Skip to content

Tainted nodes are bidding  #253

Description

@88plug

Describe the bug
Some providers have nodes that are tainted in their cluster and the Akash provider is still using them to bid when a cluster is at capacity.

[Warning] [FailedScheduling] [Pod] 0/6 nodes are available: 1 Insufficient memory, 1 node(s) had untolerated taint {node-role.kubernetes.io/control-plane: }, 4 Insufficient cpu. preemption: 0/6 nodes are available: 1 Preemption is not helpful for scheduling, 5 No preemption victims found for incoming pod..
[FailedScheduling] [Pod] 0/4 nodes are available: 1 node(s) had untolerated taint {CriticalAddonsOnly: true}, 3 Insufficient cpu. preemption: 0/4 nodes are available: 1 Preemption is not helpful for scheduling, 3 No preemption victims found for incoming pod.

I have found at least two styles of taint that are not being respected :

  1. {node-role.kubernetes.io/control-plane: },
  2. {CriticalAddonsOnly: true}

This bug causes causes the deployment never to deploy - and within 5 minute it's closed automatically by the provider.

To Reproduce
Deploy to fill each node on a provider and you will get a bid from the tainted node.

Observered in the wild on validatornode.com and various other providers.

Expected behavior
Providers who have nodes that are tainted for no-deploy/noscheduling should not bid on workloads.

Additional context
Discussed on Sep 18th support call for more detail.

Metadata

Metadata

Labels

P1repo/providerAkash provider-services repo issues

Type

No type

Fields

No fields configured for issues without a type.

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions