[Raw Msg Headers][Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

zmailer-2.99-42 /SPARC /Solaris smtp dumps core




Hello all,

zmailer-2.99-42 persists in segfaulting and dropping core for smtp
on a SPARC:

#mailq
MailQ on srv
0 entries in router queue: idle
1 message in transport queue: working (core exists)
Transport queue is empty -- or scheduler uses -Q -mode


#ls -lR /zmailer/postoffice/transport/
total 275
drwxr-xr-x   3 root          512 Nov 25 22:10 ./
drwxr-xr-x  13 root          512 Nov 25 22:08 ../
drwxr-xr-x   2 root          512 Nov 25 22:08 Q/

-rw-r--r--   1 root       268556 Nov 25 22:21 core
/zmailer/postoffice/transport/Q:
total 3
drwxr-xr-x   2 root          512 Nov 25 22:08 ./
drwxr-xr-x   3 root          512 Nov 25 22:10 ../
-r--rw----   1 jmack         661 Nov 25 22:23 26588-13351

#file /zmailer/postoffice/transport/core
/zmailer/postoffice/transport/core:     ELF 32-bit MSB core file SPARC Version 1, from 'smtp'


Just earlier on I discovered that smtp insists on looking in:

	/var/spool/postoffice/transport

(despite my config which says POSTOFFICE is /zmailer/postoffice), so I
symlinked my location to /var/spool/postoffice, but no better progress:

#zmailer-2.99.42/transports/smtp>gdb /zmailer/bin/ta/smtp \ 
				--core=/zmailer/postoffice/transport/core
GDB is free software and you are welcome to distribute copies of it
 under certain conditions; type "show copying" to see the conditions.
There is absolutely no warranty for GDB; type "show warranty" for details.
GDB 4.16 (sparc-sun-solaris2.5), Copyright 1996 Free Software Foundation, Inc...
Core was generated by `/zmailer/bin/ta/smtp -s'.
Program terminated with signal 11, Segmentation fault.
Reading symbols from /usr/lib/libsocket.so.1...done.
Reading symbols from /usr/lib/libnsl.so.1...done.
Reading symbols from /usr/lib/libc.so.1...done.
Reading symbols from /usr/lib/libdl.so.1...done.
Reading symbols from /usr/lib/libintl.so.1...done.
Reading symbols from /usr/lib/libmp.so.1...done.
Reading symbols from /usr/lib/libw.so.1...done.
#0  0xef69e3a4 in _dowrite ()
(gdb) where
#0  0xef69e3a4 in _dowrite ()
#1  0xef69e5bc in _doprnt ()
#2  0xef6aafa4 in _fprintf ()
#3  0x16b08 in smtpconn (SS=0xeffff8f0, host=0x4e290 "tshimshan.phys.ualberta.ca") at smtp.c:1661
#4  0x1687c in smtpopen (SS=0xeffff8f0, host=0x4e290 "tshimshan.phys.ualberta.ca") at smtp.c:1591
#5  0x144e0 in main (argc=0, argv=0xeffffe34) at smtp.c:625
(gdb) l 625
620               }
621               if (!SS.smtpfp) {
622                 if ((first_uid = atoi(dp->senders->misc)) < 0 ||
623                     first_uid == nobody)
624                   first_uid = daemon_uid;
625                 smtpstatus = smtpopen(&SS, host);
626                 if (smtpstatus != EX_OK) {
627                   /* SMTPOPEN() made diagnostics data -- we report.. */
628                   struct rcpt *rp;
629                   /* NOTARY: addres / action / status / diagnostic */


removing the symlink /var/spool/postoffice, and re-running gdb shows in 
addition to the above:

...
#0  0xef69e3a4 in _dowrite ()
.gdbinit:4: Error in sourced command file:
/var/spool/postoffice/transport: No such file or directory.
(gdb) 


Nowhere had I defined the postoffice to live there! I can't figure out 
where it thinks this should be the POSTOFFICE location.

my build config call  was:

./configure  	--with-bundled-libresolv \
		--with-gcc \
		--with-postoffice=/zmailer/postoffice \
		--with-prefix=/zmailer

SiteConfig essentials are:

MAILBIN=        /zmailer/bin
MAILSHARE=      /zmailer/share
MAILVAR=        /zmailer
POSTOFFICE=     /zmailer/postoffice
SMTPOPTIONS=            "-asve -l /zmailer/log/smtpserver"
SCHEDULEROPTIONS=       "-l /zmailer/log/scheduler.perflog -Q -S -H"


There are still sections of the new configure which really still don't do 
the "right" thing perhaps ( for instance, binaries get thrown into 
/usr/local/bin unless '--with-prefix=/zmailer' or some other location is
explicitly specified - my thinking on this is that they should go to 
$MAILBIN as SiteConfig requests that they should... ).


Any ideas appreciated.

back to 2.99-31 for now.

Cheers,
--
James S. MacKinnon           Office: P-139 Avahd-Bhatia Physics Lab
Team Physics                 Voice : (403) 492-8226
University of Alberta        email : Jim.MacKinnon@Phys.UAlberta.CA
Edmonton, Canada T6G 2N5
        WWW:   http://www.phys.ualberta.ca/~jmack/jmack.html