garyyang85

比如： db.num=3 db.url.0=jdbc:mysql://xxx/nacos_config?useUnicode=true&characterEncoding=utf8&useAffectedRows=true&allowMultiQueries=true db.url.1=jdbc:mysql://xxx2/nacos_config?useUnicode=true&characterEncoding=utf8&useAffectedRows=true&allowMultiQueries=true db.url.2=jdbc:mysql://xxx3/nacos_config?useUnicode=true&characterEncoding=utf8&useAffectedRows=true&allowMultiQueries=true

Support folder and dashboard permissions

1

Hi, it there a plan that we can support folder and dashboard permissions update? Thanks

Do we have any plan to deploy spring cloud alibaba components to kubernetes by chart or operator

2

**Which Component** All **Describe what problem you have encountered** Do we have any plan to deploy spring cloud alibaba components to kubernetes by chart or operator? **Describe what information you...

No devices were found in openshift

2

### 1. Quick Debug Information * OS/Version Red Hat Enterprise Linux CoreOS release 4.12 * Kernel Version: 4.18.0-372.69.1.el8_6.x86_64 * Container Runtime Type/Version: CRI-O * Openshift 4.12.29 * GPU Operator Version:...

Use FastChat to start the deepseek-coder-33b-instruct model, send a stream request and got an error response. If set stream=False, you can print a good response If change to other models,...

[Usage]: Not enough memory when run a 33b model float16 on 2 x L40 GPU (48G)

3

### Your current environment Not enough memory when run deepseek-coder-33b-instruct model on two L40 GPUs (48G) with vllm 0.4.1. Also tried --enforce-eager --gpu-memory-utilization, did not help. ``` python -m vllm.entrypoints.openai.api_server...

两块卡就没问题了，还有什么优化方法可以让他正常的跑在一块v100上么？跑两块v100有点浪费了

I would like to try 'top_k' in my scenario to see if anything different in the output, but seems the openai invoke method does not support this parameter. https://platform.openai.com/docs/api-reference/chat/create I...

[Usage]: How to increase the context length when start with vllm.entrypoints.openai.api_server

3

### Your current environment I tried deepseek-coder-v2-lite-instruct can be started on 2 x L40 GPU with vllm 0.5.1，but the context cannot reach 128K, only 9415 tokens in my test. Below...

### What is the issue? Using gemma2-27b for test, after ollama create -q Q8_0, the quality of answer is not very good. The accuracy seems more worse than origin gemma2-9b...