Re: Opinion re 2.0.30->2.0.32 in stable production servers?

John Robinson (john@intelligent.co.uk)
Wed, 19 Nov 1997 18:28:41 GMT

Stefan Monnier writes ("Re: Opinion re 2.0.30->2.0.32 in stable production servers?"):
> Lars Marowsky-Bree <lmb@pointer.teuto.de> writes:
> > which will be valid until someone decides to teardrop his servers, and then
> > he will have a hard time compiling a fixed kernel if they crash his machine
> > every 5 seconds unless of course he cuts the netconnection which might not
> > satisfy his customers ;-)

Not at all, I could (temporarily) get my firewall to block that source.

> How about an intermediate alternative like
> "compile the new kernel and install it for the next reboot" ?
> This way you don't reboot unnecessarily and if someone crashes the machine, you
> can take advantage of it to switch to the new version.

Actually I've done it the other way round. I built a new (2.0.32)
kernel and rebooted at an appropriate (idle) moment, leaving the old
kernel as the default kernel with which to boot. So if 2.0.32 goes
belly-up (22 hours with no problems so far) then without intervention
it'll be running what I was happy with before again. Seemed like a
reasonable compromise to me! The new will become the standard after a
few days.

John. _ _
-- _ | |___| |_ _ _
John Robinson 46 Bank Street, Dumfries DG1 2PA, UK | |_| / . \ ' \| ' \
+44 1387 247249 http://www.intelligent.co.uk/~john/ \___/\___/_||_|_||_|
You're footloose but you just feel limbless; life gets in the way --Del Amitri