[GEMPAK #XUF-635799]: DCMSFC Sig 11

Adam,

You probably want to allow core dumps with "limit coredumpsize unlimited" 
(assuming csh/tcsh, for bash try "ulimit -c unlimited"), run dcmsfc manually 
("dcmsfc -b 9 -d ~/dcmsfc.log -e GEMTBL=/home/ldm/nawips/gempak/tables 
~/temp.gem", assuming /home/ldm/nawips/gempak/tables is the correct tables path 
- change as needed), and if this seg faults then run "gdb decoders/dcmsfc 
core.XXXX" where "decoders/dmsfc" is the path to dcmsfc and "XXXX" is the 
appropriate core file number.  In gdb type "where" and copy/paste the stack 
trace back to me.  Sorry for the lengthy procedure but there's little to 
diagnose without knowing when and were the decoder is seg faulting, and having 
the stack trace available through the core file and debugger will help us quite 
a bit.

Michael James
Unidata User Support


> Hello,
> 
> 
> 
> I recently compiled 5.11.4 on a CentOS 5.3 X86_64 based system (RHEL5
> basically).  Everything seemed to compile fine but one decoder is doing a
> Sig 11 constantly..dcmsfc.
> 
> 
> 
> Here is the ldmd.log error:
> 
> 
> 
> Jun 25 20:04:36 tornado pqact[8265] NOTE: child 9989 terminated by signal
> 11
> (decoders/dcmsfc-b9-ddata/gempak/logs/dcmsfc.log-eGEMTBL=/home/ldm/nawips/
> gempak/tablesdata/gempak/ship/YYYYMMDDHH_sb.gem)
> 
> 
> 
> And here is the dcmsfc.log error:
> 
> 
> 
> dcmsfc.log:[9989] 090625/1504[DC 3]  Starting up. Version 5.11.4
> 
> 
> 
> That all it says about the process in the log.
> 
> 
> 
> My pqact entry is:
> 
> 
> 
> WMO
> (^S[IMN]V[^GINS])|(^S[IMN]W[^KZ])|(^S(HV|HXX|S[^X]))|(^SX(VD|V.50|US(2[0-3
> ]|08|40|82|86)))|(^Y[HO]XX84)
> 
> PIPE    decoders/dcmsfc -b 9
> 
> -d data/gempak/logs/dcmsfc.log
> 
> -e GEMTBL=/home/ldm/nawips/gempak/tables
> 
> data/gempak/ship/YYYYMMDDHH_sb.gem
> 
> #
> 
> # 6 Hour Ship Observations
> 
> WMO     (^S[IMN]V[^GINS])|(^S[IMN]W[^KZ])
> 
> PIPE    decoders/dcmsfc -d data/gempak/logs/dcmsfc_6hr.log -a 6
> 
> -e GEMTBL=/home/ldm/nawips/gempak/tables
> 
> data/gempak/ship6hr/YYYYMMDDHH_ship.gem
> 
> 
> 
> 
> 
> There are no cores..so what I just told you is all I know.
> 
> 
> 
> NOTE:  I did try to move a 32bit binary over from another system..I get
> the same Sig 11.no change.
> 
> 
> 
> Thanks,
> 
> 
> 
> Adam Taylor
> 
> Computing Center
> 
> University of Louisiana Monroe
> 
> 
> 
> 
> 


Ticket Details
===================
Ticket ID: XUF-635799
Department: Support GEMPAK
Priority: High
Status: Open