Heya all - so I ran a standard upgrade these evening, which upgraded pam; I was headed for a reboot (since the new 4.0.3 kernel was also updated) and checked my logs just because; my fcron daemon was failing before the reboot. Huh. pacman log: [2015-05-18 18:28] [ALPM] upgraded pam (1.1.8-5 -> 1.2.0-1)[2015-05-18 18:28] [ALPM] upgraded pam (1.1.8-5 -> 1.2.0-1) journal: May 18 18:28:22 grimm fcron[13534]: PAM unable to dlopen(/usr/lib/security/pam_unix.so): /usr/lib/libpam.so.0: version `LIBPAM_MODUTIL_1.1.9' not found (required by /usr/lib/security/pam_unix.so) (... followed by lots of broken pipe errors) So I had rebooted already, which means I could not test further in the same error state -- but basically, should the pam upgrade have triggered a reload of systemd? Based on lsof, that's who has this file held open: $ lsof | grep libpam systemd 1512 tengel mem REG 254,2 60008 141189 /usr/lib/libpam.so.0.84.1 My thought is if this was truly broken, then I would not have been able to log in if I had not rebooted (pure coincidence I did) since pam_unix.so was dying; fcron as a daemon doesn't have it open, so I can only assume it was attempting to do a login (it's cron after all) to run crontabs, which is systemd running the show. Anyone have a system they haven't upgraded in ~24hrs to give it a further test, see if logins are broken after the pam upgrade? -te
