Merge pull request #71 from UlfBj/master

Cleanup.

server/vissv2server/vissv2server.go

@@ -36,6 +36,7 @@ import (
 	"github.com/w3c/automotive-viss2/server/vissv2server/httpMgr"
 	"github.com/w3c/automotive-viss2/server/vissv2server/serviceMgr"
 	"github.com/w3c/automotive-viss2/server/vissv2server/wsMgr"
+	"github.com/w3c/automotive-viss2/server/vissv2server/mqttMgr"
 
 	gomodel "github.com/COVESA/vss-tools/binary/go_parser/datamodel"
 	golib "github.com/COVESA/vss-tools/binary/go_parser/parserlib"
@@ -62,7 +63,7 @@ var serverComponents []string = []string{
 	"serviceMgr",
 	"httpMgr",
 	"wsMgr",
-	"mqttMgr",
+//	"mqttMgr",  //to avoid calls to the mosquitto broker if not used anyway
 	"grpcMgr",
 	"atServer",
 }
@@ -921,8 +922,8 @@ func main() {
 			go wsMgr.WsMgrInit(1, transportMgrChannel[1])
 			go transportDataSession(transportMgrChannel[1], transportDataChan[1], backendChan[1])
 		case "mqttMgr":
-			//go mqttMgr.MqttMgrInit(2, transportMgrChannel[2])
-			//go transportDataSession(transportMgrChannel[2], transportDataChan[2], backendChan[2])
+			go mqttMgr.MqttMgrInit(2, transportMgrChannel[2])
+			go transportDataSession(transportMgrChannel[2], transportDataChan[2], backendChan[2])
 		case "grpcMgr":
 			go grpcMgr.GrpcMgrInit(3, transportMgrChannel[3])
 			go transportDataSession(transportMgrChannel[3], transportDataChan[3], backendChan[3])

tutorial/content/feeder/_index.md

@@ -10,26 +10,30 @@ A feeder is a Sw component that needs to implement three tasks:
 The SW architecture shown in figure 1 reflects the division of the three tasks in that the translation (map & scale) is done in the min process,
 which spawns two threads that implement the respective interface task.
 The architecture shown handle all its communication with the server via the state storage.
-This leads to a poling paradigm and thus a potential latency and performance weakness.
+This leads to a polling paradigm and thus a potential latency and performance weakness.
+This architecture is therefore not found on he master branch, but available on the datastore-poll branch.
 ![Feeder Sw architecture, version 1](/automotive-viss2/images/feeder-sw-design-v1.jpg?width=50pc)
 * Figure 1. Feeder software architecture version 1
 
-The translation task, is divided into a mapping of the signal name, and a possible scaling of the value.
-The instructions for how to do this is encoded into a configuation file that the feeder reads at startup.
-Currently this file only contains a signal name mapping, but design work is ongoing to evolve it to support also scaling instructions.
-
 An improved architecture that eliminates the mentioned weaknesses for data flowing in the direction from the server to the feeder (i. e client write requests)
 is shown in figure 2. For write requests the server communicates directly over an IPC channel with the feeder, thus removing the ned for the feeder to poll
 the state storage to find new write requests.
 ![Feeder Sw architecture, version 2](/automotive-viss2/images/feeder-sw-design-v2.jpg?width=50pc)
 * Figure 2. Feeder software architecture version 2
 
-A feeder implementing the 2nd version of the SwA is found at the feeder branch of the repo.
-The reason to not merge it to the main branch is that the server implements both an SQLite and a Redis interface to the state storage,
-but this feeder currently only implements the Redis state storage interface. Whe nthis is updated, it will likely me merge to the master branch.
+A feeder implementing the 2nd version of the SwA is found at the master branch.
+This feeder currently only implements the Redis state storage interface, please see the Datastore chapter for Redis details.
 
-A design for how the polling on the server side can be mitigaed are  in the planning stage.
+A design for how the polling on the server side can be mitigaed is  in the planning stage.
 It is likely to require an update of the feeder interface.
 
-An OEM wanting to deploy the VISSv2 tech stack needs to implement the Vehicle interface of the feeder, for it e. g. to implemnt a CAN bu interface.
+The feeder translation task is divided into a mapping of the signal name, and a possible scaling of the value.
+The instructions for how to do this is encoded into a configuration file that the feeder reads at startup.
+Currently this file only contains a signal name mapping, but design work is ongoing to evolve it to support also scaling instructions.
+
+An OEM wanting to deploy the VISSv2 tech stack needs to implement the Vehicle interface of the feeder, for it e. g. to implement a CAN bus interface.
 Currently the Vehicle interface contains a minimal signal simulator that generates random values for the signals it is configured to support.
+
+Besides the feeder template there is also an [rl-feeder](https://github.com/w3c/automotive-viss2/tree/master/feeder/feeder-rl)
+where the Vehicle interface is implemented to connect to a RemotiveLabs broker.
+[RemotiveLabs](https://remotivelabs.com/) has a public cloud version of its broker that can be used to replay trip data available in VSS format.