Documentation
¶
Index ¶
- Constants
- Variables
- type Server
- func (s *Server) HandleRequestBody(ctx context.Context, requestID string, requestPath string, ...) (*extProcPb.ProcessingResponse, string, *types.RoutingContext, bool, int64)
- func (s *Server) HandleRequestHeaders(ctx context.Context, requestID string, req *extProcPb.ProcessingRequest) (*extProcPb.ProcessingResponse, utils.User, int64, types.RoutingAlgorithm, ...)
- func (s *Server) HandleResponseBody(ctx context.Context, requestID string, req *extProcPb.ProcessingRequest, ...) (*extProcPb.ProcessingResponse, bool)
- func (s *Server) HandleResponseHeaders(ctx context.Context, requestID string, model string, ...) (*extProcPb.ProcessingResponse, bool, int)
- func (s *Server) Process(srv extProcPb.ExternalProcessor_ProcessServer) error
Constants ¶
View Source
const ( HeaderErrorInvalidRouting = "x-error-invalid-routing-strategy" // General Error Headers HeaderErrorUser = "x-error-user" HeaderErrorRouting = "x-error-routing" HeaderErrorRequestBodyProcessing = "x-error-request-body-processing" HeaderErrorResponseUnmarshal = "x-error-response-unmarshal" HeaderErrorResponseUnknown = "x-error-response-unknown" // Model & Deployment Headers HeaderErrorNoModelInRequest = "x-error-no-model-in-request" HeaderErrorNoModelBackends = "x-error-no-model-backends" // Streaming Headers HeaderErrorStream = "x-error-stream" HeaderErrorStreaming = "x-error-streaming" HeaderErrorStreamOptionsIncludeUsage = "x-error-no-stream-options-include-usage" // Request & Target Headers HeaderWentIntoReqHeaders = "x-went-into-req-headers" HeaderTargetPod = "target-pod" HeaderRoutingStrategy = "routing-strategy" HeaderRequestID = "request-id" HeaderModel = "model" // RPM & TPM Update Errors HeaderUpdateTPM = "x-update-tpm" HeaderUpdateRPM = "x-update-rpm" HeaderErrorRPMExceeded = "x-error-rpm-exceeded" HeaderErrorTPMExceeded = "x-error-tpm-exceeded" HeaderErrorIncrRPM = "x-error-incr-rpm" HeaderErrorIncrTPM = "x-error-incr-tpm" // Rate Limiting defaults DefaultRPM = 100 DefaultTPMMultiplier = 1000 // Envs EnvRoutingAlgorithm = "ROUTING_ALGORITHM" )
Variables ¶
View Source
var (
ErrorUnknownResponse = errors.New("unknown response")
)
Functions ¶
This section is empty.
Types ¶
type Server ¶
type Server struct {
// contains filtered or unexported fields
}
func NewServer ¶
func NewServer(redisClient *redis.Client, client kubernetes.Interface, gatewayClient *gatewayapi.Clientset) *Server
func (*Server) HandleRequestBody ¶
func (s *Server) HandleRequestBody(ctx context.Context, requestID string, requestPath string, req *extProcPb.ProcessingRequest, user utils.User, routingAlgorithm types.RoutingAlgorithm) (*extProcPb.ProcessingResponse, string, *types.RoutingContext, bool, int64)
func (*Server) HandleRequestHeaders ¶
func (s *Server) HandleRequestHeaders(ctx context.Context, requestID string, req *extProcPb.ProcessingRequest) (*extProcPb.ProcessingResponse, utils.User, int64, types.RoutingAlgorithm, string)
func (*Server) HandleResponseBody ¶
func (*Server) HandleResponseHeaders ¶
Source Files
¶
Click to show internal directories.
Click to hide internal directories.