[30075]: NLDN2MD - Flash events in file: 
> 58101 
> Aug 03 04:06:55 weather nldn2md[30075]: NLDN2MD -- DONE 
> Aug 03 04:09:33 weather rpc.ldmd[7861]: child 7865 terminated by signal 11 
> Aug 03 04:09:33 weather rpc.ldmd[7861]: Killing (SIGINT) process group 
> Aug 03 04:09:33 weather rpc.ldmd[7861]: Interrupt 
> Aug 03 04:09:33 weather rpc.ldmd[7861]: Exiting 
> Aug 03 04:09:33 weather pqbinstats[7862]: Interrupt 
> Aug 03 04:09:33 weather pqbinstats[7862]: Exiting 
> Aug 03 04:09:33 weather pqact[7864]: Interrupt 
> 
> And here is the process:
> 
> Aug 03 01:22:54 weather pqsurf[7865]: Starting Up (7861) 
> 
> >From address@hidden Sat Aug  3 00:25:18 2002
> >Subject: LDM 5.2 problem---question
> 
> Hello again,
> 
> Maybe it would help, if pqsurf is really causing the problem...if you 
> could check out my pqsurf.conf file.
> 
> #=============================================================================
> # pqsurf.conf
> #
> # Physical Plant
> # Northern Illinois University
> #
> # Last changed:  December 19, 1998
> #
> #=============================================================================
> # Surface reports (SAOs and METARs) are extracted from their reports and 
> individually
> # filed in two files; one for U.S., Canada, and Mexico, and other.
> #
> # AMOS observations (we think).
> #WMO  ^SXUS03 K... ([0-3][0-9])([0-2][0-9])
> #     FILE    domestic/sao/amos_stuff.(\1:yy)(\1:mmm)\1
> # SAOs.
> #WMO  ^sao (.*) (.*) ([0-3][0-9])([0-2][0-9])
> #     FILE    domestic/sao/sao.(\3:yy)(\3:mmm)\3_\4
> # METARs.
> #WMO  ^metar (.*) ([0-3][0-9])([0-2][0-9])
> #     FILE    domestic/metar/metar.(\2:yy)(\2:mmm)\2_\3
> #-----------------------------------------------------------------------------
> # Dumping both SAOs and METARs into one file.
> # SAOs.
> #WMO  ^sao (.*) (.*) ([0-3][0-9])([0-2][0-9])
> #     FILE    domestic/surface_obs/(\3:yy)(\3:mmm)\3_\4
> # METARs.
> #WMO  ^metar (.*) ([0-3][0-9])([0-2][0-9])
> #     FILE    domestic/surface_obs/(\2:yy)(\2:mmm)\2_\3
> # METARs.
> #WMO  ^speci (....) ([0-3][0-9])([0-2][0-9])
> #     FILE    domestic/surface_obs/(\2:yy)(\2:mmm)\2_\3
> #
> #Doing it the right way.
> #
> WMO   ^(sao .. ...|metar ....|speci ....) ([0-3][0-9])([0-2][0-9])
>       FILE    domestic/surface_obs/(\2:yy)(\2:mmm)\2_\3
> 
> *******************************************************************************
> Gilbert Sebenste                                                     ********
> Internet: address@hidden    (My opinions only!)                     ******
> Staff Meteorologist, Northern Illinois University                      ****
> E-mail: address@hidden                                 ***
> web: http://weather.admin.niu.edu                                      **
> Work phone: 815-753-5492                                                *
> *******************************************************************************
> 
> ------- Message 2
> 
> >To: General Support <address@hidden>
> >From: Gilbert Sebenste <address@hidden>
> >Subject: LDM 5.2 crashed...this time, on 5.2!
> >Organization: UCAR/Unidata
> >Keywords: 200208031656.g73GuI910481
> 
> Hey gang,
> 
> Well, so much for the theory that it only crashes when feeding from
> 5.1.4 machines. Weather2, solely feeding off a 5.2 machine (atm) crashed 
> last night 15 minutes after I checked it around 8Z. I restarted it about
> 16Z. Again, it complained about pqsurf.
> 
> *******************************************************************************
> Gilbert Sebenste                                                     ********
> Internet: address@hidden    (My opinions only!)                     ******
> Staff Meteorologist, Northern Illinois University                      ****
> E-mail: address@hidden                                 ***
> web: http://weather.admin.niu.edu                                      **
> Work phone: 815-753-5492                                                *
> *******************************************************************************
> 
> 
> ------- Message 3
> 
> >To: General Support <address@hidden>
> >From: Gilbert Sebenste <address@hidden>
> >Subject: LDM (fwd)
> >Organization: UCAR/Unidata
> >Keywords: 200208040516.g745Gk917401
> 
> Wow. Well, I'm at 5.1.4 and also feeding from COD now. atm.geo was 45 
> minutes behind on everything tonight. Weird...see also message 
> attached below.
> 
> P.S. Will be at a conference from Tuesday evening on next week, so 
> anything you want for me to try has to be done before 4 PM Tuesday.
> 
> *******************************************************************************
> Gilbert Sebenste                                                     ********
> Internet: address@hidden    (My opinions only!)                     ******
> Staff Meteorologist, Northern Illinois University                      ****
> E-mail: address@hidden                                 ***
> web: http://weather.admin.niu.edu                                      **
> Work phone: 815-753-5492                                                *
> *******************************************************************************
> 
> ---------- Forwarded message ----------
> Date: Sat, 3 Aug 2002 23:31:28 -0500
> From: Chris Novy <address@hidden>
> To: address@hidden
> Subject: LDM
> 
> Glad you're back on 5.1.4.  Your problems appear to cause queue 
> corruption on my end resulting in multiple mailings to the WX-***** 
> lists.
> 
> ..Chris..
> 
> 
> ------- End of Forwarded Messages
> 
> 

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