d3204b82f0
A Bedrock provider could never answer a discovery request. The router
routed GET /inference-profiles to the record's upstream, which has to be
bedrock-runtime.<region> for InvokeModel to work, and that host does not
implement the operation — AWS answers <UnknownOperationException/>.
ListInferenceProfiles lives on the control plane at bedrock.<region>.
Give the route a discovery host, taken from the catalog's declaration
with the region read back out of the configured upstream, and send the
listing — and only the listing — there. Inference is untouched, and a
proxied or self-hosted Bedrock endpoint gets no discovery host at all
rather than a guessed one, since inventing a host would send the
operator's credential somewhere they never configured.
Two things had to follow for the listing to be usable once it arrives.
The response filter only understood OpenAI's {data:[{id:…}]}, so a
Bedrock listing was forwarded whole — offering every profile in the
account whatever the policy said. It now recognises the
inferenceProfileSummaries envelope, and matches a listing id against the
record's models after stripping the region prefix and version suffix, so
the two spellings of one model line up.
The policy bound had the same problem from the other side: it intersected
by exact string, so a record registering the raw profile id while a
guardrail names the catalog key intersected to nothing and would have
bounded a working provider's listing down to empty. routeClaimsModel
already normalises the candidate for this reason; the bound now agrees
with it.
The live discovery e2e flips from asserting the 404 to asserting a real
filtered listing. The mock upstream cannot cover any of this: it answers
/inference-profiles on the same listener as everything else, so a
mock-based test passes whichever host the request went to.
netbird Management Server
netbird management server will control and synchronize peers configuration within your Netbird account and network.
Command Options
The CLI accepts the command management with the following options:
start Netbird Management Server
Usage:
netbird-mgmt management [flags]
Flags:
--cert-file string Location of your SSL certificate. Can be used when you have an existing certificate and don't want a new certificate be generated automatically. If letsencrypt-domain is specified this property has no effect
--cert-key string Location of your SSL certificate private key. Can be used when you have an existing certificate and don't want a new certificate be generated automatically. If letsencrypt-domain is specified this property has no effect
--datadir string server data directory location
-h, --help help for management
--letsencrypt-domain string a domain to issue Let's Encrypt certificate for. Enables TLS using Let's Encrypt. Will fetch and renew certificate, and run the server with TLS
--port int server port to listen on (default 33073)
Global Flags:
--config string Netbird config file location to write new config to (default "/etc/netbird")
--log-file string sets Netbird log path. If console is specified the the log will be output to stdout (default "/var/log/netbird/management.log")
--log-level string (default "info")
Run Management service (Docker)
You can run service in 2 modes - with TLS or without (not recommended).
Run with TLS (Let's Encrypt).
By specifying the --letsencrypt-domain the daemon will handle SSL certificate request and configuration.
In the following example 33073 is the management service default port, and 443 will be used as port for Let's Encrypt challenge and HTTP API.
The server where you are running a container has to have a public IP (for Let's Encrypt certificate challenge).
Replace with your server's public domain (e.g. mydomain.com or subdomain sub.mydomain.com).
# create a volume
docker volume create netbird-mgmt
# run the docker container
docker run -d --name netbird-management \
-p 33073:33073 \
-p 443:443 \
-v netbird-mgmt:/var/lib/netbird \
-v ./config.json:/etc/netbird/config.json \
netbirdio/management:latest \
--letsencrypt-domain <YOUR-DOMAIN>
An example of config.json can be found here management.json
Trigger Let's encrypt certificate generation:
curl https://<YOUR-DOMAIN>
The certificate will be persisted in the datadir/letsencrypt/ folder (e.g. /var/lib/netbird/letsencrypt/) inside the container.
Make sure that the datadir is mapped to some folder on a host machine. In case you used the volume command, you can run the following to retrieve the Mountpoint:
docker volume inspect netbird-mgmt
[
{
"CreatedAt": "2021-07-25T20:45:28Z",
"Driver": "local",
"Labels": {},
"Mountpoint": "/var/lib/docker/volumes/mgmt/_data",
"Name": "netbird-mgmt",
"Options": {},
"Scope": "local"
}
]
Consequent restarts of the container will pick up previously generated certificate so there is no need to trigger certificate generation with the curl command on every restart.
Run without TLS.
# create a volume
docker volume create netbird-mgmt
# run the docker container
docker run -d --name netbird-management \
-p 33073:33073 \
-v netbird-mgmt:/var/lib/netbird \
-v ./config.json:/etc/netbird/config.json \
netbirdio/management:latest
Debug tag
We also publish a docker image with the debug tag which has the log-level set to default, plus it uses the gcr.io/distroless/base:debug image that can be used with docker exec in order to run some commands in the Management container.
shell $ docker run -d --name netbird-management-debug \
-p 33073:33073 \
-v netbird-mgmt:/var/lib/netbird \
-v ./config.json:/etc/netbird/config.json \
netbirdio/management:debug-latest
shell $ docker exec -ti netbird-management-debug /bin/sh
container-shell $
For development purposes:
Install golang gRpc tools:
#!/bin/bash
go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.26
go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@v1.1
Generate gRpc code:
#!/bin/bash
protoc -I proto/ proto/management.proto --go_out=. --go-grpc_out=.