[PKG-Openstack-devel] openstack debian image improvements

[Thomas Goirand]

On 05/21/2015 08:24 PM, Gaudenz Steinlin wrote:
> Thomas Goirand <thomas at goirand.fr> writes:
>
>> On 05/20/2015 09:46 AM, Gaudenz Steinlin wrote:
>>> It's the nature of future failures that we don't know what they are and
>>> how we would debug them. So IMO it's best to have as much debug output
>>> as possible available at a place that's not obstrusive.
>>
>> It's my opinion that kernel debug output is very verbose and not giving
>> any clues of the type of failures we may have. If we can redirect that
>> to ttyS0 or something, then why not, but I don't believe it's convenient
>> on tty0
>
> I was always talking about ttyS0, not tty0. That's where the nova
> console-log output comes from. I agree that it's not very helpful on
> tty0 and don't care if the output is there or not.

Ah, cool! Then we agree. Then how would you send the kernel output to 
only ttyS0?

>> I wonder how fast you must clicking on horizon to be able to reach the
>> menu before the timeout. This doesn't seem a realistic way to handle
>> things in the cloud use case.
>
> With the default timeout of grub this works just fine if you first
> connect to the console and then reboot/reset the instance.

IMO, increasing the timeout too much just because you want to be able to 
reach the rescue shell is definitively not a nice option for the cloud 
usage, where you want instances to start as fast as possible.

Thomas