By reading the docs, it is clear that the only way to interact with corda is via RPC. If you want to interact via http, then we have to write a web server exposing specific endpoints.
I am trying to write a rpc client to start a flow in the cordapp without webserver.
rpcOps.startTrackedFlowDynamic(ExampleFlow.Initiator.class, iouValue, otherParty)
I couldn't understand properly here. Should I duplicate the ExampleFlow class both on client end and in the cordapp? What is the structure of rpcclient and cordapp in this case of not having a web server?
tl;dr Write a client to start a flow on already running corda node without webserver? Thanks
Yes - currently, the client must depend on ExampleFlow.Initiator and have it available on the classpath. This is true whether it's a webserver or a regular command line client.
Related
We have used "Yo!CorDapp" example (https://github.com/corda/spring-observable-stream) to build a POC.
In this POC, can we replace angular by .NET for frontend and use IIS webserver in place of springboot webserver to talk to Corda platform?
Thanks
You can use any front-end technology you want.
As of Corda 3, your backend must be JVM-based, for two reasons:
You need to load various flow, state and other class definitions onto the classpath to pass as arguments to flows, retrieve objects from the vault, etc.
You need to use the CordaRPCClient library to create an RPC connection to the node
If you really need to write your back-end in another language, there are a few workarounds:
Create a thin Java webserver that sits between your main webserver and the node. The Java webserver translates HTTP requests from the main webserver into RPC calls to the node, and RPC responses from the node into HTTP responses to the main webserver
This is the approach taken by libraries such as Braid
Use a library such as GraalVM to compile non-JVM languages to JVM bytecode
An example of writing a JVM webserver in Javascript using GraalVM is available here: https://github.com/nitesh7sid/cordapp-example-nodejs-server-graalvm
Background-
I am extremely new to CORDA and Blockchain Platform. In the past few months i have had my share of experience working on a small project on Ethereum as platform. Ethereum blockchain was leveraged as ledger to mark Transaction as a proof of existence. It means for some action (success/failure) we have marked respective transaction on Blockchain. We may consider it as a proof of concept to show knowledge of interaction with nodes running on Ethereum Blockchain.
Infrastructure - Node.js web services, two ethereum (PoA) nodes
Question-
I would now like to port this running example on to CORDA blockchain. How would i achieve this with bare minimum changes. That means if i have a Corda network with two nodes running on my system and i want my web services to connect to one or both of the nodes and save the transaction (in its state). I understand that this certainly is not what CORDA might be meant for. Consider this question as my first step to interact with CORDA from Node.js web services.
Any inputs highly appreciated.
I recommend you go through the documentation first. your Tx will be a state. you need to build contracts and flows for a Tx to happen. Tx will happen using flows which will be initiated using Crash Shell or RPC Client. AFAIK, this client is in Kotlin or Java. so you'll have to create a JAVA or Kotlin application to instantiate this client. now in the Java application, you'll have to expose rest endpoints to communicate with the client which will initiate your flows. you can call these rest endpoints from your node application. All this has to be created in CodaApp. This is the bare minimum.
I just found there is a library.
look at this: https://gitlab.com/bluebank/braid
This can help you.
I would like to run a kafka consumer inside of a cordapp as a service, and start flows in response to messages. Can I do this without basically forking corda and exposing the node local rpcops?
This will be supported in Corda V3 via the introduction of the AppServiceHub interface: https://github.com/corda/corda/blob/master/core/src/main/kotlin/net/corda/core/node/AppServiceHub.kt.
As per the KAA references, I understand that once should only use the Notification feature, When it required to send data from server (External apps) to endpoints and Events are only used when there is a need for endpoint to endpoint communication (kind of device binding requirement)
So, To achieve request/response functionality using KAA. I need to implement any hybrid solutions like as below.
1) In my server, I can run one KAA SDK instance and use the event feature for request to the endpoint and response from the endpoint.
OR
2) From my server, I use the notification REST API for request and get the response back through the data logger feature using any in-build appender by configuring "LogUploadStrategy" as to uploads every log record as soon as it is created.
Notes For Point 1
As per Andrew, Solutions Architect of Kaa IoT platform
"You can always embed an SDK to a standalone application and host in
on the same server where kaa-node is present. This application may
receive REST API calls and forward them to particular endpoints via
Kaa events feature. However, this is useful for test purposes. I
would not recommend this solution in production because it is hard to
scale and has potential security issues"
Notes For Point 2
It satisfies the KAA reference document as well as Andrew's suggestion for request only but how can i achieve the response.
Questions For Point 1
1) What causes to scale the application and what type of security issues it faces even through it uses RSA 2048 encryption for communication?
2) Can we embed more then one SDK in standalone application and host in on the same server where kaa-node is present.
Questions For Point 2
3) if device sends the notification response along with the telemetry data, can it increase the latency and any other performance issue.
Common Questions
4) Which one is the better approach to achieve request/response functionality?
Any help or suggestion is really appreciated.
1) What causes to scale the application and what type of security issues it faces even though it uses RSA 2048 encryption for communication?
It makes the EP on the server side as a single point of failure and does not allow load balancing.
About security issues, Andrew meant: This application may receive REST API calls and this forces one to provide additional security for this REST API calls and better use your first hybrid solutions using solely event feature.
2) Can we embed more then one SDK in standalone application and host in on the same server where kaa-node is present.
No, you can't use more than one SDK in one application, but you can run a couple of instance on one machine in different directories
in order to prevent collisions of autogenerated security keys and other files.
3) if device sends the notification response along with the telemetry data, can it increase the latency and any other performance issue.
Of course, you will face some delays if start sending very frequently and big portions of data on both sides. If you have a lot of devices that sends in total a big amount of telemetry data, you can increase performance on the server side by start-up KAA in the cluster mode or add new nodes for processing requests.
4) Which one is the better approach to achieve request/response functionality?
The second hybrid solution – data collection and notification features. This doesn't cause any problem with scale and you can easily launch Kaa server in cluster mode.
I need to pause a SFTP Connector from polling and then restart it after a specified interval, in Mule ESB CE 3.2.1.
Do you know if it's possible, by using Mule's API if necessary?
And, if it was possible, how could I disconnect and reconnect an SFTP client?
Thanks in advance,
Gabriele
I suggest you first try to practice pausing/stopping/starting through JMX, in order to determine exactly what Mule component you want to interact with (the SFTP connector? the flow that uses the connector? just the SFTP inbound endpoint?).
When you have decided exactly what it is you want to do, you can then either do it:
Through JMX calls in the JVM directly on the Mule MBeans,
Through Mule API calls starting by locating the target components in the registry.