You are not logged in.
Pages: 1
I noticed the following in the my logs, happening since yesterday. After logging to the machine, the systemd unit user@$USERID goes into failed state, yet the login completes and the system can be used just fine:
$ ssh root@host
# systemctl status user@0
× user@0.service - User Manager for UID 0
Loaded: loaded (/usr/lib/systemd/system/user@.service; static)
Active: failed (Result: exit-code) since Thu 2022-08-04 14:54:54 CEST; 3s ago
Docs: man:user@.service(5)
Process: 2654366 ExecStart=/usr/lib/systemd/systemd --user (code=exited, status=224/PAM)
Main PID: 2654366 (code=exited, status=224/PAM)
CPU: 8ms
Aug 04 14:54:54 systemd[1]: Starting User Manager for UID 0...
Aug 04 14:54:54 systemd[2654366]: PAM unable to dlopen(/usr/lib/security/pam_unix.so): /usr/lib/libc.so.6: version `GLIBC_ABI_DT_RELR' not found (required by /usr/lib/libresolv.so.2)
Aug 04 14:54:54 systemd[2654366]: PAM adding faulty module: /usr/lib/security/pam_unix.so
Aug 04 14:54:54 systemd[2654366]: PAM failed: Module is unknown
Aug 04 14:54:54 systemd[2654366]: user@0.service: Failed to set up PAM session: Operation not permitted
Aug 04 14:54:54 systemd[2654366]: user@0.service: Failed at step PAM spawning /usr/lib/systemd/systemd: Operation not permitted
Aug 04 14:54:54 systemd[1]: user@0.service: Main process exited, code=exited, status=224/PAM
Aug 04 14:54:54 systemd[1]: user@0.service: Failed with result 'exit-code'.
Aug 04 14:54:54 systemd[1]: Failed to start User Manager for UID 0.
# objdump -x /usr/lib/libc.so.6 | grep -C2 GLIBC_ABI_DT_RELR
37 0x00 0x069691b6 GLIBC_2.36
GLIBC_2.35
38 0x00 0x00fd0e42 GLIBC_ABI_DT_RELR
GLIBC_2.36
39 0x00 0x0963cf85 GLIBC_PRIVATE
# objdump -x /usr/lib/libresolv.so.2 | grep -B1 -A6 GLIBC_ABI_DT_RELR
required from libc.so.6:
0x00fd0e42 0x00 12 GLIBC_ABI_DT_RELR
0x06969194 0x00 11 GLIBC_2.14
0x0d696914 0x00 10 GLIBC_2.4
0x069691b4 0x00 09 GLIBC_2.34
0x0963cf85 0x00 08 GLIBC_PRIVATE
0x09691a75 0x00 07 GLIBC_2.2.5
0x0d696913 0x00 06 GLIBC_2.3
Package "glibc" has been upgraded from 2.35-6 to 2.36-1 yesterday, so I guess I'll have to wait until the "pam" package has been rebuilt to 2.36?
Last edited by ckujau (2022-08-04 15:02:33)
Offline
libc.so.6 in 2.36 defines that version. I would guess you're overriding /usr/lib/libc.so.6 somewhere?
Offline
Hm, what do you mean by "override"?
# ls -l /usr/lib/libc.so.6
-rwxr-xr-x 1 root root 1936696 Jul 31 22:47 /usr/lib/libc.so.6
# pacman -Q -kk glibc
backup file: glibc: /etc/locale.gen (Modification time mismatch)
backup file: glibc: /etc/locale.gen (Size mismatch)
backup file: glibc: /etc/locale.gen (MD5 checksum mismatch)
backup file: glibc: /etc/locale.gen (SHA256 checksum mismatch)
glibc: 1596 total files, 0 altered files
The rest of the package appears to be OK, same for the "pam" package.
Offline
I mean you have another libc.so.6 somewhere else higher in your linker path, so that /usr/lib/libc.so.6 isn't being used.
Offline
No, not that I know of. This is a pretty standard (hah!) Arch installation, no other glibc present here.
# locate libc.so.6
/usr/lib/libc.so.6
# find / -xdev -name libc.so.6 -ls
138883 1896 -rwxr-xr-x 1 root root 1936696 Jul 31 22:47 /usr/lib/libc.so.6
Offline
readelf -sW /tmp/libc.so.6 | grep GLIBC_ABI_DT_RELR
readelf -sW /tmp/libc.so.6 | grep 2.36
ldd /usr/lib/libresolv.so.2
Offline
# readelf -sW /usr/lib/libc.so.6 | grep -e GLIBC_ABI_DT_RELR -Fe 2.36
53: 0000000000108ca0 37 FUNC GLOBAL DEFAULT 15 open_tree@@GLIBC_2.36
418: 0000000000000000 0 OBJECT GLOBAL DEFAULT ABS GLIBC_2.36
815: 0000000000000000 0 OBJECT GLOBAL DEFAULT ABS GLIBC_ABI_DT_RELR
898: 0000000000039bd0 17 FUNC WEAK DEFAULT 15 arc4random_buf@@GLIBC_2.36
1400: 0000000000039c40 342 FUNC WEAK DEFAULT 15 arc4random_uniform@@GLIBC_2.36
1425: 0000000000108cd0 37 FUNC GLOBAL DEFAULT 15 pidfd_open@@GLIBC_2.36
1701: 00000000001089d0 40 FUNC GLOBAL DEFAULT 15 fsconfig@@GLIBC_2.36
1852: 0000000000108dc0 37 FUNC GLOBAL DEFAULT 15 process_mrelease@@GLIBC_2.36
1903: 00000000000bc460 413 FUNC GLOBAL DEFAULT 15 c8rtomb@@GLIBC_2.36
2046: 0000000000108d60 40 FUNC GLOBAL DEFAULT 15 pidfd_send_signal@@GLIBC_2.36
2211: 00000000000bc250 516 FUNC GLOBAL DEFAULT 15 mbrtoc8@@GLIBC_2.36
2270: 0000000000108d90 40 FUNC GLOBAL DEFAULT 15 process_madvise@@GLIBC_2.36
2288: 0000000000108a60 37 FUNC GLOBAL DEFAULT 15 fspick@@GLIBC_2.36
2314: 0000000000108a30 37 FUNC GLOBAL DEFAULT 15 fsopen@@GLIBC_2.36
2404: 0000000000108c10 40 FUNC GLOBAL DEFAULT 15 mount_setattr@@GLIBC_2.36
2546: 0000000000108c40 40 FUNC GLOBAL DEFAULT 15 move_mount@@GLIBC_2.36
2814: 0000000000108a00 37 FUNC GLOBAL DEFAULT 15 fsmount@@GLIBC_2.36
2952: 0000000000039bf0 70 FUNC WEAK DEFAULT 15 arc4random@@GLIBC_2.36
2955: 0000000000108d00 37 FUNC GLOBAL DEFAULT 15 pidfd_getfd@@GLIBC_2.36
# ldd /usr/lib/libresolv.so.2
linux-vdso.so.1 => linux-vdso.so.1 (0x00006c56dbeb4000)
libc.so.6 => /usr/lib/libc.so.6 (0x00006c56dbca9000)
/usr/lib64/ld-linux-x86-64.so.2 => /usr/lib64/ld-linux-x86-64.so.2 (0x00006c56dbeb6000)
Offline
I separated the greps so I wouldn't have to look for it in your reply, but oh, well.
So the symbol is there and you're also on the 2.36 library - did you reboot after the update (speculating for some filesystem cache here)?
Offline
No, I have not rebooted yet, just checked with "lsof" for stale file handles, but nothing related to glibc turned up.
Offline
Sadl^W Interestingly, a reboot did the trick, and the errors are gone now :-\ Maybe someone replaced my system with a Windows VM, hm? :-)
@all: Thanks for your help here, much appreciated.
Offline
btrfs?
Offline
No, ext4. The only "exotic" thing about this machine would be the linux-hardened kernel from Arch, apart from that it's pretty boring, I think.
$ mount | grep ext
/dev/xvda2 on / type ext4 (rw,relatime)
/dev/xvda1 on /boot type ext4 (rw,nosuid,nodev,relatime,stripe=4)
(Huh, that stripe option is weird though, but not related to that topic here.)
Last edited by ckujau (2022-08-04 16:19:45)
Offline
Pages: 1