Hi all! I was myself experiencing bug #407466. Every time I shut my chrony-equipped box down the hwclock failed to be updated; as a consequence, as soon as I booted the box up again, I received a local mail message (to root) saying that the chrony correction exceeded the threshold. These corrections got bigger and bigger (more than 150 s, recently).
By reading this bug log, I found the suggestion by Daniel Vetter: http://bugs.debian.org/407466#25 I tried it out and it seems to work: my box booted up and the correction that chrony had to apply was really small (about 1.2 s). IMHO, the following commands are the manual fix that users may apply to an already installed chrony package: # update-rc.d -f chrony remove # update-rc.d chrony defaults 83 22 These commands set the stop script order for chrony to 22, which is before the hwclock stop script is executed (25). The long term fix for future users of the chrony package requires a change in its postinst script, I think: the line update-rc.d chrony defaults 83 >/dev/null should become update-rc.d chrony defaults 83 22 >/dev/null Please note that, IIUC, this fix would work for users who will install the new version of chrony from scratch, but would not fix things for users who upgrade from a previous version. Maybe a note in the README.Debian could refer to this bug and explain things. John Hasler, please make this modification as soon as possible, if you agree that it would solve the issue. -- New location for my website! Update your bookmarks! http://www.inventati.org/frx ..................................................... Francesco Poli . GnuPG key fpr == C979 F34B 27CE 5CD8 DC12 31B5 78F4 279B DD6D FCF4
pgpHOxzli1XrJ.pgp
Description: PGP signature
