Re: "superblock last write time is in the future" on boot

On Fri, 06 Jan 2006, Ray Lanza wrote:
> Is using UTC the last word for this problem? I must dual-boot with windows on my machine.

No, of course not. It is the _easiest_ fix. It is becoming aparent that we
can do a much better fix in glibc, but I need to investigate more. And
there is always the workaround of setting TZ in /etc/init.d/hwclockfirst.sh
for the meanwhile.

> My linux configuration is relatively simple with everything on a single
> partition. Time zone is set properly, is not a symbolic link and is in the
> same filesystem as root.

So it should just work. Make sure /etc/default/rcS has UTC=no in it, and
that TZ is not set anywhere in the /etc/init.d/hwclock* scripts. Look at
what the system says while booting up, it will output the system time twice
(one for hwclockfirst.sh, the other for hwclock.sh). They must make sense
(also pay attention to the timezone it reports).

> I've noticed that when I have the system set for local time it reports a time that is really GMT.

"date -u" will tell you what the system thinks is the UTC time. If the
output is different from plain "date", then it certainly thinks it is in
some timezone.

> Everything worked fine until the last time I updated (debian/testing).

The current bugs in util-linux should not matter much if /usr is in the same
partition as /, so we will need a bit more info to help you.

--
"One disk to rule them all, One disk to find them. One disk to bring
them all and in the darkness grind them. In the Land of Redmond
where the shadows lie." -- The Silicon Valley Tarot
Henrique Holschuh


--
To UNSUBSCRIBE, email to debian-user-REQUEST@xxxxxxxxxxxxxxxx
with a subject of "unsubscribe". Trouble? Contact listmaster@xxxxxxxxxxxxxxxx

Relevant Pages

  • Re: "superblock last write time is in the future" on boot
    ... > for instance but the log entries are in UTC. ... programs with that broken TZ setting in the environment will run in UTC. ... "One disk to rule them all, ... To UNSUBSCRIBE, email to debian-user-REQUEST@xxxxxxxxxxxxxxxx ...
    (Debian-User)
  • Re: fsck Fails On Reboot After Partially Completed Testing Upgrade
    ... >If you cannot, well, we are fixing the bug but it will take some time. ... clock to UTC. ... "One disk to rule them all, ... To UNSUBSCRIBE, email to debian-user-REQUEST@xxxxxxxxxxxxxxxx with a subject of "unsubscribe". ...
    (Debian-User)
  • Re: fsck Fails On Reboot After Partially Completed Testing Upgrade
    ... >>If you cannot, well, we are fixing the bug but it will take some time. ... clock to UTC. ... "One disk to rule them all, ... To UNSUBSCRIBE, email to debian-user-REQUEST@xxxxxxxxxxxxxxxx ...
    (Debian-User)
  • Re: smartctl trackrecord
    ... reports "aborted by host" with 90% remaining. ... "One disk to rule them all, ... To UNSUBSCRIBE, email to debian-user-REQUEST@xxxxxxxxxxxxxxxx ...
    (Debian-User)
  • Re: ATA error messages
    ... Martin Schweizer writes: ... These are all various kinds of reports of the disk not reporting back ... To unsubscribe, ...
    (freebsd-questions)