[exim-dev] [Bug 2274] exim 4.91: segfault ... error 4 in libc-2.17.so

[admin--- via Exim-dev]

https://bugs.exim.org/show_bug.cgi?id=2274

--- Comment #5 from Alex  ---
Hello Heiko,

Thanks for your reply.

(In reply to Heiko Schlittermann from comment #1)
> So, the next time the same message is delivered via the same transport, it
> does not segfault?

Yes, as soon as we restart exim and spamd all the new emails flow fine, the
same senders and recipients as well.

Coredumps were configured, but the fact exim is a suid binary slipped out of my
mind, so a coredump was never generated.

re-configured with the setting:

fs.suid_dumpable = 1

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-- 
## List details at https://lists.exim.org/mailman/listinfo/exim-dev Exim 
details at http://www.exim.org/ ##

[exim-dev] [Bug 2274] exim 4.91: segfault ... error 4 in libc-2.17.so

[admin--- via Exim-dev]

https://bugs.exim.org/show_bug.cgi?id=2274

--- Comment #6 from Alex  ---
Hello Jeremy,

Thanks for your reply.

(In reply to Jeremy Harris from comment #4)
> To get more info, configure your system to take corefiles (see "man core",
> including the setuid specifics), and install the debuginfo package for glibc.
> Once you have a coredump, run (as root) gdb with the Exim binary and the
> corefile, and run the "bt" command to get a stack trace.
> 
> Then turn suid coredumps off again; they're a security problem.

It's now re-configured with

fs.suid_dumpable = 1 

so need to wait for a next crash.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-- 
## List details at https://lists.exim.org/mailman/listinfo/exim-dev Exim 
details at http://www.exim.org/ ##

[exim-dev] [Bug 2274] exim 4.91: segfault ... error 4 in libc-2.17.so

[admin--- via Exim-dev]

https://bugs.exim.org/show_bug.cgi?id=2274

Heiko Schlittermann  changed:

   What|Removed |Added

 CC||h...@schlittermann.de

--- Comment #1 from Heiko Schlittermann  ---
So, the next time the same message is delivered via the same transport, it does
not segfault?

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-- 
## List details at https://lists.exim.org/mailman/listinfo/exim-dev Exim 
details at http://www.exim.org/ ##

[exim-dev] [Bug 2274] exim 4.91: segfault ... error 4 in libc-2.17.so

[admin--- via Exim-dev]

https://bugs.exim.org/show_bug.cgi?id=2274

--- Comment #4 from Jeremy Harris  ---
To get more info, configure your system to take corefiles (see "man core",
including the setuid specifics), and install the debuginfo package for glibc.
Once you have a coredump, run (as root) gdb with the Exim binary and the
corefile, and run the "bt" command to get a stack trace.

Then turn suid coredumps off again; they're a security problem.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-- 
## List details at https://lists.exim.org/mailman/listinfo/exim-dev Exim 
details at http://www.exim.org/ ##

[exim-dev] [Bug 2274] exim 4.91: segfault ... error 4 in libc-2.17.so

[admin--- via Exim-dev]

https://bugs.exim.org/show_bug.cgi?id=2274

--- Comment #2 from Heiko Schlittermann  ---
Do you know how to enable core dumps on set-uid binaries? This could help for
further tracing.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-- 
## List details at https://lists.exim.org/mailman/listinfo/exim-dev Exim 
details at http://www.exim.org/ ##

[exim-dev] [Bug 2274] New: exim 4.91: segfault ... error 4 in libc-2.17.so

[admin--- via Exim-dev]

https://bugs.exim.org/show_bug.cgi?id=2274

Bug ID: 2274
   Summary: exim 4.91: segfault ... error 4 in libc-2.17.so
   Product: Exim
   Version: 4.91
  Hardware: x86-64
OS: Linux
Status: NEW
  Severity: bug
  Priority: medium
 Component: Content Scanning
  Assignee: t...@duncanthrax.net
  Reporter: ale...@nsk21.ru
CC: exim-dev@exim.org

Hello,

I've got Exim 4.91 segfaulting on CentOS 7, the case is not reproducible by
hands. It occurs without any noticeable logic.

In exim's maillog we have:

2018-05-07 16:16:30 1fFgwH-nJ-4f ** u...@domain.com F=
R=spamcheck_director T=spamcheck: Child process of spamcheck transport (running
command "/usr/sbin/exim -oMr spam-scanned -bS") was terminated by signal 11
(Segmentation fault)

In messages we have:

exim[3061]: segfault at 0 ip 7f9a3620fc8f sp 7ffc3eb39a18 error 4 in
libc-2.17.so[7f9a3617b000+1b8000]

The exim version:

# exim -bV
Exim version 4.91 #4 built 09-May-2018 15:25:19
Copyright (c) University of Cambridge, 1995 - 2018
(c) The Exim Maintainers and contributors in ACKNOWLEDGMENTS file, 2007 - 2018
Berkeley DB: Berkeley DB 5.3.21: (May 11, 2012)
Support for: crypteq IPv6 Perl OpenSSL move_frozen_messages Content_Scanning
DKIM DNSSEC Event OCSP PRDR SPF TCP_Fast_Open Experimental_SRS
Lookups (built-in): lsearch wildlsearch nwildlsearch iplsearch cdb dbm dbmjz
dbmnz dnsdb
Authenticators: cram_md5 dovecot plaintext spa
Routers: accept dnslookup ipliteral manualroute queryprogram redirect
Transports: appendfile/maildir/mailstore/mbx autoreply lmtp pipe smtp
Malware: f-protd f-prot6d drweb aveserver fsecure kavdaemon sophie clamd mksd
avast sock cmdline
Fixed never_users: 0
Configure owner: 0:0
Size of off_t: 8
2018-05-09 15:44:04 cwd=/usr/local/directadmin/custombuild 2 args: exim -bV
Configuration file is /etc/exim.conf

Spamcheck transport:

spamcheck:
  driver = pipe
  batch_max = 100
  command = /usr/sbin/exim -oMr spam-scanned -bS
  current_directory = "/tmp"
  group = mail
  home_directory = "/tmp"
  log_output
  message_prefix =
  message_suffix =
  return_fail_output
  no_return_path_add
  transport_filter = /usr/bin/spamc -u
${lookup{$domain}lsearch*{/etc/virtual/domainowners}{$value}}
  use_bsmtp
  user = mail



OS kernel: 3.10.0-693.21.1.el7.x86_64
Perl: natively installed from official repo: perl-5.16.3-292.el7.x86_64


Could you let me know how to trace further the segfult and fix it?

Regards,
Alex.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-- 
## List details at https://lists.exim.org/mailman/listinfo/exim-dev Exim 
details at http://www.exim.org/ ##