smitmistry77 commented on issue #2486:
URL:
https://github.com/apache/apisix-ingress-controller/issues/2486#issuecomment-3103134926
Sure @juzhiyuan ,
Let me explain you my usecase: I want to divide the traffic based on the
lite customers (Consumers) and Premium Consumers.
So firstly via Admin API,
Step1: I am creating two consumer groups: LiteConsumerGroup and
PremiumConsumerGroup.
Step2: I am creating one consumer in lite group and associating the
LiteConsumerGroup Id to it, one for PremiumConsumer and associating the
PremiumConsumerGroup id to it.
Step3: I am create a route for a simple call to Upstream.
Step4: I am executing the API.
Below are the API Curls:
**Create lite Consumer Group:**
curl --request PUT \
--url https://localhost:9180/apisix/admin/consumer_groups/lite \
--header 'X-API-KEY: {{Admin_Apikey}}' \
--header 'content-type: application/json' \
--data '{
"plugins": {
"ai-proxy-multi": {
"fallback_strategy": "instance_health_and_rate_limiting",
"instances": [
{
"name": "gemma3-instance",
"provider": "openai-compatible",
"priority": 2,
"weight": 0,
"auth": {
"header": {
"Authorization": ""
}
},
"options": {
"model": "google/gemma-xxxx"
},
"override": {
"endpoint": "http://abcd/v1/chat/completions";
}
},
{
"name": "llama4-instance",
"provider": "openai-compatible",
"priority": 1,
"weight": 0,
"auth": {
"header": {
"Authorization": ""
}
},
"options": {
"model": "llama4"
},
"override": {
"endpoint": "http://xyz/v1/chat/completions";
}
}
]
},
"ai-rate-limiting": {
"instances": [
{
"name": "gemma3-instance",
"limit": 20,
"time_window": 60
}
],
"limit_strategy": "prompt_tokens"
},
"limit-count": {
"count": 1,
"time_window": 60,
"rejected_code": 429,
"key_type": "var",
"key": "consumer_name"
}
}
}'
**Create a Premium Consumer Group**
curl --request PUT \
--url https://localhost:9180/apisix/admin/consumer_groups/premium \
--header 'X-API-KEY: {{Admin_Apikey}}' \
--header 'content-type: application/json' \
--data '{
"plugins": {
"ai-proxy-multi": {
"fallback_strategy": "instance_health_and_rate_limiting",
"instances": [
{
"name": "gemma3-instance",
"provider": "openai-compatible",
"priority": 2,
"weight": 0,
"auth": {
"header": {
"Authorization": ""
}
},
"options": {
"model": "google/gemma-xxxx"
},
"override": {
"endpoint": "http://abcd/v1/chat/completions";
}
},
{
"name": "llama4-instance",
"provider": "openai-compatible",
"priority": 1,
"weight": 0,
"auth": {
"header": {
"Authorization": ""
}
},
"options": {
"model": "meta-llama/Llama-xxxx"
},
"override": {
"endpoint": "http://xyz/v1/chat/completions";
}
}
]
},
"ai-rate-limiting": {
"instances": [
{
"name": "gemma3-instance",
"limit": 30,
"time_window": 60
}
],
"limit_strategy": "prompt_tokens"
},
"limit-count": {
"count": 3,
"time_window": 60,
"rejected_code": 429,
"key_type": "var",
"key": "consumer_name"
}
}
}'
**Lite Consumer**
curl --request PUT \
--url https://localhost:9180/apisix/admin/consumers \
--header 'X-API-KEY: {{Admin_Apikey}}' \
--header 'content-type: application/json' \
--data '{
"username": "consumer-lite",
"plugins": {
"key-auth": {
"key": "xxxxx"
}
},
"group_id":"lite"
}'
**Premium Consumer**
curl --request PUT \
--url https://localhost:9180/apisix/admin/consumers \
--header 'X-API-KEY: {{Admin_Apikey}}' \
--header 'content-type: application/json' \
--data '{
"username": "consumer-premium",
"plugins": {
"key-auth": {
"key": "xxxx"
}
},
"group_id":"premium"
}'
**Create Route**
curl --request PUT \
--url https://localhost:9180/apisix/admin/routes \
--header 'X-API-KEY: {{Admin_Apikey}}' \
--header 'content-type: application/json' \
--data '{
"id": "route-consumer-category",
"uri": "/consumercategory",
"methods": [
"POST"
],
"plugins": {
"key-auth": {}
}
}'
**Execute Req with Lite Consumer**
curl --request POST \
--url https://localhost:9180/consumercategory \
--header 'apikey: lite' \
--header 'content-type: application/json' \
--data '{
"model": "google/gemma-xxxx",
"messages": [
{
"role": "user",
"content": "What is APISIX"
}
]
}'
**Execute Req with Premium Consumer**
curl --request POST \
--url https://localhost:9180/consumercategory \
--header 'apikey: lite' \
--header 'content-type: application/json' \
--data '{
"model": "google/gemma-xxxx",
"messages": [
{
"role": "user",
"content": "What is APISIX"
}
]
}'
Now my question is, can I do the above setup in ADC format with the
apisix-ingress-controller 2.0.0-rc2 image?
If yes, how? and if no, then why?
In the github repo where the CRD yaml file is present, I don't see the
Consumer_Groups CRD. correct me here if I am wrong.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscr...@apisix.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
