暗能星系

zhanglu

Type Reason Age From Message

Normal NodeReady 47m (x11 over 6h56m) kubelet Node node1 status is now: NodeReady
Normal NodeNotReady 44m (x12 over 7h3m) kubelet Node node1 status is now: NodeNotReady
Normal Starting 37m kubelet Starting kubelet.
Normal NodeHasSufficientMemory 37m kubelet Node node1 status is now: NodeHasSufficientMemory
Normal NodeHasNoDiskPressure 37m kubelet Node node1 status is now: NodeHasNoDiskPressure
Normal NodeHasSufficientPID 37m kubelet Node node1 status is now: NodeHasSufficientPID
Normal NodeAllocatableEnforced 37m kubelet Updated Node Allocatable limit across pods
Normal NodeReady 37m kubelet Node node1 status is now: NodeReady
Normal NodeNotReady 34m kubelet Node node1 status is now: NodeNotReady
Normal Starting 31m kubelet Starting kubelet.
Normal NodeHasSufficientMemory 31m kubelet Node node1 status is now: NodeHasSufficientMemory
Normal NodeHasNoDiskPressure 31m kubelet Node node1 status is now: NodeHasNoDiskPressure
Normal NodeHasSufficientPID 31m kubelet Node node1 status is now: NodeHasSufficientPID
Normal NodeAllocatableEnforced 31m kubelet Updated Node Allocatable limit across pods
Normal NodeReady 9m29s (x2 over 31m) kubelet Node node1 status is now: NodeReady
Normal NodeNotReady 6m28s (x2 over 28m) kubelet Node node1 status is now: NodeNotReady

zhanglu

安装NVIDIA仓库配置包（适用于CentOS 8）

distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.repo | sudo tee /etc/yum.repos.d/nvidia-docker.repo

sudo dnf install -y nvidia-container-toolkit

zhanglu

nvidia-ctk runtime configure --runtime=docker

zhanglu

kubectl get pods --all-namespaces -o json | jq -r '.items[] | select(.spec.containers[].resources.limits["<gpu-resource-name>"] != null) | .metadata.namespace + " " + .metadata.name'

zhanglu

2025/10/13 02:31:43 Starting FS watcher.
2025/10/13 02:31:43 Starting OS watcher.
2025/10/13 02:31:43 Starting Plugins.
2025/10/13 02:31:43 Loading configuration.
2025/10/13 02:31:43 Initializing NVML.
2025/10/13 02:31:43 Failed to initialize NVML: could not load NVML library.
2025/10/13 02:31:43 If this is a GPU node, did you set the docker default runtime to nvidia?
2025/10/13 02:31:43 You can check the prerequisites at: https://github.com/NVIDIA/k8s-device-plugin#prerequisites
2025/10/13 02:31:43 You can learn how to set the runtime at: https://github.com/NVIDIA/k8s-device-plugin#quick-start
2025/10/13 02:31:43 If this is not a GPU node, you should set up a toleration or nodeSelector to only deploy this plugin on GPU nodes

zhanglu

https://cn.download.nvidia.com/XFree86/Linux-x86_64/580.95.05/NVIDIA-Linux-x86_64-580.95.05.run

zhanglu

2025/10/13 07:23:31 Failed to initialize NVML: could not load NVML library.
2025/10/13 07:23:31 If this is a GPU node, did you set the docker default runtime to nvidia?

zhanglu

wget https://developer.download.nvidia.com/compute/cuda/13.0.2/local_installers/cuda_13.0.2_580.95.05_linux.run

zhanglu

wget https://developer.download.nvidia.com/compute/cuda/11.6.2/local_installers/cuda_11.6.2_510.47.03_linux.run

zhanglu

Failed to initialize NVML: could not load NVML library

zhanglu

https://raw.githubusercontent.com/NVIDIA/k8s-device-plugin/v0.17.1/deployments/static/nvidia-device-plugin.yml

zhanglu

https://www.nvidia.cn/drivers/details/252785/

zhanglu

$ curl -s -L https://nvidia.github.io/nvidia-docker/centos8/nvidia-docker.repo | sudo tee /etc/yum.repos.d/nvidia-docker.repo
yum install -y nvidia-container-toolkit

zhanglu

d?????????? ? ? ? ? ? cephfs_data

zhanglu

root
Tzzs@2025*
220.185.228.106
30001

zhanglu

http://192.168.30.202:31237/api/workflows/v1/8a6aa1f8-0ab1-4518-bc22-06390d1c7494/abort

zhanglu

curl -X POST "http://192.168.30.202:31237/api/workflows/v1/8a6aa1f8-0ab1-4518-bc22-06390d1c7494/abort" -H "accept: application/json"

zhanglu

zhanglu

import json
import argparse
import requests
from requests.exceptions import RequestException # 处理网络请求异常

def fetch_json_from_url(url, timeout=10):
"""通过GET请求从HTTP接口获取JSON数据并返回解析后的字典"""
try:
# 发送GET请求，设置超时时间
response = requests.get(url, timeout=timeout)

    # 检查HTTP响应状态码（200表示成功）
    response.raise_for_status()
    
    # 解析JSON响应内容
    return response.json()

except requests.exceptions.ConnectTimeout:
    raise ValueError(f"错误：连接接口 '{url}' 超时（{timeout}秒），请检查网络或接口可用性")
except requests.exceptions.ConnectionError:
    raise ValueError(f"错误：无法连接到接口 '{url}'，请检查URL是否正确或接口是否在线")
except requests.exceptions.HTTPError as e:
    # 捕获404、500等HTTP错误状态码
    raise ValueError(f"错误：接口返回 HTTP {response.status_code} 错误，详情：{str(e)}")
except json.JSONDecodeError:
    raise ValueError(f"错误：接口返回的数据不是有效的JSON格式，请检查接口响应")
except RequestException as e:
    # 捕获其他网络请求异常
    raise ValueError(f"请求接口时发生错误：{str(e)}")

def parse_running_task_ids(json_data):
"""从JSON数据中提取状态为'Running'的任务ID，返回ID列表"""
if "results" not in json_data:
return [] # 无results字段时返回空列表

running_ids = []
for task in json_data["results"]:
    # 仅收集状态为"Running"的任务ID
    if task.get("status") == "Running":
        running_ids.append(task.get("id", "N/A"))
return running_ids

def generate_abort_commands(task_ids):
"""根据任务ID生成中止任务的curl命令并打印"""
if not task_ids:
print("未找到状态为'Running'的任务，无需生成中止命令")
return

print("生成的中止任务命令：")
for task_id in task_ids:
    if task_id != "N/A":
        print(f'curl -X POST "http://192.168.30.202:31237/api/workflows/v1/{task_id}/abort" -H "accept: application/json"')
    else:
        print(f"警告：存在状态为'Running'的任务，但未获取到有效ID，跳过")

def main():
# 解析命令行参数（指定HTTP接口URL）
parser = argparse.ArgumentParser(description="从HTTP接口获取任务信息，生成Running状态任务的中止命令")
parser.add_argument("url", help="JSON数据所在的HTTP接口URL（例如：http://example.com/tasks）")
parser.add_argument("--timeout", type=int, default=10, help="请求超时时间（秒），默认10秒")
args = parser.parse_args()

try:
    # 从接口获取并解析JSON数据
    json_data = fetch_json_from_url(args.url, timeout=args.timeout)
    # 提取Running状态的任务ID
    running_task_ids = parse_running_task_ids(json_data)
    # 生成并打印中止命令
    generate_abort_commands(running_task_ids)
except ValueError as e:
    print(f"处理失败：{e}")
    exit(1)

if name == "main":
main()

zhanglu

-o Report/sample_info.xlsx`
################# retry : Some(9977) ################################## retry : Some(9933) ################################## retry : Some(9921) ################################## retry : Some(9992) ################################## retry : Some(9919) ################################## retry : Some(9921) ################################## retry : Some(9952) ################################## retry : Some(9867) ################################## retry : Some(9978) ################################## retry : Some(9962) ################################## retry : Some(9958) ################################## retry : Some(9948) ################################## retry : Some(9839) ################################## retry : Some(9994) ################################## retry : Some(9959) ################################## retry : Some(9964) ################################## retry : Some(9985) ################################## retry : Some(9953) ################################## retry : Some(9899) ################################## retry : Some(9990) ################################## re

tmp

安装NVIDIA仓库配置包（适用于CentOS 8）