[Linux-HA] "Clones, Stonith and Suicide" The SysAdmin who had a nervous breakdown.
beekhof at gmail.com
Fri Oct 5 02:55:50 MDT 2007
i believe dejan has answered most of the questions here, i just wanted
to re-iterate a point about suicide plugins...
On 10/2/07, Peter Farrell <peter.d.farrell at gmail.com> wrote:
> On 02/10/2007, Dejan Muhamedagic <dejanmm at fastmail.fm> wrote:
> > The suicide stonith device is not exactly the best approach.
> > Ultimately it is not reliable, so it should not be used on the
> > production clusters. If you can afford it, get a real (hardware)
> > stonith device.
> Can't. No budget. Advice taken - I'll have to kill these via SSH or suicide.
the inherent problem with a suicide plugin is that you're relying on a
sick node to shoot itself... depending on exactly how sick it is, that
may or may not be impossible.
wrt. to ssh... the "meatware" plugin would arguably be better a better
option, at least from a reliability perspective, as it will work in
all cases (providing the admin is prepared to get up at 3am or
whenever the node goes bad :-)
but if you're "only" trying to protect against resource failures (not
node failures), then ssh is probably sufficient.
More information about the Linux-HA