[PATCH conductor 0/2] #786844 - Reporting on deleted instances (v2)

Matt Wagner matt.wagner at redhat.com
Tue Feb 14 14:08:14 UTC 2012


On Tue, Feb 14, 2012 at 10:09:01AM +0100, Jan Provaznik wrote:
> When reading it now, one question I should have probably asked earlier:
> since we support only stateless instances, wouldn't be better to
> change state to 'stopped' instead of deleting a vanished instance?
> This behaviour reflects EC2 state change better - if an instance is
> stopped, it 'disappears', and RHEVM (or other stateful) instances
> can't be started again anyway. Advantage of this is that a user has
> access to the instance's events and he can track what happened.

So I think there are actually two things here.

The first is that I prefer something like "vanished" over "stopped", as
"stopped" to me implies that the user took some action to stop the
instance. Although "stopped" with a log entry could communicate the same
information, I suppose. But I think it's important to, in some manner,
distinguish between instances that were stopped on purpose and instances
that somehow got into a weird state.

The second is that the deletion of instances was meant as a stopgap
solution, since this came in as a bug report long after feature freeze.
We'd like to keep instances in a state indicating this error, and have
some UI treatment to indicate the problem. So I largely agree with you,
but I would argue that it's not in scope for the current sprint.

-- Matt



More information about the aeolus-devel mailing list