Open Stack

On 23 January 2014 09:19, Fox, Kevin M <Kevin.Fox at pnnl.gov> wrote:
> I think most of the time taken to reboot is spent in bringing down/up the services though, so I'm not sure what it really buys you if you do it all. It may let you skip the crazy long bootup time on "enterprise" hardware, but that could be worked around with kexec on the full reboot method too.

kexec is not the answer you are looking for - we did give it a spin,
but because the hardware is not left in the state the new OS expects,
it is super fragile: when it works, it works, when it doesn't its OMG
pain.

Also no, restarting all the openstack related services is a few seconds:
real    0m41.683s
user    0m3.747s
sys     0m1.451s

(thats on a control plane node)

OCP might power on faster than that, but still, its only slightly
longer than SYN timeouts.

And yes, enterprise hardware boot times are crazy long :)


-Rob

-- 
Robert Collins <rbtcollins at hp.com>
Distinguished Technologist
HP Converged Cloud