Hi,
On 23-05-16 20:04, Manuel Bilderbeek wrote:
3. My BIOS told me at boot today that no RTC clock time was set. I guess
my RTC battery is dead. Will replace it.
Replaced it last night. Then booted again and set the hwclock. Then
booted Debian, no fsck, as I still have the e2fsck.conf in place.
Then I removed the file and turned off the PC.
Booted again today and... yet another set of fsck's, but now at least
the reason is clear:
mei 24 19:08:51 sonata systemd-fsck[326]: /dev/sda1 has filesystem last
checked time in the future, check forced.
mei 24 19:13:06 sonata systemd-fsck[332]: /dev/sda6 has filesystem last
checked time in the future, check forced.
Then a bit later the timesync kicks in:
mei 24 19:33:51 sonata dbus[506]: [system] Failed to activate service
'org.bluez': timed out
mei 24 19:33:52 sonata systemd-timesyncd[462]: Synchronized to time
server 94.228.220.14:123 (2.debian.pool.ntp.org).
mei 24 19:34:28 sonata realmd[1318]: quitting realmd service after timeout
There is no notification that time changed now, so I guess the hwclock
is fine now.
Let's see what happens the coming days. I hope that the replaced CMOS
battery avoids the issues I had.
The main thing of this bug report is still the following: if a 'simple'
user runs into this problem (hwclock starts failing, but BIOS doesn't
report it's bad yet), he's doomed to end up in getting fsck's all the
time. Such a 'simple' user won't understand and probably run away from
Debian. What could be done for such user?
A little more advanced user would check the logs, but as I posted in my
original report: the log did NOT mention any reason why the fsck was done.
He could probably find out that something is wrong with his hwclock
though, due to the time changes that show up in the logs. But I don't
think he could reasonably know that he could work around it with that
e2fsck.conf. (But he should fix his hwclock anyway.) He may think of
installing fake-hwclock though.
Anyway, thanks for your help so far.
--
Kind regards,
Manuel
