4paradigm / k8s-vgpu-scheduler

OpenAIOS vGPU device plugin for Kubernetes is originated from the OpenAIOS project to virtualize GPU device memory, in order to allow applications to access larger memory space than its physical capacity. It is designed for ease of use of extended device memory for AI workloads.
Apache License 2.0
526 stars 93 forks source link

vgpu-scheduler container kube-scheduler error #43

Open G-Kolls opened 3 months ago

G-Kolls commented 3 months ago

I0816 05:35:33.465569 1 flags.go:64] FLAG: --leader-elect-resource-namespace="kube-system" I0816 05:35:33.465572 1 flags.go:64] FLAG: --leader-elect-retry-period="2s" I0816 05:35:33.465576 1 flags.go:64] FLAG: --log-flush-frequency="5s" I0816 05:35:33.465580 1 flags.go:64] FLAG: --log-json-info-buffer-size="0" I0816 05:35:33.465591 1 flags.go:64] FLAG: --log-json-split-stream="false" I0816 05:35:33.465595 1 flags.go:64] FLAG: --logging-format="text" I0816 05:35:33.465598 1 flags.go:64] FLAG: --master="" I0816 05:35:33.465602 1 flags.go:64] FLAG: --permit-address-sharing="false" I0816 05:35:33.465605 1 flags.go:64] FLAG: --permit-port-sharing="false" I0816 05:35:33.465609 1 flags.go:64] FLAG: --pod-max-in-unschedulable-pods-duration="5m0s" I0816 05:35:33.465612 1 flags.go:64] FLAG: --profiling="true" I0816 05:35:33.465615 1 flags.go:64] FLAG: --requestheader-allowed-names="[]" I0816 05:35:33.465622 1 flags.go:64] FLAG: --requestheader-client-ca-file="" I0816 05:35:33.465625 1 flags.go:64] FLAG: --requestheader-extra-headers-prefix="[x-remote-extra-]" I0816 05:35:33.465635 1 flags.go:64] FLAG: --requestheader-group-headers="[x-remote-group]" I0816 05:35:33.465643 1 flags.go:64] FLAG: --requestheader-username-headers="[x-remote-user]" I0816 05:35:33.465649 1 flags.go:64] FLAG: --secure-port="10259" I0816 05:35:33.465653 1 flags.go:64] FLAG: --show-hidden-metrics-for-version="" I0816 05:35:33.465657 1 flags.go:64] FLAG: --tls-cert-file="" I0816 05:35:33.465661 1 flags.go:64] FLAG: --tls-cipher-suites="[]" I0816 05:35:33.465669 1 flags.go:64] FLAG: --tls-min-version="" I0816 05:35:33.465673 1 flags.go:64] FLAG: --tls-private-key-file="" I0816 05:35:33.465679 1 flags.go:64] FLAG: --tls-sni-cert-key="[]" I0816 05:35:33.465690 1 flags.go:64] FLAG: --v="4" I0816 05:35:33.465701 1 flags.go:64] FLAG: --version="false" I0816 05:35:33.465709 1 flags.go:64] FLAG: --vmodule="" I0816 05:35:33.465714 1 flags.go:64] FLAG: --write-config-to="" I0816 05:35:33.917294 1 serving.go:348] Generated self-signed cert in-memory E0816 05:35:33.917711 1 run.go:74] "command failed" err="no kind \"KubeSchedulerConfiguration\" is registered for version \"kubescheduler.config.k8s.io/v1beta2\" in scheme \"pkg/scheduler/apis/config/scheme/scheme.go:30\""

lzh-1625 commented 3 months ago

Same problem, have you solved it yet?

lzh-1625 commented 3 months ago

I solved it,you can try the new project https://github.com/Project-HAMi/HAMi?tab=readme-ov-file