we want to ensure that the platform 'survives' even if an application
is not 'drainable'
it already happened that some applications are getting stuck in a
termination process and won't recover anymore. This can happen for
various reasons (volume not detachable, podDisruptionBudget ...)
as we are using the drain during every update, we are likely to run
into this issue multiple times per cluster
we want to ensure that the platform 'survives' even if an application is not 'drainable'
it already happened that some applications are getting stuck in a termination process and won't recover anymore. This can happen for various reasons (volume not detachable, podDisruptionBudget ...)
as we are using the drain during every update, we are likely to run into this issue multiple times per cluster
Signed-off-by: Maximilian Meister mmeister@suse.de