Say I have a domainname and I want to host multiple applications under that domain name:
https://domainname/app1 : Ruby on rails
https://domainname/app2 : Python Django
https://domainname/app3 : ASP.Net core
I'll go as far to say that they all could run on a linux machine, so we're not talking about IIS support.
The idea is that you can have nested applications that all have their own domain bounded context and code framework but they all comprise a larger application.
Related
I have developed a POS system using ASP.NET Core 5 MVC. I want it to work offline in case there is no internet connectivity.
How can we make an ASP.NET Core MVC app work with a local and remote database?
According to your needs description, I think the following scheme will be more suitable for you.
The first situation:
1. You need to have one physical server (used to deploy the website and install the database), or two, depending on your performance requirements.
2. If you have an independent public network IP, it is best. If you don't, you need to use an intranet penetration tool, such as ngrok. When users have a network, they can access the public network.
If the above two conditions are met, when the Internet gateway service is disconnected, users in your intranet can still access the website and can work normally. However, if you leave the intranet, such as when you return home, you will not be able to access it. At this time, it is equivalent to offline.
When the public network environment of the intranet is restored, it can be accessed from the public network again.
The second situation:
If you are using a cloud server, this situation does not exist. I don't think it makes sense to use offline and remote databases. When it is stored in the local database when offline, there will always be data out of sync.
Following a Pluralsight tutorial I built 3 Asp.Net Core projects within a Solution and all 3 of them run on different ports through which they also communicate with each other:
Project1/localhost:28851,
Project2/localhost:59418 and
Project3/localhost:33918
How does the port thing really work on production? Does this mean I will have to buy subdomains for each port or they can all run within a single domain.
I would like to know how it works in details, if possible.
Thank you
when I studied 3-tier architecture, I was told that it takes three servers:
1- web server
2- application server
3- BD server
By cons when I work in JEE, there is a single server that Glassfish or JBOS under which runs all the modules,
My question is how does 3-tiers architecture appear JEE, is it necessary to extract the web module from multi-third party applications and put it specially on a web server, for example apache and leave the EJB on glassfish?
enter image description here
Your diagram visualizes application and webcontainer separately but in case of application server, it contains itself web container/server. yes you can use only services which only rely on webserver instead of EJB. So if your specifically talking about java EE then Distributed Multitiered Applications and What is the difference between application server and web server?
helps you to understand.
Description
I am curious about how a DNX load balanced setup will work. I am accustomed to working with an ARR (Application Request Routing)/Server Farm setup with IIS like this. Typically there is nothing done on the code side with the ARR/Server Farm setup. However, with the cross platform support that DNX provides, while getting rid of IIS, this sort of load balanced setup seems like it will now need to be handled in the code. Or at least in my case I would be responsible for code deploys as opposed to a Network Admin that would install updates. The closest that I could find is session management. Can you do load balancing with this, potentially, I just do not see it there yet.
Functional Impact
Decide to download a third party ARR, use MS ARR, or Role your own ARR?
What I need
I would like to keep this from "I like this best" responses as much as possible and see responses that are "Here is why you need this" or "Here is why I found this to work well". I know that dnx is a new environment, but the experience of setting up LB servers in OSX and Linux is not. With the move to cross platform, I do not expect that the MS ARR will work on OSX or Linux. I am not very familiar with those OS's so I do not know where to begin.
Question(s)
1) Is the https://github.com/aspnet/Session/ project (or another project) expected to contain an ARR?
If no on 1
2) Is there an ARR that is cross platform (Windows, OSX, Linux)?
There's no out-of-the-box load balancer for ASP.NET 5.
Your solution is the right one, add a load balancer in front of DNX. On Windows you can use IIS, on cross plat you can use something like ngnix
No changes and no support for in-app.
ASP.NET 5 is not where this should be going. IIS will still exist and the ARR module is a part of IIS itself. Not the app.
On the page itself it says:
Works With: IIS 7, IIS 7.5, IIS 8, IIS 8.5, IIS 10
As for supporting different kind of servers? I don't see why not since the ARR module is basically just a reverse-proxy.
Nothing need to change. You can even compile to CoreCLR and have it hosted on a farm of Linux machines with Apache but have an IIS server with the ARR Module set in Reverse-Proxy to forward load balance the requests.
My problem relates to having several different web applications (PHP & Rails) running on one server and deciding which server to use for Rails since it will be coexisting with a few PHP apps currently running on apache2 and there are a few choices out there when it comes to serving Rails apps.
We have a new Debian "Squeeze" server set up running Apache2. This server hosts PHP applications and we were asked to bring over a Ruby on Rails project and run it on this box.
The Rails app will not get too much traffic because it's a micro site for some of our business partners.
With that type of environment, what's the best app server choice for the Rails app knowing that Apache2 is already serving up PHP apps?
1) Apache2 with Phusion Passenger
2) Mongrel
3) Webrick
4) Something else?
Take a look at thin. It has native SSL support, its very lightweight and easy to setup (a line in your gemfile) and wouldn't interact with any other servers running on the host.
The Rails app will not get too much traffic because it's a micro site for some of our business partners.
Seems like Passenger would be a good fit as it's easy to configure to spawn/kill workers depending on the current traffic so you might save some resources most of the time.