绵阳网站,服务器的作用和用途,事业单位门户网站建设的建议,巫山网站开发0.问题说明
想要让RAGFLOW利用GPU资源跑起来#xff0c;可以选择docker-compose-gpu.yml启动。#xff08;但是官网启动案例是86平台的不是NVIDIA GPU的#xff0c;docker-compose-gpu.yml又是第三方维护#xff0c;所以稍有问题#xff09;
1.问题
docker利用docker-c…0.问题说明
想要让RAGFLOW利用GPU资源跑起来可以选择docker-compose-gpu.yml启动。但是官网启动案例是×86平台的不是NVIDIA GPU的docker-compose-gpu.yml又是第三方维护所以稍有问题
1.问题
docker利用docker-compose-gpu.yml启动RAGFLOW文档解析出错 报错 18:10:23 [ERROR][Exception]: NCCL Error 2: unhandled system error (run with NCCL_DEBUGINFO for details) 2.解决方案
1修改docker-compose-gpu.yml文件稍作改动
下面是修改后的完整docker-compose-gpu.yml文件可以直接复制。
# The RAGFlow team do not actively maintain docker-compose-gpu.yml, so use them at your own risk.
# However, you are welcome to file a pull request to improve it.
include:- ./docker-compose-base.ymlservices:ragflow:depends_on:mysql:condition: service_healthyimage: ${RAGFLOW_IMAGE}container_name: ragflow-serverports:- ${SVR_HTTP_PORT}:9380- 80:80- 443:443volumes:- ./ragflow-logs:/ragflow/logs- ./nginx/ragflow.conf:/etc/nginx/conf.d/ragflow.conf- ./nginx/proxy.conf:/etc/nginx/proxy.conf- ./nginx/nginx.conf:/etc/nginx/nginx.confenv_file: .envipc: hostshm_size: 8genvironment:- TZ${TIMEZONE}- HF_ENDPOINT${HF_ENDPOINT}- MACOS${MACOS}- NCCL_DEBUGINFOnetworks:- ragflowrestart: on-failure# https://docs.docker.com/engine/daemon/prometheus/#create-a-prometheus-configuration# If youre using Docker Desktop, the --add-host flag is optional. This flag makes sure that the hosts internal IP gets exposed to the Prometheus container.extra_hosts:- host.docker.internal:host-gatewaydeploy:resources:reservations:devices:- driver: nvidiacount: allcapabilities: [gpu]参数解释 ipc: host允许容器共享主机的IPC命名空间解决NCCL多卡通信问题 shm_size: 8g增大共享内存容量默认64MB不足
2通过docker-compose-gpu.yml重新启动RAGFLOW
docker compose -f docker-compose-gpu.yml up -d3运行ragflow-server服务器
docker logs -f ragflow-server4检查是否成功进行文档解析
成功解析如下结果 到此问题解决
