Closed rschalo closed 3 months ago
Name | Link |
---|---|
Latest commit | 5525628c4ac860dcfba53f91b63712d5b9d5d1a8 |
Latest deploy log | https://app.netlify.com/sites/karpenter-docs-prod/deploys/6669ed525494310008768b23 |
Files with Coverage Reduction | New Missed Lines | % | ||
---|---|---|---|---|
pkg/providers/amifamily/ami.go | 1 | 90.56% | ||
<!-- | Total: | 1 | --> |
Totals | |
---|---|
Change from base Build 9475149132: | -0.01% |
Covered Lines: | 5550 |
Relevant Lines: | 6730 |
I'm on board with changing this to be the default, but should we surface a configuration option in the helm chart to override this default? If we did this from the start I'd defer to user demand, but this would now be a breaking change for customers who may have infra already set up relying on this being a ScheduleAnyway
TSC.
So, my thinking is that if leave this as is and don't surface an option to override, there are two kinds of cx that would break. The first are those that would already break given a zonal outage, which would not different behavior with DoNotSchedule
as the default. The other case, is that Karpenter is running two replicas on the same node, the leader pod goes unhealthy, loses leadership, and then there is not a second node because of the DoNotSchedule
TSC.
This PR changes the value which can be changed by cx, setting this by default requires a chart default update that we can revisit later but I think changing the values and making this recommendation is sufficient for now.
Fixes #N/A
Description Change default behavior so that both replicas cannot schedule to the same availability zone.
How was this change tested?
Does this change impact docs?
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.