[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[LDM #DCW-448836]: pqsurf dying on SunOS 5.11, LDM 6.6.5
- Subject: [LDM #DCW-448836]: pqsurf dying on SunOS 5.11, LDM 6.6.5
- Date: Fri, 21 Dec 2007 12:46:09 -0700
Art,
The last messages from the "pqact" child-process of "pqsurf":
Dec 21 19:20:13 ls3 pqact[12793] DEBUG: Delay: 0.0011 sec
Dec 21 19:20:13 ls3 pqact[12793] DEBUG: pq_sequence():
time(insert)-time(create): 0.1378 s
Dec 21 19:20:13 ls3 pqact[12793] INFO: 0910dc2d976a186a97f5662edac4abba
83 20071221192013.203 IDS|DDPLUS 191502569 metar KDTL 211920
Dec 21 19:20:13 ls3 pqact[12793] INFO: dbfile:
data/weathernew/20071221/2007122119.metar_gdb KDTL
Dec 21 19:20:13 ls3 pqact[12793] DEBUG: ldmdb_prodput: 2 metar KDTL 211920
Dec 21 19:20:13 ls3 pqsurf[12789] DEBUG: bbc4d544979b42e0eb9c4bc01ac9a7ad
68 20071221192013.203 IDS|DDPLUS 191502582 metar KMRJ 211920
Dec 21 19:20:13 ls3 pqact[12793] DEBUG: Concatenating data under key
KDTL
would seem to indicate that it died while trying to concatenate
a data-product during a DBFILE action.
Your intuition might be right and the DBFILE action is broken
under SunOS 5.11.
Can you get a core-dump of the "pqact" child-process? Remove the
"-vx" option from the EXEC of "pqsurf" and ensure that corefiles
are allowed in your session (via the "limit" or "ulimit" command).
Then start an LDM session. The corefile should be in the LDM
user's home-directory. Then use the "dbx" utility to get a trace
of the stack.
Regards,
Steve Emmerson
Ticket Details
===================
Ticket ID: DCW-448836
Department: Support LDM
Priority: Normal
Status: On Hold