mirror of
https://github.com/katanemo/plano.git
synced 2026-06-17 15:25:17 +02:00
Add prem support for a2a agents
This commit is contained in:
Adil Hafeez
parent
2e346143dd
commit
299f183e66
23 changed files with 2544 additions and 16 deletions
|
|
@ -4,8 +4,7 @@ RUN rustup -v target add wasm32-wasip1
|
|||
WORKDIR /arch
|
||||
COPY crates .
|
||||
|
||||
RUN cd prompt_gateway && cargo build --release --target wasm32-wasip1
|
||||
RUN cd llm_gateway && cargo build --release --target wasm32-wasip1
|
||||
RUN cargo build --release --target wasm32-wasip1
|
||||
|
||||
# copy built filter into envoy image
|
||||
FROM docker.io/envoyproxy/envoy:v1.32-latest as envoy
|
||||
|
|
@ -17,6 +16,7 @@ RUN apt-get update && apt-get install -y gettext-base curl && apt-get clean && r
|
|||
|
||||
COPY --from=builder /arch/target/wasm32-wasip1/release/prompt_gateway.wasm /etc/envoy/proxy-wasm-plugins/prompt_gateway.wasm
|
||||
COPY --from=builder /arch/target/wasm32-wasip1/release/llm_gateway.wasm /etc/envoy/proxy-wasm-plugins/llm_gateway.wasm
|
||||
COPY --from=builder /arch/target/wasm32-wasip1/release/agent_gateway.wasm /etc/envoy/proxy-wasm-plugins/agent_gateway.wasm
|
||||
COPY --from=envoy /usr/local/bin/envoy /usr/local/bin/envoy
|
||||
WORKDIR /app
|
||||
COPY arch/requirements.txt .
|
||||
|
|
@ -29,4 +29,4 @@ RUN pip install requests
|
|||
RUN touch /var/log/envoy.log
|
||||
|
||||
# ENTRYPOINT ["sh","-c", "python config_generator.py && envsubst < /etc/envoy/envoy.yaml > /etc/envoy.env_sub.yaml && envoy -c /etc/envoy.env_sub.yaml --log-level trace 2>&1 | tee /var/log/envoy.log"]
|
||||
ENTRYPOINT ["sh","-c", "python config_generator.py && envsubst < /etc/envoy/envoy.yaml > /etc/envoy.env_sub.yaml && envoy -c /etc/envoy.env_sub.yaml --component-log-level wasm:info 2>&1 | tee /var/log/envoy.log"]
|
||||
ENTRYPOINT ["sh","-c", "python config_generator.py && envsubst < /etc/envoy/envoy.yaml > /etc/envoy.env_sub.yaml && envoy -c /etc/envoy.env_sub.yaml --component-log-level wasm:debug 2>&1 | tee /var/log/envoy.log"]
|
||||
|
|
|
|||
|
|
@ -3,6 +3,10 @@ type: object
|
|||
properties:
|
||||
version:
|
||||
type: string
|
||||
tools:
|
||||
type: object
|
||||
agents:
|
||||
type: object
|
||||
listeners:
|
||||
type: object
|
||||
additionalProperties: false
|
||||
|
|
|
|||
|
|
@ -452,6 +452,125 @@ static_resources:
|
|||
typed_config:
|
||||
"@type": type.googleapis.com/envoy.extensions.filters.http.router.v3.Router
|
||||
|
||||
- name: ingress_traffic_agent
|
||||
address:
|
||||
socket_address:
|
||||
address: 0.0.0.0
|
||||
port_value: 14000
|
||||
traffic_direction: INBOUND
|
||||
filter_chains:
|
||||
- filters:
|
||||
- name: envoy.filters.network.http_connection_manager
|
||||
typed_config:
|
||||
"@type": type.googleapis.com/envoy.extensions.filters.network.http_connection_manager.v3.HttpConnectionManager
|
||||
{% if "random_sampling" in arch_tracing and arch_tracing["random_sampling"] > 0 %}
|
||||
generate_request_id: true
|
||||
tracing:
|
||||
provider:
|
||||
name: envoy.tracers.opentelemetry
|
||||
typed_config:
|
||||
"@type": type.googleapis.com/envoy.config.trace.v3.OpenTelemetryConfig
|
||||
grpc_service:
|
||||
envoy_grpc:
|
||||
cluster_name: opentelemetry_collector
|
||||
timeout: 0.250s
|
||||
service_name: ingress_traffic_agent
|
||||
random_sampling:
|
||||
value: {{ arch_tracing.random_sampling }}
|
||||
{% endif %}
|
||||
stat_prefix: ingress_traffic_agent
|
||||
codec_type: AUTO
|
||||
scheme_header_transformation:
|
||||
scheme_to_overwrite: https
|
||||
access_log:
|
||||
- name: envoy.access_loggers.file
|
||||
typed_config:
|
||||
"@type": type.googleapis.com/envoy.extensions.access_loggers.file.v3.FileAccessLog
|
||||
path: "/var/log/access_ingress_agent.log"
|
||||
route_config:
|
||||
name: local_routes
|
||||
virtual_hosts:
|
||||
- name: local_service
|
||||
domains:
|
||||
- "*"
|
||||
routes:
|
||||
{% for provider in arch_llm_providers %}
|
||||
# if endpoint is set then use custom cluster for upstream llm
|
||||
{% if provider.endpoint %}
|
||||
{% set llm_cluster_name = provider.name %}
|
||||
{% else %}
|
||||
{% set llm_cluster_name = provider.provider_interface %}
|
||||
{% endif %}
|
||||
- match:
|
||||
prefix: "/"
|
||||
headers:
|
||||
- name: "x-arch-llm-provider"
|
||||
string_match:
|
||||
exact: {{ llm_cluster_name }}
|
||||
route:
|
||||
auto_host_rewrite: true
|
||||
cluster: {{ llm_cluster_name }}
|
||||
timeout: 60s
|
||||
{% endfor %}
|
||||
http_filters:
|
||||
- name: envoy.filters.http.compressor
|
||||
typed_config:
|
||||
"@type": type.googleapis.com/envoy.extensions.filters.http.compressor.v3.Compressor
|
||||
compressor_library:
|
||||
name: compress
|
||||
typed_config:
|
||||
"@type": type.googleapis.com/envoy.extensions.compression.gzip.compressor.v3.Gzip
|
||||
memory_level: 3
|
||||
window_bits: 10
|
||||
- name: envoy.filters.http.wasm_agent
|
||||
typed_config:
|
||||
"@type": type.googleapis.com/udpa.type.v1.TypedStruct
|
||||
type_url: type.googleapis.com/envoy.extensions.filters.http.wasm.v3.Wasm
|
||||
value:
|
||||
config:
|
||||
name: "http_config"
|
||||
root_id: agent_gateway
|
||||
configuration:
|
||||
"@type": "type.googleapis.com/google.protobuf.StringValue"
|
||||
value: |
|
||||
{{ arch_config | indent(32) }}
|
||||
vm_config:
|
||||
runtime: "envoy.wasm.runtime.v8"
|
||||
code:
|
||||
local:
|
||||
filename: "/etc/envoy/proxy-wasm-plugins/agent_gateway.wasm"
|
||||
- name: envoy.filters.http.wasm_llm
|
||||
typed_config:
|
||||
"@type": type.googleapis.com/udpa.type.v1.TypedStruct
|
||||
type_url: type.googleapis.com/envoy.extensions.filters.http.wasm.v3.Wasm
|
||||
value:
|
||||
config:
|
||||
name: "http_config"
|
||||
root_id: llm_gateway
|
||||
configuration:
|
||||
"@type": "type.googleapis.com/google.protobuf.StringValue"
|
||||
value: |
|
||||
{{ arch_llm_config | indent(32) }}
|
||||
vm_config:
|
||||
runtime: "envoy.wasm.runtime.v8"
|
||||
code:
|
||||
local:
|
||||
filename: "/etc/envoy/proxy-wasm-plugins/llm_gateway.wasm"
|
||||
- name: envoy.filters.http.decompressor
|
||||
typed_config:
|
||||
"@type": type.googleapis.com/envoy.extensions.filters.http.decompressor.v3.Decompressor
|
||||
decompressor_library:
|
||||
name: decompress
|
||||
typed_config:
|
||||
"@type": "type.googleapis.com/envoy.extensions.compression.gzip.decompressor.v3.Gzip"
|
||||
window_bits: 9
|
||||
chunk_size: 8192
|
||||
# If this ratio is set too low, then body data will not be decompressed completely.
|
||||
max_inflate_ratio: 1000
|
||||
- name: envoy.filters.http.router
|
||||
typed_config:
|
||||
"@type": type.googleapis.com/envoy.extensions.filters.http.router.v3.Router
|
||||
|
||||
clusters:
|
||||
- name: openai
|
||||
connect_timeout: 0.5s
|
||||
|
|
|
|||
|
|
@ -49,6 +49,7 @@ def docker_start_archgw_detached(
|
|||
port_mappings = [
|
||||
f"{prompt_gateway_port}:{prompt_gateway_port}",
|
||||
f"{llm_gateway_port}:{llm_gateway_port}",
|
||||
"14000:14000",
|
||||
"9901:19901",
|
||||
]
|
||||
port_mappings_args = [item for port in port_mappings for item in ("-p", port)]
|
||||
|
|
@ -56,7 +57,7 @@ def docker_start_archgw_detached(
|
|||
volume_mappings = [
|
||||
f"{logs_path_abs}:/var/log:rw",
|
||||
f"{arch_config_file}:/app/arch_config.yaml:ro",
|
||||
# "/Users/adilhafeez/src/intelligent-prompt-gateway/crates/target/wasm32-wasip1/release:/etc/envoy/proxy-wasm-plugins:ro",
|
||||
"/Users/adilhafeez/src/intelligent-prompt-gateway/crates/target/wasm32-wasip1/release:/etc/envoy/proxy-wasm-plugins:ro",
|
||||
]
|
||||
volume_mappings_args = [
|
||||
item for volume in volume_mappings for item in ("-v", volume)
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue