Re: Crash problem

compiled with the CFLAGS now

no systemd no

yes, have tried this now but no core file is dumped

ulimit -c is "unlimited" so it is enabled.

have done a echo 1 > /proc/sys/fs/suid_dumpable (it was 0)

it now crashes (but continue the queuescanner, so not stopping) with a 
signal 139

2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 SIGSEGV (fault address: (nil))
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 SIGSEGV (null pointer 
indirection)
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 SIGSEGV (qrun-p1-delivery: 
55577 delivering 1vFj5z-0000000016v-2hc1 (queue run pid 54750)
)
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 backtrace
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 ---
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 
/usr/exim/bin/exim-4.99-2(stackdump+0x1f) [0x46c834]
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 /usr/exim/bin/exim-4.99-2() 
[0x46ca13]
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 
/lib64/libpthread.so.0(+0xf770) [0x7fe1a7179770]
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 
/lib64/libc.so.6(strlen+0x12) [0x7fe1a6e77622]
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 /usr/exim/bin/exim-4.99-2() 
[0x4558d6]
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 /usr/exim/bin/exim-4.99-2() 
[0x455955]
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 /usr/exim/bin/exim-4.99-2() 
[0x45c2e2]
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 
/usr/exim/bin/exim-4.99-2(deliver_message+0x38e5) [0x466b7c]
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 
/usr/exim/bin/exim-4.99-2(queue_run+0xd5c) [0x4a2cc2]
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 
/usr/exim/bin/exim-4.99-2(single_queue_run+0x123) [0x4a33b4]
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 
/usr/exim/bin/exim-4.99-2(main+0x74f6) [0x475e03]
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 
/lib64/libc.so.6(__libc_start_main+0xfd) [0x7fe1a6e13b6d]
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 /usr/exim/bin/exim-4.99-2() 
[0x441b99]
2025-11-04 13:06:06 1vFj5z-0000000016v-2hc1 ---
2025-11-04 13:06:06 queue run: process 55577 crashed with signal 139 
while delivering 1vFj5z-0000000016v-2hc1

tried to run it directly via gdb, after finishing with "Program exited 
normally" and typing bt I get "no stack".

On 2025-11-04 11:00, Jeremy Harris via Exim-users wrote:
> OK, that didn't detect anything.  We need to investigate a core.
>
> Do a build with "CFLAGS += -O0 -ggdb".
>
> Does Slackware have the systemd-ish coredumpctl stuff?
> I suspect not; so we need to enable suid-coredumps.
>  (Over here, Fedora, the hints are in "man core" which mentions 
> /proc/sys/fs/suid_dumpable)
> and check that "ulimit" does not restrict the size of corefile.
>
> Run a test to get a crash, and locate the core file.
>
> Run up gdb - "sudo gdb <exim-binary-location> <corefile-location>".
>   Ignore all the offers to load debuginfo file for libraries.
>   At the prompt:   "bt"
> You should get a more comprehensive stacktrace, with file and line 
> number info.
>
>
> Copy that here.  Keep the corefile; we may want to pull some
> variable values out of it.
>
> (If you can do IRC, I'm in #exim on irc.libera.chat)

-- 
## subscription configuration (requires account):
##   https://lists.exim.org/mailman3/postorius/lists/exim-users.lists.exim.org/
## unsubscribe (doesn't require an account):
##   [email protected]
## Exim details at http://www.exim.org/
## Please use the Wiki with this list - http://wiki.exim.org/