Re: Scary bugs
On Sun, 30 Jan 2000, Thierry Laronde wrote:
> On Sun, Jan 30, 2000 at 12:07:09PM -0200, Henrique M Holschuh wrote:
> > I've just noticed an *EVIL* piece of misinformation (for Debian) in man
> > hwclock, which implies that the user should use hwclock to set the RTC
> > without mangling the system clock and reboot to update the system clock.
> > THIS might be one of the causes of all this mess, since we currently have
> > --systohc enabled. Added to the list of small doc updates.
> I think that this is a *bad* idea to modify a man page because of some
> specificities of the installation (as you point out, the man page can be
Hmm... you're right, of course. I was going to add a small Debian users:
paragraph, but I can either leave it alone, or add a one line to the top of
the man page: "Debian users: Please read (file) before using this program,
or you risk clock nightmares."
And (file) tells the user in a simple language that unless he chages the rc
scripts, any changes done with hwclock _might_ be undone on shutdown, and
points to another file with the full explanation. It also tells the user to
simply use 'date' if he wants to change the system date.
> If the doc is put in front of the user (one or two `echo' lines in
> `hwclock.sh', the READMEs) the clueless user has no reason to go reading
> the man page if he doesn't know what he is trying to do.
For #3 (no --adjust at all, --systohc still there) I propose:
"Setting the system clock using the hardware clock as reference...
Clock set. Local time: ..."
"Saving the system clock to the hardware clock...
"Hardware clock updated to (date)."
This should help people figure out that indeed something is touching a
"clock" during startup and shutdown. The old shutdown message "CMOS clock
updated" is not as good because hwclock docs always talk about the "Hardware
Would that be enough, or should "Edit /etc/init.d/hwclock.sh to disable" be
added to the shutdown messages? I think that would be going too far.
BTW, would such messages violate the policy?
> So, here is my humble opinion :
> > Okay, which one? #1, #2 or #3?
> Okay for #3 (actual case) but WITH the basic explanations you wrote, and,
> in addition, as a kind of morality : "a unix system doesn't allow the
> direct access to the devices, so, if you have no idea about what's going
> on, access the clock via the system --- ie via `date'".
I'll start working on doc patches for #3 then.
"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