Re: [exim] 4.95-RC0 - SIGSEGV (maybe attempt to write to immutable memory) & other oddities
Quick and dirty follow up as I don't have a laptop with me right now. This is the 12.2 machine coredumps so far (2 distinct failures from what I can see). I see someone else has run into this on FreeBSD 13.0 too. I will try to help them if this doesn't once I am back on a real keyboard! Matthew core.exim-4.95-RC0-2.40.41367.core (lldb) target create "/usr/exim/bin/exim-4.95-RC0-2" --core "core.exim-4.95-RC0-2.40.41367.core" Core file '/var/spool/exim/core.exim-4.95-RC0-2.40.41367.core' (x86_64) was loaded. (lldb) bt exim-4.95-RC0-2 was compiled with optimization - stepping may behave oddly; variables may not be available. * thread #1, name = 'exim-4.95-RC0-2', stop reason = signal SIGSEGV * frame #0: 0x002cdb46 exim-4.95-RC0-2`smtp_start_session [inlined] string_from_gstring(g=0x0004) at functions.h:919:4 [opt] frame #1: 0x002cdb41 exim-4.95-RC0-2`smtp_start_session at smtp_in.c:3090 [opt] frame #2: 0x00260ff1 exim-4.95-RC0-2`daemon_go at daemon.c:528:8 [opt] frame #3: 0x00260eca exim-4.95-RC0-2`daemon_go at daemon.c:2594 [opt] frame #4: 0x00279d5e exim-4.95-RC0-2`main(argc=, cargv=) at exim.c:4947:3 [opt] frame #5: 0x00254f00 exim-4.95-RC0-2`_start(ap=, cleanup=) at crt1.c:76:7 core.exim-4.95-RC0-2.40.41404.core (lldb) target create "/usr/exim/bin/exim-4.95-RC0-2" --core "core.exim-4.95-RC0-2.40.41404.core" Core file '/var/spool/exim/core.exim-4.95-RC0-2.40.41404.core' (x86_64) was loaded. (lldb) bt exim-4.95-RC0-2 was compiled with optimization - stepping may behave oddly; variables may not be available. * thread #1, name = 'exim-4.95-RC0-2', stop reason = signal SIGSEGV * frame #0: 0x002cdb46 exim-4.95-RC0-2`smtp_start_session [inlined] string_from_gstring(g=0x001c) at functions.h:919:4 [opt] frame #1: 0x002cdb41 exim-4.95-RC0-2`smtp_start_session at smtp_in.c:3090 [opt] frame #2: 0x00260ff1 exim-4.95-RC0-2`daemon_go at daemon.c:528:8 [opt] frame #3: 0x00260eca exim-4.95-RC0-2`daemon_go at daemon.c:2594 [opt] frame #4: 0x00279d5e exim-4.95-RC0-2`main(argc=, cargv=) at exim.c:4947:3 [opt] frame #5: 0x00254f00 exim-4.95-RC0-2`_start(ap=, cleanup=) at crt1.c:76:7 core.exim-4.95-RC0-2.40.41540.core (lldb) target create "/usr/exim/bin/exim-4.95-RC0-2" --core "core.exim-4.95-RC0-2.40.41540.core" Core file '/var/spool/exim/core.exim-4.95-RC0-2.40.41540.core' (x86_64) was loaded. (lldb) bt exim-4.95-RC0-2 was compiled with optimization - stepping may behave oddly; variables may not be available. * thread #1, name = 'exim-4.95-RC0-2', stop reason = signal SIGSEGV * frame #0: 0x002cdb46 exim-4.95-RC0-2`smtp_start_session [inlined] string_from_gstring(g=0x014c) at functions.h:919:4 [opt] frame #1: 0x002cdb41 exim-4.95-RC0-2`smtp_start_session at smtp_in.c:3090 [opt] frame #2: 0x00260ff1 exim-4.95-RC0-2`daemon_go at daemon.c:528:8 [opt] frame #3: 0x00260eca exim-4.95-RC0-2`daemon_go at daemon.c:2594 [opt] frame #4: 0x00279d5e exim-4.95-RC0-2`main(argc=, cargv=) at exim.c:4947:3 [opt] frame #5: 0x00254f00 exim-4.95-RC0-2`_start(ap=, cleanup=) at crt1.c:76:7 core.exim-4.95-RC0-2.40.41549.core (lldb) target create "/usr/exim/bin/exim-4.95-RC0-2" --core "core.exim-4.95-RC0-2.40.41549.core" Core file '/var/spool/exim/core.exim-4.95-RC0-2.40.41549.core' (x86_64) was loaded. (lldb) bt exim-4.95-RC0-2 was compiled with optimization - stepping may behave oddly; variables may not be available. * thread #1, name = 'exim-4.95-RC0-2', stop reason = signal SIGSEGV * frame #0: 0x002cdb46 exim-4.95-RC0-2`smtp_start_session [inlined] string_from_gstring(g=0x022c) at functions.h:919:4 [opt] frame #1: 0x002cdb41 exim-4.95-RC0-2`smtp_start_session at smtp_in.c:3090 [opt] frame #2: 0x00260ff1 exim-4.95-RC0-2`daemon_go at daemon.c:528:8 [opt] frame #3: 0x00260eca exim-4.95-RC0-2`daemon_go at daemon.c:2594 [opt] frame #4: 0x00279d5e exim-4.95-RC0-2`main(argc=, cargv=) at exim.c:4947:3 [opt] frame #5: 0x00254f00 exim-4.95-RC0-2`_start(ap=, cleanup=) at crt1.c:76:7 core.exim-4.95-RC0-2.40.41588.core (lldb) target create "/usr/exim/bin/exim-4.95-RC0-2" --core "core.exim-4.95-RC0-2.40.41588.core" Core file '/var/spool/exim/core.exim-4.95-RC0-2.40.41588.core' (x86_64) was loaded. (lldb) bt exim-4.95-RC0-2 was compiled with optimization - stepping may behave oddly; variables may not be available. * thread #1, name = 'exim-4.95-RC0-2', stop reason = signal SIGSEGV * frame #0: 0x002cdb46 exim-4.95-RC0-2`smtp_start_session [inlined] string_from_gstring(g=0x06cc) at functions.h:919:4 [opt] frame
Re: [exim] 4.95-RC0 - SIGSEGV (maybe attempt to write to immutable memory) & other oddities
On 24/07/2021 11:17, Evgeniy Berdnikov via Exim-users wrote: The setuid bit is meaningful for local submissions. Your logs show that problem is observed for SMTP submissions, so your could run other Exim binary (without setuid bit) as a listening daemon. You generally need to be root to bind to a priviliged port, too - so unless you can duplicate on a nonstandard port using a custome client (eg. telnet), or use inetd-style listener mode... -- Cheers, Jeremy -- ## List details at https://lists.exim.org/mailman/listinfo/exim-users ## Exim details at http://www.exim.org/ ## Please use the Wiki with this list - http://wiki.exim.org/
Re: [exim] Exim 4.95-RC1 released
On Wed, Jul 28, 2021 at 11:30 AM Jeremy Harris via Exim-users < exim-users@exim.org> wrote: > On 28/07/2021 09:14, Odhiambo Washington via Exim-users wrote: > > On FreeBSD 13-RELEASE (clang version 11.0.1 -f that matters), I had to > > back out of RC0 to 4.94.2 because my paniclog was filled with several > lines > > of: > > > > *2021-07-24 18:16:23 SIGSEGV (maybe attempt to write to immutable > memory)* > > I think someone else also reported that (also on a BSD) and was trying to > get us a stack backtrace. We'll really need one to have an chance of > locating > this. > > @Jeremy Harris I could have followed any instructions geared towards identifying the cause, but unfortunately, I do not have enough time on my hands at the moment. The other person is running inside a jail, while I am not. He says: 1. sysctl kern.sugid_coredump=1 is necessary in FreeBSD for the core dumps. My system already has that enabled. 2. He used "-glldb and stopped the exim binary being stripped" - and that is an option passed to gcc. I don't know the equiv for clang. If you'd like to build a version that would coredump, I am happy to give you access to my server for that purpose. -- Best regards, Odhiambo WASHINGTON, Nairobi,KE +254 7 3200 0004/+254 7 2274 3223 "Oh, the cruft.", egrep -v '^$|^.*#' :-) -- ## List details at https://lists.exim.org/mailman/listinfo/exim-users ## Exim details at http://www.exim.org/ ## Please use the Wiki with this list - http://wiki.exim.org/
Re: [exim] Exim 4.95-RC1 released
Good Morning,
> On FreeBSD 13-RELEASE (clang version 11.0.1 -f that matters), I had to
> back out of RC0 to 4.94.2 because my paniclog was filled with several lines
> of:
>
> *2021-07-24 18:16:23 SIGSEGV (maybe attempt to write to immutable memory)*
>
> I have just installed RC1 and will report if anything strange happens.
I'm afraid the issue isn't solved yet, see the thread starting at
https://lists.exim.org/lurker/message/20210723.150306.145a081e.en.html
Best regards from Dresden/Germany
Viele Grüße aus Dresden
Heiko Schlittermann
--
SCHLITTERMANN.de internet & unix support -
Heiko Schlittermann, Dipl.-Ing. (TU) - {fon,fax}: +49.351.802998{1,3} -
gnupg encrypted messages are welcome --- key ID: F69376CE -
signature.asc
Description: PGP signature
--
## List details at https://lists.exim.org/mailman/listinfo/exim-users
## Exim details at http://www.exim.org/
## Please use the Wiki with this list - http://wiki.exim.org/
Re: [exim] Exim 4.95-RC1 released
On 28/07/2021 09:14, Odhiambo Washington via Exim-users wrote: On FreeBSD 13-RELEASE (clang version 11.0.1 -f that matters), I had to back out of RC0 to 4.94.2 because my paniclog was filled with several lines of: *2021-07-24 18:16:23 SIGSEGV (maybe attempt to write to immutable memory)* I think someone else also reported that (also on a BSD) and was trying to get us a stack backtrace. We'll really need one to have an chance of locating this. -- Cheers, Jeremy -- ## List details at https://lists.exim.org/mailman/listinfo/exim-users ## Exim details at http://www.exim.org/ ## Please use the Wiki with this list - http://wiki.exim.org/
Re: [exim] Exim 4.95-RC1 released
On Wed, Jul 28, 2021 at 10:53 AM Heiko Schlittermann via Exim-users <
exim-users@exim.org> wrote:
> Hi *,
>
> Thank you for the feedback we got since RC0, especially to Andreas
> Metzler who helped discovering and testing stuff around dbbc1c20b.
>
> We just released the next release candidate. Commits since RC0 are:
>
> * 03fc05ca1 - (HEAD -> master, tag: exim-4.95-RC1, origin/master,
> origin/HEAD) Docs: tidy variables lists (5 days ago)
> * 27d03dca1 - DKIM: fix build with older GnuTLS (5 days ago) Harris>
> * dbbc1c20b - TLS: fix tls_verify_certificates handling of "system" (5
> days ago)
> * 1c18b2f73 - Testsuite: output chnges resulting (5 days ago) Harris>
> * 25f3b885d - typo (7 days ago)
> * 651acf8b3 - Docs: enhance SPF description (7 days ago) Metzler>
> * 32451e8a6 - Docs: remove extraneous file copy (8 days ago) Harris>
> * f9d167e05 - typo (10 days ago)
> * 9138b6973 - ALPN: not supported under LibreSSL (10 days ago) Harris>
> * 32c45e838 - Fix no-TLS bulid (10 days ago)
> * d083e3f2a - ALPN: feature macro (10 days ago)
> * b634f8eaf - typo (10 days ago)
> * dbad58950 - typo (10 days ago)
> * c4b408623 - TLS: ALPN options (10 days ago)
> * f7ea5ba10 - Remove the must-helo check from the example config given
> that there is now a default-set option and hard code (2f8e0a5f6b) (11 days
> ago)
> * c968a17cc - NewStuff typo (12 days ago) (HS12-RIPE)>
> * 26916dc75 - Docs: Clarify $acl_verify_message lifetime (12 days ago)
>
> * 1f76af318 - Docs: fix formatting (2 weeks ago)
>
> As usual we ask *you* to do as much testing as possible and provide us
> feedback. This covers build issues with unusual libraries, runtime
> issues in unusual environments and any other kind of things that should
> be fixed.
>
> The Exim 4.95-RC1 is available
>
> - as tarball:https://ftp.exim.org/pub/exim/exim4/test
> - directly from Git: https://git.exim.org
> tag exim-4.95-RC1
>
> The tarball checksums are signed using the same GPG key as I used to
> sign this message (Key-ID: D0BFD6B9ECA5694A6F149DCEAF4CC676A6B6C142),
> as the tag and tagged commit are.
>
> This is the last chance for new features, as starting with RC2 (planned
> around Aug 5th) we won't add any new features to the upcoming release.
> If you have anything that should make it into the official release, and
> is mature enough to be included in this phase, please contact us.
>
> For changes since the previous stable version, please see the RC0
> announcement:
> https://lists.exim.org/lurker/message/20210715.212328.6bec444b.en.html
>
> @David Restall, @Thomas Noll: I'm not sure if we'll manage to fix the
> source of the compiler warnings for this release, but we'll try.
>
> Best regards from Dresden/Germany
> Viele Grüße aus Dresden
> Heiko Schlittermann
> --
> SCHLITTERMANN.de internet & unix support -
> Heiko Schlittermann, Dipl.-Ing. (TU) - {fon,fax}: +49.351.802998{1,3} -
> gnupg encrypted messages are welcome --- key ID: F69376CE -
> --
> ## List details at https://lists.exim.org/mailman/listinfo/exim-users
> ## Exim details at http://www.exim.org/
> ## Please use the Wiki with this list - http://wiki.exim.org/
>
On FreeBSD 13-RELEASE (clang version 11.0.1 -f that matters), I had to
back out of RC0 to 4.94.2 because my paniclog was filled with several lines
of:
*2021-07-24 18:16:23 SIGSEGV (maybe attempt to write to immutable memory)*
I have just installed RC1 and will report if anything strange happens.
--
Best regards,
Odhiambo WASHINGTON,
Nairobi,KE
+254 7 3200 0004/+254 7 2274 3223
"Oh, the cruft.", egrep -v '^$|^.*#' :-)
--
## List details at https://lists.exim.org/mailman/listinfo/exim-users
## Exim details at http://www.exim.org/
## Please use the Wiki with this list - http://wiki.exim.org/
[exim] Exim 4.95-RC1 released
Hi *,
Thank you for the feedback we got since RC0, especially to Andreas
Metzler who helped discovering and testing stuff around dbbc1c20b.
We just released the next release candidate. Commits since RC0 are:
* 03fc05ca1 - (HEAD -> master, tag: exim-4.95-RC1, origin/master,
origin/HEAD) Docs: tidy variables lists (5 days ago)
* 27d03dca1 - DKIM: fix build with older GnuTLS (5 days ago)
* dbbc1c20b - TLS: fix tls_verify_certificates handling of "system" (5 days
ago)
* 1c18b2f73 - Testsuite: output chnges resulting (5 days ago)
* 25f3b885d - typo (7 days ago)
* 651acf8b3 - Docs: enhance SPF description (7 days ago)
* 32451e8a6 - Docs: remove extraneous file copy (8 days ago)
* f9d167e05 - typo (10 days ago)
* 9138b6973 - ALPN: not supported under LibreSSL (10 days ago)
* 32c45e838 - Fix no-TLS bulid (10 days ago)
* d083e3f2a - ALPN: feature macro (10 days ago)
* b634f8eaf - typo (10 days ago)
* dbad58950 - typo (10 days ago)
* c4b408623 - TLS: ALPN options (10 days ago)
* f7ea5ba10 - Remove the must-helo check from the example config given that
there is now a default-set option and hard code (2f8e0a5f6b) (11 days ago)
* c968a17cc - NewStuff typo (12 days ago)
* 26916dc75 - Docs: Clarify $acl_verify_message lifetime (12 days ago)
* 1f76af318 - Docs: fix formatting (2 weeks ago)
As usual we ask *you* to do as much testing as possible and provide us
feedback. This covers build issues with unusual libraries, runtime
issues in unusual environments and any other kind of things that should
be fixed.
The Exim 4.95-RC1 is available
- as tarball:https://ftp.exim.org/pub/exim/exim4/test
- directly from Git: https://git.exim.org
tag exim-4.95-RC1
The tarball checksums are signed using the same GPG key as I used to
sign this message (Key-ID: D0BFD6B9ECA5694A6F149DCEAF4CC676A6B6C142),
as the tag and tagged commit are.
This is the last chance for new features, as starting with RC2 (planned
around Aug 5th) we won't add any new features to the upcoming release.
If you have anything that should make it into the official release, and
is mature enough to be included in this phase, please contact us.
For changes since the previous stable version, please see the RC0
announcement:
https://lists.exim.org/lurker/message/20210715.212328.6bec444b.en.html
@David Restall, @Thomas Noll: I'm not sure if we'll manage to fix the
source of the compiler warnings for this release, but we'll try.
Best regards from Dresden/Germany
Viele Grüße aus Dresden
Heiko Schlittermann
--
SCHLITTERMANN.de internet & unix support -
Heiko Schlittermann, Dipl.-Ing. (TU) - {fon,fax}: +49.351.802998{1,3} -
gnupg encrypted messages are welcome --- key ID: F69376CE -
signature.asc
Description: PGP signature
--
## List details at https://lists.exim.org/mailman/listinfo/exim-users
## Exim details at http://www.exim.org/
## Please use the Wiki with this list - http://wiki.exim.org/
