I have an ASP.NET MVC 2 Beta application where I need to block incoming requests for a specific action until I have some data available to return or just release the request after 30 seconds with no new data available.
In order to accomplish this, I'm using AutoResetEvent.WaitOne(30000);
The big issue is that IIS does not seem to be accepting any new request while the thread is blocked at the WaitOne instruction. New requests get hung till the thread releases.
I need to be able to parallelize the requests while still keeping the WaitOne behavior.
Async handlers are what you're looking for. If you're building a comet solution, you may want to check out our .NET implementation of a comet server here, it'll save you some time. If you're wanting to roll your own, you'll definately need to use the async handlers to avoid hitting upper concurrency limits by the time you get past 60 or 70 users, but even with the async handlers, you'll still have to do some fancy footwork. Basically, you're still going to hit some upper limits in the threadpool unless you hand off the requests into a bounded thread pool that can basically manage all the incoming requests for you.
Good luck!
You should not be blocking incoming requests at all. If the data you need are not ready, then return an empty response, or perhaps return an error code.
For a web application, it is more advisable (not a hard rule) to return a message to tell the users to retry again later due to whatever reason you want to call it.
Stalling/blocking the requests by 'waiting' doesn't really help much as the wait is undeterministic, unless of course you have a mechanism to make it so.
I do not know the nature/context/traffic pattern of your website. 30 seconds can be a number that works for you. Perhaps my points above are not really relevant, just my 2 cents.
Actually, it turns out that this behavior only happens with ASP.NET MVC 2 Beta. I had this working fine with MVC 2 Preview 2 and rolled back to this version to re-test and confirmed that the application worked fine with that version.
Now, the question is: Why am I seeing this different behavior between these two MVC release versions, and what is the correct behavior I should expect to get in this scenario?
Related
I've been hunting around my issue for a while, probably the best I've come up with is another Stack Overflow question: How should I perform a long-running task in ASP.NET 4?
I'm in a similar place in that I'm wanting to understand what my options are, but I don't feel I know enough specifically about MVC to come to a view. I'm using MVC 5 but with the 4.8 framework, plus I note that technologies such as SignalR have become available since this question was asked. I was wondering if any experienced MVC'ers could give me a view?
I too have a long running process. More specifically, the user is importing a file. The file is delimited so the import happens line by line. The file might be thousands of lines long. Each line will be parsed and imported in a fraction of a second but the whole operation might take several minutes.
I don't particularly need behaviour to be asynchronous, but because of the length of the entire process I want to regularly update the user on progress. I'm wondering what options I have?
I've got a vague recollection that I might have looked at this problem 20-odd years ago (Classic ASP), and solved it by regular flushes, sending a bit more of the page to the client every few seconds, but I'm trying also to use a _Layout page now, so I've sent the page back already. So I don't think I have that option, even assuming such a mechanism still exists. A bit more recently, but still a while ago, I might have used javascript to poll but everything I'm reading now seems to point me to newer technologies which I'm not sure I fully understand yet.
I'm just wondering how would you solve this problem?
I would not be performing any of the file parsing on the web server, especially if it's thousands of rows long. I would delegate this to a background service of sorts, whether that be a Lambda service in the cloud or a Windows service or a scheduled task. You could then call your SignalR hub from the background task (whatever that might be) to update the progress of the import.
I use ASP.Net MVC 5 and I have a long running action which have to poll webservices, process data and store them in database.
For that I want to use TPL library to start the task async.
But I wonder how to do 3 things :
I want to report progress of this task. For this I think about SignalR
I want to be able to left the page where I start this task from and be able to report the progression across the website (from a panel on the left but this is ok)
And I want to be able to cancel this task globally (from my panel on the left)
I know quite a few about all of technologies involved. But I'm not sure about the best way to achieve this.
Is someone can help me about the best solution ?
The fact that you want to run long running work while the user can navigate away from the page that initiates the work means that you need to run this work "in the background". It cannot be performed as part of a regular HTTP request because the user might cancel his request at any time by navigating away or closing the browser. In fact this seems to be a key scenario for you.
Background work in ASP.NET is dangerous. You can certainly pull it off but it is not easy to get right. Also, worker processes can exit for many reasons (app pool recycle, deployment, machine reboot, machine failure, Stack Overflow or OOM exception on an unrelated thread). So make sure your long-running work tolerates being aborted mid-way. You can reduce the likelyhood that this happens but never exclude the possibility.
You can make your code safe in the face of arbitrary termination by wrapping all work in a transaction. This of course only works if you don't cause non-transacted side-effects like web-service calls that change state. It is not possible to give a general answer here because achieving safety in the presence of arbitrary termination depends highly on the concrete work to be done.
Here's a possible architecture that I have used in the past:
When a job comes in you write all necessary input data to a database table and report success to the client.
You need a way to start a worker to work on that job. You could start a task immediately for that. You also need a periodic check that looks for unstarted work in case the app exits after having added the work item but before starting a task for it. Have the Windows task scheduler call a secret URL in your app once per minute that does this.
When you start working on a job you mark that job as running so that it is not accidentally picked up a second time. Work on that job, write the results and mark it as done. All in a single transaction. When your process happens to exit mid-way the database will reset all data involved.
Write job progress to a separate table row on a separate connection and separate transaction. The browser can poll the server for progress information. You could also use SignalR but I don't have experience with that and I expect it would be hard to get it to resume progress reporting in the presence of arbitrary termination.
Cancellation would be done by setting a cancel flag in the progress information row. The app needs to poll that flag.
Maybe you can make use of message queueing for job processing but I'm always wary to use it. To process a message in a transacted way you need MSDTC which is unsupported with many high-availability solutions for SQL Server.
You might think that this architecture is not very sophisticated. It makes use of polling for lots of things. Polling is a primitive technique but it works quite well. It is reliable and well-understood. It has a simple concurrency model.
If you can assume that your application never exits at inopportune times the architecture would be much simpler. But this cannot be assumed. You cannot assume that there will be no deployments during work hours and that there will be no bugs leading to crashes.
Even if using http worker is a bad thing to run long task I have made a small example of how to manage it with SignalR :
Inside this example you can :
Start a task
See task progression
Cancel task
It's based on :
twitter bootstrap
knockoutjs
signalR
C# 5.0 async/await with CancelToken and IProgress
You can find the source of this example here :
https://github.com/dragouf/SignalR.Progress
PS: I was doing to some random search and then I got detrusion.com.
Whats this web application firewall ?
How it works ?
Any performance hit, if yes then how much?
Should I use this destruction.com or anything else better available.
Anybody??
I quickly glanced at the code and it doesnt appear to be doing all that much. Basically it maintains a white and black list of IPs. While it cannot be that much of a crazy performance hit you'd probably be better off doing this kind of request analyzing in a Rack middleware, that is before it even gets to the Rails request handling.
That being said, I dont like the fact that it will re-sync every 5 minutes DURING processing a given request. That is, it will block the current request while it re-syncs its ruleset / and lists. Which means that you're at the mercy of the Detrusion.com team to keep their site/API up. So when they go down you go down.
While its not as real-timey, I'd feel more comfortable to have the updating process be out of bound. Maybe you store the rules/lists in a flat file or a local DB (Redis would be perfect) which you load on app start. Then you have a frequent cron which reloads the ruleset from Detrusion and writes it locally.
Something like that. Just anything to de-couple your request handling from a Detrusion API check.
I have a particularly long running method that I need to execute from my controller. The Method is in it's own Model. I am using an async controller, and I have the method setup using asyncFunc library to make it asynchronous. I have also tried invoking it on it's own process. The problem is I want to controller to go ahead and return a view so the user can continue doing other things as the method will notify the user it is completed or has any errors via e-mail.
The problem is even thogh it is an asynchronous method the controller will not move forward to return the view until the process is done. 15+ mins. and if you navigate to a different page the method stops trying to execute.
so how can I get the method to execute as a worker and free up the controller?
Any Help would be greatly appreciated.
all the best,
Chase Q, Aucoin
Use ThreadPool.QueueUserWorkItem() as a fire-and-forget approach in the ASPX page.
Do the long-running work in the WaitCallback you pass to QUWI.
when the work is complete, that WaitCallback can send an email, or whatever it wants.
You need to take care to handle the case that the w3wp.exe is stopped during the 15 minute run. What will you do if the work is 2/3 complete? Some options are, making the work restartable, or just allowing the interrupted work to be forgotten.
Making it restartable might mean, when w3wp.exe restarts, your ASP.NET logic makes sure to begin again, any work that was interrupted. It might mean that your ASP.NET logic sets "syncpoints" so that it knows where to restart.
If you want the restartable option, you might think about Workflow, which is specifically designed for this purpose - maintaining state of long-running workflows, restarting automatically, and so on. If you use Workflow, you can set it to run asynchronously, and you may decide you do not need QueueUserWorkItem.
see also:
Moving a time taking process away from my asp.net application
the Workflow Foundation tag
This will help > http://msdn.microsoft.com/en-us/library/ms227433.aspx
It is the standard way of running a background process on the server in the .NET stack.
I don't know why, but I still live in conviction that this should not be done. Executing background threads in ASP.NET smells. You will also steal threads from ASP.NET thread pool which is controlled by IIS. It can decide that something is wrong with your worker process and restart it any time just to keep memory consumption, processing time consumption or thread consumption low. If you need background logic create custom NT service and call the process on that service either via old .NET remoting or WCF.
Btw. approach I described is used frequently in commercial applications and those which doesn't use it often self-host the whole web server.
I used System.Timers.timer in global.as in asp.net to set a timer for scheduling execute a
function
let' say transferMoney().
But it seems that this timer might stop after several hours unexpected.
And this cause that all the actions are pending.
I want to know whether there are any better methods to set up a timer in asp.net, MVC 1.0?
Thanks in advance!
It might just be because the application got recycled. Global.ashx is not really the right place to do long running tasks because if your AppDomain gets recycled your timer will die. I suggest making a job windows service instead.
Edit: Well, it's fairly easy to create a windows service project in Visual Studio just do [File] > [Add] > [New Project...] > [Windows] > [Windows Service] and you will get the stub code for the project.
It's hard to come up with a complete example so i suggest you google it. ;) There are tons of samples out there for you to look at.
This article on CodeProject seems to be a good introduction to Windows Services.
Any timer you'll use in ASP.NET apps will eventually "terminate", but this a very expected behavior due to process recycling.
The timer will never work because IIS will reschedule the worker process regularly based on Application Pool settings, so when it recycles your timer will get destroyed and you might need to reopen it.
You can put a check on whether timer object is still available or not, if not available then create it !!, using any other timer object will not work. But this still has a problem, because if you dont have any web request for particular period of time, it will still get destroyed. Best is to setup a ping monitor from other place which can keep your website alive.
You can't reliably run a timer in ASP.NET. If there are no requests coming in, the IIS can shut down the application, and it will not start until the next request arrives.
Why do you think that you need a timer? In most web applications this is not needed at all to do periodical updates unless they depend on an external source.
If you are just moving data around inside your application, the actual transactions doesn't have to happen at an exact interval, you only have to calculate what the result would be if they had happened. Whenever a request comes in, you calculate how many transactions would have happened since the last request, and do them to catch up to the current state.
If your transactions rely on an external source so that they actually has to run at a specific time, you simply can't do it with ASP.NET alone. You need an application that runs outside IIS, for example started periodically by the windows scheduler.
You could try the system.threading.timer
http://msdn.microsoft.com/en-us/library/system.threading.timer.aspx