Hello,
> Else:
>
> Configure your system to permit setuid programs to dump core; set
> rlimits accordingly; grab coredump, run:
> $ gdb /path/to/exim /path/to/exim.core
> and issue the "bt" command, which will tell you where it died, and
> suggest which library it was in.
>
I try version with coredump. Run:
gdb ./exim-bad /var/log/dumps/core.12383
Reading symbols from /usr/sbin/exim-bad...(no debugging symbols
found)...done.
warning: core file may not match specified executable file.
Reading symbols from /lib64/libresolv.so.2...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libresolv.so.2
Reading symbols from /lib64/libnsl.so.1...(no debugging symbols
found)...done.
Loaded symbols for /lib64/libnsl.so.1
.
.
.
.
.
Loaded symbols for /lib64/libnss_files.so.2
Core was generated by `/usr/sbin/exim -Mc 1R2Kw7-0003Dd-HH'.
Program terminated with signal 11, Segmentation fault.
#0 0x000000000046b660 in smtp_read_response ()
then I run "bt"
(gdb) bt
#0 0x000000000046b660 in smtp_read_response ()
#1 0x0000000000491950 in smtp_deliver ()
#2 0x000000000049451e in smtp_transport_entry ()
#3 0x00000000004232c1 in do_remote_deliveries ()
#4 0x00000000004265ea in deliver_message ()
#5 0x000000000042f884 in main ()
