I am working on a project that involves an embedded system which runs a non-microsoft OS with a C program for the application and am developing .NET software for its end user applications. For remote configuring with the .NET software (which can go across firewalls), I am considering using WCF. I know only a little about WCF so far but I've read that it is supposed to be interoperable with environments other than .NET. The embedded environment has an HTTP stack but no built in support for web services. Does anyone have any experience with this kind of thing or know if it would be appropriate at all? If so please provide some advice or point me in the right direction.
Thanks!
WCF is interoperable because it's accessed over HTTP. Visual Studio can help you build client libraries very quickly for WCF, but client access to WCF doesn't require anything other than HTTP calls with the proper payload. If you're looking at a remote server call and your built-in support in your embedded environment is basic HTTP, look at building your server-side as REST-formatted methods. Your debugger will thank you.
What kinds of data are you planning on transferring back and forth? For something this low level and proprietary I would recommend sticking with good old fashioned Sockets.
I will be passing configuration data back and forth...basically to enable technical support staff to remotely program the device. If I were using sockets this could be binary data, but there is a requirement that customers with firewalls shouldn't need to open any ports. Because of this I was thinking of sending XML over HTTP. So, is it better to use WCF or REST on the server side? Or WCF with REST?
I'm curious about your "customers with firewalls" requirement. Sockets with binary data or XML over HTTP can use any port (not just port 80), and I'm curious if your device will be "listening" on the network, or just making an outbound connection. If your device is listening, you will need to open a port on the firewall. Making an outbound connection ("phoning home") is much easier on the firewall.
So I think you could use sockets and binary data. However, I have faced similar issues on the last two projects, and I really wanted to implement WCF using REST on the embedded device, but no one else wanted to do it - I'm hoping someone else will be first, and publish some results!
Good Luck! (and post your results!)
Related
Advanced Attacks Detection in a Platform-as-a-Service(PaaS) Environment
In the first part of this project, i'm supposed to monitor incoming packets
in a web service, accept only HTTP & HTTPS (TCP)packets for later analysis and, drop the rest.
I was thinking doing this in JAVA, because i think it's a very flexible and
complete language and, it's present in every PaaS Environment! So, my idea is
to build a simple web page in JSP/JSF with a bean to attend this first step
of the project.
This is where i need some guidance! Because i've started considering
libpcap JAVA wrappers like jNetPcap, Jpcap and Pcap4J. But none of them is able to drop packets!
Forgetting JAVA, i also have red about other libraries like: libnet, libdnet and libcrafter.
libnet can not handle the task!
libdnet has network firewall rule manipulation capabilities, but it's a very old library and, i'm not sure it can handle integration with iptables!
libcrafter is the best! Because it's an actual updated project and, it allows the use of iptables rules in the code.
And, of course, working directly with netfilter would be the ideal scenario!
But working with libcrafter or netfilter, to follow my simple idea of a web service with a JAVA bean, i would have to write my own java wrapper by JNI! Which i assume NOT to be a simple task!
Now, what is raising many doubts in my mind, is the fact that this has to be
done in a PaaS environment! None of them (PaaS providers) seem to have the
same restrictions. There are some more flexible like AWS and Microsoft Azure that let you choose and manage a VM with the OS distro you want. Others like OpenShift, BlueMix or Cloud Foundry, in a project, only give you the option of defining the programming language, application server and, that's it! So, one might not have permissions to install libraries and control network & transport layers to manage the packets! Since the hole OS administration is handled by the provider.
Considering only the main purpose of this project, which is managing the packet flow pointed to a domain located in a PaaS environment, without the help of other servers like tcp proxies, i am desperately in need of someone pointing me a direction to start from! Because with that, i can dig as deep as needed to get a solution. Please HELP!
Thank you very much for your time and consideration.
i'm about to decide on technology choices for an agent based application used in the transportaion systems domain.
basically there will be a central system hosting the backend, and multiple agents located across town (installed on desktops) that communicate with devices/kiosks collecting data and then transmitting them back to the central server. the central server could also be hosted on the cloud.
following are important
securing the data and communications between the device and the agent
and the agent and central server.
agents should be easily installable with little or no configuration.
near 100% uptime and availability
Does WCF fit the bill here?
if so what binding types should i go for? netTCP or wsHttp with SSL/HTTPS?
WCF is definitely a fit choice for this kind of scenario. For your bindings, the actual question is what technology you are going to use. Do you want to make the agents run in a non .NET environment like Java, then you should chose for wsHttpBinding. This binding communicates through SOAP and is very interoperable.
If you chose to use .NET agents, you might as well use netTcpBinding because they use the same WCF frameworks. It also supports binary encoding. If you really need to make a choice, take a look at the MSDN Documentation.
For your agents you could use a simple console application that runs in the background as a Windows service. WIX can help you with that (install an application as windows service), but thats all I know. WIX can also help you with basic installing and configure everything for you but it has a high learning curve so you might need to invest time in it.
I have a vb6-mysql client-server desktop application which is distributed as a setup file.
It uses DLL all logical operations as well as database operations. The EXE and the DLL are installed in the server as well as the client machines. When I say server which only means the database resides in that machine no other difference in EXE or DLLs.
As all the database operations are done in the DLL when connected from a client machine, performance would be less. It is not possible now to change all the logic into database.
Is it possible to store the DLL in the server machine only and use the same DLL by the client machine also so that database connection will always be from the server itself?
Is converting the DLL to a windows service the possible solution for this?
How can I to convert it to a service?
And finally, if it is possible to make the DLL act as a service, what would the connection issues be?
You appear to be trying to rediscover n-tier application development.
The usual way this would be done using VB6 within a LAN would be to create an ActiveX EXE instead of a DLL so you can use DCOM. However DCOM isn't something you'd want to expose over the Internet.
For such cases it is more typical to use a commonly-open-port protocol such as HTTP or HTTPS. Almost everyone has firewall settings permitting outbound HTTP and HTTPS connections and most of the major Web servers undergo regular hardening to make them safer to expose to the Internet.
The classic way to do this with VB6 was to use IIS to host the Remote Data Service, which uses a form of Web Service "under the covers" where your program doesn't deal with the gory details. However this is a deprecated approach, and today configuring IIS and the RDS components can be a chore since they are locked down hard by default.
This leaves you with such things as the deprecated SOAP Toolkit or 3rd party tools such as those in the PocketSOAP suite... or you can roll your own.
Doing this from scratch can be a bit of work but is more flexible, allowing REST instead of SOAP which can have advantages in itself. You could use whatever Web server you choose that can work with VB6 (via CGI, etc.).
The hardest approach to justify might seem the simplest on the surface: create your own protocol over TCP and write a Windows Service. This can be the most flexible of all but it can be more work than other options and you are on your own as far as making it and keeping it secure. You'll probably also face firewall issues depending on where your clients are and what the local firewall policies are there.
When we could rely on DCOM the issues were relatively small aside from security configuration headaches. With the Internet in the picture it is an entirely different story.
This really isn't something you undertake casually. Even assuming that your database is safe exposed to the Internet is naive and should be rethought.
I need a way to exchange data between a process and a windows service.
The process (Windows Form Application, Console Application, in the future also a Web Solution) needs to instruct and interact with the windows service.
I want to know which way is the best to accompplish this.
I'll write the solution in C#, .NET Framework version does not matter.
In the past I've used Remoting (Activator), WCF Interface with Contracts, Inter Process Exchange IPC and some named pipe implementation. What is your experience? Other ways?
I would choose WCF. It is most modern and probably best supported approach at the moment. It "replaced" older technologies in most scenarios. Nice feature of the WCF is that if you need to move your service to other protocol you can do that simply in configuration.
If you expect that windows service will always run on the same machine as other application you can use WCF with netNamedBinding. If you decide to move your service to other machine you will have to change configuration (probably to netTcpBinding) because Named pipes in WCF are limited only to IPC.
My previous experiences have always been over an IPCChannel, mainly because the only code I've had to be involved in that does any form of inter-process communication. It's never caused me any problems and the code is working away quite merrily as I type.
The only real answer to this question is, whichever you're most comfortable with.
I am working with an electronics appliance manufacturer to embed LAN based control systems into the products. The idea is to serve up a system configuration/control interface through a web browser so clients never need to install software. We can communicate with the appliance by sending and receiving serial data through the embedded module. Since the appliance can also be controlled from a front panel UI, it creates a challenge to keep a remote web interface in sync with very low latency. It seems like websockets or some sort of Push is what we need for handling real time events from the server to clients.
I am using a Lantronix Mathport AR embedded device server. Out of the box the unit will serve up any custom HTML and java servlets/applets. We have the option to install a lightweight Linux distro if we need more flexibility. I am not sure how to implement any server side apps since the device is not running standard Apache. I believe it is using Boa.
Can anyone guide me in the right direction of how to do this?
Some general info...The WebSocket protocol (draft spec here) is a simple layer on top of TCP. What this means is that, if you already have a TCP server for your platform, implementing the WebSocket is just a matter of hours. The protocol specifies a handshake and two ways of sending data frames.
I strongly suggest you start by reading the 39 pages spec.
As Tihauan already mentioned, start by reading the spec, and also note that there are still some changes ongoing, although websockets is now more stable than it was 1 year ago.
Key point for me was the requirement that websocket data is entirely UTF-8 text, which lends itself nicely to JSON based message definitions.
Our system uses a form of embedded linux, so we then added and made use of the following libraries:
"libwebsockets" from:
http://git.warmcat.com/cgi-bin/cgit/libwebsockets/
"jansson" from:
http://www.digip.org/jansson/
Using the above as support libraries, we created an internal lightweight "client/server" that allowed our other software modules to register for certain, applicable, websocket messages, and respond as needed. Worked great.
Good luck and best regards,
I'm a bit late, but Mozilla posted a guide entitled "Writing WebSocket servers", which literally guides you through writing a websocket server.
You will need to already know how HTTP works and have medium programming experience. Depending on language support, knowledge of TCP sockets may be required. The scope of this guide is to present the minimum knowledge you need to write a WebSocket server.
https://developer.mozilla.org/en-US/docs/Web/API/WebSockets_API/Writing_WebSocket_servers