Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 21 Sep 1997 14:30:51 +0300 (EET DST)
From:      mika ruohotie <bsdchat@shadows.aeon.net>
To:        scrappy@hub.org (The Hermit Hacker)
Cc:        wosch@cs.tu-berlin.de, chat@FreeBSD.ORG
Subject:   Re: uptime on hub.freebsd.org
Message-ID:  <199709211130.OAA21317@shadows.aeon.net>
In-Reply-To: <Pine.NEB.3.95.970919122818.22049G-100000@hub.org> from The Hermit Hacker at "Sep 19, 97 01:00:29 pm"

next in thread | previous in thread | raw e-mail | index | archive | help
> > i prefer my machines up to date, since rebooting new kernel
> > takes machine off for just a minute or two.
> 	That's great for non-production machines that don't need 24/7 uptime,

i do that on those 24/7 too...

> even longer if that new kernel has the slightly bug in it, something I 
> experienced with a 'stable' kernel on a 'not-so-stable' machine awhile back :(

ick. i've tried to solved it with a development desktop. one that'll
run same system than the production machines... yes, probs on higher
loads are not seen that way, but it's likely in few days someone on
the lists have seen them...

first i cvsup all into the same repository, then i make world & kernel
in the development machine. and if all goes fine for few days, i
go updating the other machines.

so far, i havent gotten bitten to my leg by that.

also, my home machine running current is somewhat "production" since it's
required to get me into internet, and to me getting my machine into a
state i can not get connected quolifyes multiple times more serious
than anything anything else i can think... but, even with current, in
past 3 years, software related problems havent been the cause to take
me off the net. for that machine i read the lists, and then compile
if it might be safe... occasionally i've been forced to reboot with 
the old kernel again, but even in those times i've gotten myself to net,
and server have been able to operate "as usual", coz i update frequently,
kernel.old works okish (top/w/ipfw might not work, that's all).

yes, i agree, production server might suffer from a reboots to a
new kernel. but it also might suffer from _not_ rebooting a new
kernel. and it's something every admin probably sees differently.

> Marc G. Fournier                                 scrappy@hub.org


mickey



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199709211130.OAA21317>