Bug#1004577: ldconfig -p coredumps
control: severity -1 minor
control: retitle -1 libc-bin: ldconfig aborts when using linux 2.6 personality
control: found -1 glibc/2.26-1
On 2022-01-30 19:03, Christoph Berg wrote:
> Package: libc-bin
> Version: 2.33-3
> Severity: important
>
> In https://salsa.debian.org/python-team/packages/python-telethon/-/jobs/2413916
> there is a diff generated between the two builds because a core file
> from `ldconfig -p` appears as /usr/lib/python3.10/dist-packages/core.
>
> Backtrace:
>
> [0] 18:56 myon@sid-amd64.maxwell:~/de/py/debian/output/reprotest 1j $ gdb /sbin/ldconfig core
> GNU gdb (Debian 10.1-2) 10.1.90.20210103-git
> Copyright (C) 2021 Free Software Foundation, Inc.
> License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
> This is free software: you are free to change and redistribute it.
> There is NO WARRANTY, to the extent permitted by law.
> Type "show copying" and "show warranty" for details.
> This GDB was configured as "x86_64-linux-gnu".
> Type "show configuration" for configuration details.
> For bug reporting instructions, please see:
> <https://www.gnu.org/software/gdb/bugs/>.
> Find the GDB manual and other documentation resources online at:
> <http://www.gnu.org/software/gdb/documentation/>.
>
> For help, type "help".
> Type "apropos word" to search for commands related to "word"...
> Reading symbols from /sbin/ldconfig...
> Reading symbols from /usr/lib/debug/.build-id/f3/4bf815c30c307353fa1703469d5f1f4b3c9356.debug...
> [New LWP 12315]
> Core was generated by `/sbin/ldconfig -p'.
> Program terminated with signal SIGABRT, Aborted.
> #0 0x00007f205bb9f621 in ?? ()
> (gdb) bt
> #0 0x00007f205bb9f621 in ?? ()
> #1 0x0000000000000000 in ?? ()
This just points to the code that exit with SIGABRT. This happens
because in that job ldconfig is run with the linux 2.6 personality
(using setarch --uname-2.6) while the minimum supported kernel version
is 3.2 since stretch. This is what happens in practice:
$ setarch --uname-2.6 /sbin/ldconfig
FATAL: kernel too old
Aborted
This only happens for static binaries like ldconfig which get the faked
kernel version through uname, while dynamically linked binaries are able
to get the real kernel version from the ELF auxiliary vectors.
> The build artifacts are available from the salsa page; I don't have
> any access to the system there.
>
> There is another ldconfig segfault reported in #806911, I don't
> know if that is related.
It's related in the sense that they both try to use the linux 2.6
personality to test for reproductible builds. That said in the case of
#806911 ldconfig crashed with a segmentation fault, and that bug got
fixed in the meantime (I just realized that now and closed the bug),
while in your case it exits properly with an abort.
I do believe it's actually a minor issue, we ship > 3.x kernels since
Wheezy, so support for the 2.6 personality is now rather pointless.
Regards,
Aurelien
--
Aurelien Jarno GPG: 4096R/1DDD8C9B
aurelien@aurel32.net http://www.aurel32.net
Reply to: