Skip to content

Fix timeouts #66

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
nicosuter merged 2 commits into from
Apr 15, 2026
Merged

Fix timeouts #66

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6e0a1fe
add instance type for debugging
nicosuter Apr 11, 2026
7ea7544
increase timeouts
nicosuter Apr 15, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
36 changes: 32 additions & 4 deletions modules/aws/eks-addons/karpenter.tf
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -116,7 +116,7 @@ resource "kubectl_manifest" "nodepool_gpus" {
taints = [{ key = "nvidia.com/gpu", value = "true", effect = "NoSchedule" }]
}
}
limits = { "nvidia.com/gpu" = "8" }
limits = { "nvidia.com/gpu" = "32" }
disruption = { consolidationPolicy = "WhenEmpty", consolidateAfter = "5m" }
}
})
Expand Down Expand Up @@ -144,7 +144,7 @@ resource "kubectl_manifest" "nodepool_gpum" {
taints = [{ key = "nvidia.com/gpu", value = "true", effect = "NoSchedule" }]
}
}
limits = { "nvidia.com/gpu" = "8" }
limits = { "nvidia.com/gpu" = "32" }
disruption = { consolidationPolicy = "WhenEmpty", consolidateAfter = "5m" }
}
})
Expand Down Expand Up @@ -172,7 +172,7 @@ resource "kubectl_manifest" "nodepool_gpul" {
taints = [{ key = "nvidia.com/gpu", value = "true", effect = "NoSchedule" }]
}
}
limits = { "nvidia.com/gpu" = "8" }
limits = { "nvidia.com/gpu" = "32" }
disruption = { consolidationPolicy = "WhenEmpty", consolidateAfter = "5m" }
}
})
Expand Down Expand Up @@ -200,7 +200,7 @@ resource "kubectl_manifest" "nodepool_h100" {
taints = [{ key = "nvidia.com/gpu", value = "true", effect = "NoSchedule" }]
}
}
limits = { "nvidia.com/gpu" = "8" }
limits = { "nvidia.com/gpu" = "32" }
disruption = { consolidationPolicy = "WhenEmpty", consolidateAfter = "5m" }
}
})
Expand Down Expand Up @@ -261,5 +261,33 @@ resource "kubectl_manifest" "nodepool_a100_80" {
}
})

depends_on = [kubectl_manifest.karpenter_node_class]
}


resource "kubectl_manifest" "nodepool_g6e8xlarge" {
yaml_body = yamlencode({
apiVersion = "karpenter.sh/v1"
kind = "NodePool"
metadata = { name = "g6e8xlarge" }
spec = {
template = {
metadata = { labels = { "node-tier" = "g6e8xlarge" } }
spec = {
nodeClassRef = { group = "karpenter.k8s.aws", kind = "EC2NodeClass", name = "default" }
expireAfter = var.gpu_node_max_lifetime
requirements = [
{ key = "karpenter.sh/capacity-type", operator = "In", values = ["on-demand"] },
{ key = "node.kubernetes.io/instance-type", operator = "In", values = ["g6e.8xlarge"] },
{ key = "kubernetes.io/arch", operator = "In", values = ["amd64"] },
]
taints = [{ key = "nvidia.com/gpu", value = "true", effect = "NoSchedule" }]
}
}
limits = { "nvidia.com/gpu" = "32" }
disruption = { consolidationPolicy = "WhenEmpty", consolidateAfter = "5m" }
}
})

depends_on = [kubectl_manifest.karpenter_node_class]
}
2 changes: 1 addition & 1 deletion variables.tf
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -128,7 +128,7 @@ variable "s3_bucket_arns" {
variable "gpu_node_max_lifetime" {
description = "Hard TTL for gpus/gpum/gpul nodes. Karpenter drains and terminates any node running longer than this duration, regardless of workload state. Go duration syntax (e.g. \"24h\", \"72h\"). Set to \"Never\" to disable."
type = string
default = "16h"
default = "48h"
}

variable "ml_data_bucket_name" {
Expand Down
Loading