At 14:27 01/12/2000 +0100, Christian Carstensen wrote:
>On Wed, 12 Jan 2000, Donn Miller wrote:
>
>> My guess is that once -current gets closer to the release date, it becomes
>> more and more stable. I guess the period of greatest instability occurs
>> somewhere about 1/4 to 1/2 through the -current life cycle. We could do a
>> chart plotting stability vs. time for the life cycle of a given
>> -current. That could help people decide whether or not they want to run
>> -current.
>
>This would be great, but I wonder from what source we could take reliable
>data about -current's stability.
>But what I've meant was: I've had these ugly system freezes not perfactly
>reproducable, but very often. From what I've read on current list, the
>problems still exist, but not on my system. At least this system runs
>stable for 1 day now. I'm wondering, why.
>
How 'bout some sort of client program that is run via the rc.d and
rc.shutdown scripts?
When run on bootup it checks dmesg for "WARNING: / was not properly
dismounted", and tells a master server whether or not the last reboot was
intentional.
When run at shutdown it tells the master server the machine's uptime.
Of course it would also help to send a 'uname -v' in both situations. This
system would have statistical flaws, but it is still an interesting idea.
Tom Embt
[EMAIL PROTECTED]
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message
