You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, waitUntilCondition still polls the current state from the API server every 500ms(!!) statically, which is pretty damn often IMHO.
It seems to me like this is pretty harsh on the API server to be queried that frequently when waiting for a condition like a roll-out of a Deployment finishing.
Can you overload the method to add a parameter to configure the poll rate, or a function parameter to allow custom back-off functions, or just straight implement exponential back-off that resets when the resource changes or something like that?
The text was updated successfully, but these errors were encountered:
Oh sure, sorry. I am using 4.10.1, so it makes sense that this is outdated on my part.
I've been encountering a problem with 4.10.2 that prohibited me from upgrading to the latest version and therefore i have not upgraded yet:
I'm using DefaultKubernetesClient#load(InputStream) to load a rendered template, which worked in 4.10.1 but throws a Jackson exception like No resource type found for:Deployment in 4.10.2. Should i raise an issue with a reproducer for this instead?
Currently,
waitUntilCondition
still polls the current state from the API server every 500ms(!!) statically, which is pretty damn often IMHO.It seems to me like this is pretty harsh on the API server to be queried that frequently when waiting for a condition like a roll-out of a Deployment finishing.
Can you overload the method to add a parameter to configure the poll rate, or a function parameter to allow custom back-off functions, or just straight implement exponential back-off that resets when the resource changes or something like that?
The text was updated successfully, but these errors were encountered: