ES

Results 8 comments of ES

算法都是假需求,背就好了。。

> 通过哪种方式安装并接入的? 通过 docker 启动 rainbond 的 allinone 镜像之后,选择 平台管理 > 集群 -> 添加集群 -> 接入自建集群,然后按照 接入自建集群 的指示操作完成安装之后,在 rainbond 的集群接入等待页面依旧提示错误,查看 rainbond 的日志发现如问题所示的错误

> 看这里 `error execution phase kubelet-start: a Node with name "lvscare.node.ip" and status "Ready" already exists in the cluster. You must delete the existing Node or change the name of...

这是完整日志 ```shell (venv) root@scene_llm4:projects/xinference # xinference-local --log-level DEBUG --host 0.0.0.0 --port 9997 INFO 11-07 14:17:05 [__init__.py:235] Automatically detected platform cuda. Detected accelerate version: 1.3.0 Detected bitsandbytes version: 0.45.0 Detected coloredlogs...

新起一个环境安装可以用,是不是哪没更新到

> 我的命令是这样的: docker run --hostname localhost --log-opt max-size=100m --log-opt max-file=2 -v /data/llm:/root/Ilm -e XINFERENCE_HOME=/root/llm -p 9997:9997 --gpus all --shm-size 30g -d --restart=always --privileged xinference_1_12_0:v1 /root/anaconda3/envs/xinference/bin/xinference-local -H 0.0.0.0 我是通过重新创建python虚拟环境,重新安装,从命令行启动的,数据应该是在默认路径的,我没动。docker我就不清楚了。

> I encountered the same problem. May I ask if you have solved it? Use tag v1.0.1, and [install ](https://github.com/OpenDriveLab/UniAD/tree/v1.0.1?tab=readme-ov-file)

> # Thank you very much for your answer. This issue still occurs when the version is reduced to 1.0.1. There is no such error when using version 0.11.4, but...