AliyunContainerService gpushare-device-plugin issues

AliyunContainerService / gpushare-device-plugin

GPU Sharing Device Plugin for Kubernetes Cluster

Apache License 2.0

468 stars 144 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

ALIYUN_COM_GPU_MEM_IDX in the annotation is different than ALIYUN_COM_GPU_MEM_IDX inside the pod

#61 wokalski closed 11 months ago
1
Can it take effect on the window node?

#60 RowgerGo opened 1 year ago
0
节点有多个不同型号GPU(显存也不一致)时会以第一个识别到的GPU为准

#59 SakuraAxy opened 1 year ago
0
feat: upgrade go version && go mod

#58 swartz-k opened 1 year ago
1
ResourceExhausted desc = grpc: received message larger than max (4986010 vs. 4194304)

#57 k0nstantinv opened 1 year ago
0
trivy image scan lists critical and high vulnerability against latest image k8s-gpushare-plugin:v2-1.11-aff8a23

#56 carlwang87 opened 1 year ago
0
NVIDIA_VISIBLE_DEVICES wrong value in OCI spec

#55 k0nstantinv opened 1 year ago
0
该程序在 k8s .1.25中无法使用

#54 blankxyz opened 2 years ago
0
support for vgl

#52 umberto10 opened 2 years ago
0
containerd and nvidia-container-runtime instead of nvidia-docker2

#51 Frank-17 opened 2 years ago
2
set gpu index rather than gpu uuid for env NVIDIA_VISIBLE_DEVICES

#50 happy2048 closed 2 years ago
0
No Devices found. Waiting indefinitely.

#49 clennpillo opened 2 years ago
1
exit the device plugin when creating failed

#48 happy2048 closed 2 years ago
0
插件能获取GPU的个数，但是获取不了GPU的显存，共享无法调度

#47 gxwangit opened 3 years ago
2
skip to search pod when count of gpu devices is only one

#46 happy2048 closed 3 years ago
0
How to install on Mac?

#45 2811299 opened 3 years ago
0
fix kubelet pod annotation wrong

#44 RongbinZ opened 3 years ago
0
Question: request gpushare on the same GPU

#43 xhejtman opened 3 years ago
3
fix: modify the naming method to reduce the gRPC's request bytes

#42 qmloong closed 3 years ago
2
fix #39: get pods from kubelet client rather than list cluster-scope pods from apiserver

#41 qmloong closed 3 years ago
4
修复大量list all-namespaces pods的缺陷、MiB单位下名称长度可能引起grpc调用失败从而导致node的gpumem资源清0

#40 qmloong closed 3 years ago
2
集群内pod数量过多的情况有可能会引起集群高负载从而雪崩，另外MiB单位有可能会引起kubelet grpc单位失败

#39 qmloong closed 3 years ago
0
fix bug: update pod error when allocate

#38 tzzcfrank closed 3 years ago
0
PLEASE GIVE US A BINARY

#37 mariusehr1 closed 3 years ago
0
GPU device not detected with nvidia driver > 430.XX

#36 ptonelli closed 3 years ago
1
节点重启后，发现gpu显存超分了

#35 zlingqu opened 3 years ago
0
fix bug: display with multiple devices

#34 happy2048 closed 3 years ago
0
kubectl-inspect-gpushare supports gpushare2.0

#33 happy2048 closed 3 years ago
0
detect node label to disable cgpu

#32 happy2048 closed 4 years ago
0
Metrics

#31 dontmint closed 4 years ago
0
GPU registered to Kubelet but not available in `kubectl inspect gpushare` and not schedulable when using --memory-unit=Mib

#30 dontmint opened 4 years ago
0
Allocate 中getCandidatePods是否多此一举？

#29 malixian opened 4 years ago
0
Unable to schedule pod with: Insufficient aliyun.com/gpu-mem

#28 k0nstantinv closed 3 years ago
1
kubectl-inspect-gpushare: fatal with OpenID as auth-provider

#27 k0nstantinv opened 4 years ago
0
Concurrently create sharegpu instance will cause creation to fail

#26 guunergooner opened 4 years ago
0
some problem about auto allocate GPU card

#25 guobingithub opened 4 years ago
1
How to guarantee the pod does not use more memory than allocation?

#24 hyc3z opened 4 years ago
1
nvidia-container-cli: device error: unknown device id: no-gpu-has-256MiB-to-run\\\\n\\\"\"": unknown

#23 zhaogaolong opened 4 years ago
3
Update api version

#22 danieltanyouzu closed 4 years ago
1
Fix for kubernetes 1.17

#21 danieltanyouzu closed 4 years ago
0
update base image

#20 happy2048 closed 4 years ago
0
No assume timestamp for pod tf-jupyter-... so it's not GPUSharedAssumed assumed pod.

#19 jear closed 4 years ago
2
make missed error be handled

#18 mozhata closed 4 years ago
1
device plugin failed to detect gpu info correctly

#17 pan87232494 closed 4 years ago
7
[问题] Device Plugin allocate 选出的 pod 是否会跟 Kubelet 绑定的不一致

#16 orainxiong opened 5 years ago
4
自动给node添加gpuType标签

#15 lunar-knights opened 5 years ago
0
Add MPSUserGuide and integrate GPUShare with MPS

#14 Sakuralbj opened 5 years ago
0
Add env and volume information in MPS situation.

#13 Sakuralbj opened 5 years ago
1
question about pods with multiple containers

#12 monstercy opened 5 years ago
3
Can't pull your docker image

#11 morecoffee101 opened 5 years ago
1