Re: Strang shutdown mechanism with Debian
>>>>> "Martin" == Martin Schulze <joey@tapiola.Infodrom.North.DE> writes:
Martin> At K90 the last "normal" script is run, so we really only
Martin> have K91 up to K99. If we move K89 (atd) to K79 and K90
Martin> (sysklogd) to K80, we'll have another 10 sequence numbers.
I agree utterly with this. My basic question was: why can't the
script sequences be renumbered? The fact that apache decided to be
K91 some time ago does not make forcing ourselves into a technically
sub-optiomal solution right. I would suggest moving atd to K69 or
K70. That way we don't souddenly run short again.
Does Debian need a shut-down scripts order in the polocy document? :))
Martin> How do other unixes achieve this? E.g. a Slowaris box at
Martin> work uses 5 scripts that need to run a particular order,
Martin> we only have 2 of them and 3 that they doesn't have.
Martin> Fortunately they don't have Sxx scripts that need to be
Martin> called with the "stop" argument. :-)
I have 14 /etc/rc0.d/K* scripts on each of the first two solaris boxes
I bothered to check here at work. They don not run the same sequence
of commands at shut-down and there is no problem with adding
additional commands to either sequence, i.e. there' space between the
numbers.
Martin> Right at the beginning he admitted that the mechanism is
Martin> broken and that it would be dangerous to change it now as
Martin> it would break several systems.
OK, my apologies Miguel if I seemed inflamatory - I didn't intend to.
Maybe I shouldn't send e-mail right after drinking a double latte. :)
--
Stephen
---
all coders are created equal; that they are endowed with certain
unalienable rights, of these are beer, net connectivity, and the
pursuit of bugfixes... - Gregory R Block
--
To UNSUBSCRIBE, email to debian-devel-request@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Reply to: