Erratic Invalid Viewstate issue in a .NET applicat

Question

I seem to be getting a "invalid viewstate" every now and then in the event viewer for my ASP.NET application.

Most of them (95%) seem to be referencing ScriptResource.axd (the application uses the ASP.NET AJAX library). There is no way I can remove the Ajax library either as Ajax is used everywhere..

How can I reduce these errors? I'm getting ~ 100-200 errors a day and I have no idea how to fix them! They come from different browsers, different IPs and geographical locations.

It's difficult for me to reproduce the problem because it barely even happened to me, it has only happened to me 3-4 times out of the blue.

Error:

Process information: 
    Process ID: 4004 
    Process name: w3wp.exe 
    Account name: NT AUTHORITY\NETWORK SERVICE 

Exception information: 
    Exception type: HttpException 
    Exception message: Invalid viewstate. 

Request information: 
    Request URL: http://domainnamehere/ScriptResource.axd?d=W1R6x9VzZ2C9SKnIkOmX9VRLhSjJ3nOF1GSQvPwKS3html 
    Request path: /ScriptResource.axd 
    User host address: 124.177.170.75 
    User:  
    Is authenticated: False 
    Authentication Type:  
    Thread account name: NT AUTHORITY\NETWORK SERVICE 

Thread information: 
    Thread ID: 1 
    Thread account name: NT AUTHORITY\NETWORK SERVICE 
    Is impersonating: False 
    Stack trace:    at System.Web.UI.Page.DecryptStringWithIV(String s, IVType ivType)
   at System.Web.UI.Page.DecryptString(String s)
   at System.Web.Handlers.ScriptResourceHandler.DecryptParameter(NameValueCollection queryString)
   at System.Web.Handlers.ScriptResourceHandler.ProcessRequestInternal(HttpResponse response, NameValueCollection queryString, VirtualFileReader fileReader)
   at System.Web.Handlers.ScriptResourceHandler.ProcessRequest(HttpContext context)
   at System.Web.Handlers.ScriptResourceHandler.System.Web.IHttpHandler.ProcessRequest(HttpContext context)
   at System.Web.HttpApplication.CallHandlerExecutionStep.System.Web.HttpApplication.IExecutionStep.Execute()
   at System.Web.HttpApplication.ExecuteStep(IExecutionStep step, Boolean& completedSynchronously)


Custom event details: 

For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.

I also get this error every now and then in my .NET code which happens at the same time which might be related:

Exception raised in GLOBAL.ASAX.Application_Error(): 'Padding is invalid and cannot be removed.' at System.Security.Cryptography.RijndaelManagedTransform.DecryptData(Byte[] inputBuffer, Int32 inputOffset, Int32 inputCount, Byte[]& outputBuffer, Int32 outputOffset, PaddingMode paddingMode, Boolean fLast)
   at System.Security.Cryptography.RijndaelManagedTransform.TransformFinalBlock(Byte[] inputBuffer, Int32 inputOffset, Int32 inputCount)
   at System.Security.Cryptography.CryptoStream.FlushFinalBlock()
   at System.Web.Configuration.MachineKeySection.EncryptOrDecryptData(Boolean fEncrypt, Byte[] buf, Byte[] modifier, Int32 start, Int32 length, IVType ivType, Boolean useValidationSymAlgo)
   at System.Web.UI.ObjectStateFormatter.Deserialize(String inputString)

Answer 1

The problem seems to be the lookahead downloader in IE8.

It seems to only affect IE8 in a fairly obscure set of circumstances. One of the reasons it can go unnoticed is that a 4k chunk of data appended to a URL is often discarded by the server. One of the things that seems to make it more likely is a slow network connection. Someone in one of the below resources noted that he only had the issue with his clients in New Zealand.

Lots of people explaining two separate problems, one of which is described in the question above (malformed URLs sent to server):

http://connect.microsoft.com/VisualStudio/feedback/details/434997/invalid-webresource-axd-parameters-being-generated

Article explaining that the lookahead downloader is fixed:

http://blogs.msdn.com/b/ieinternals/archive/2010/04/01/ie8-lookahead-downloader-fixed.aspx

KB980182 – Cumulative update in which issue is fixed:

http://support.microsoft.com/kb/980182

NOTE: Because the script is automatically re-downloaded if it couldn’t be retrieved by the lookahead downloader, it should be possible to change back to the old validation mode in which .axd files were not checked for validity and remove the symptoms of the issue:

<httpRuntime requestValidationMode="2.0" />

Answer 2

Use a fixed machine key (even when doing single server).

The issue occurs when using the auto configuration for the machine key, you get a new one each time the app domain is recycled. This affects viewstate validation, dynamic resources query string decryption and authentication tickets [insert other uses of the machine key].

Answer 3

I am also having this issue, and I've tried everything mentioned in all the blogs I've found (fixed machine key, viewstate size, etc). 99% of the time the error is logged on requests to ScriptResource.axd. I am using .net 3.5 SP1, on Win 2003 server. The app is hosted on two parallel identical servers, balanced 50/50. Each server has the same machine key.

Normally this error does not concern me much, however, over a 3 month period, the trend on occurance has been going way up.

Does anyone think this error is related to the Viewstate not being UrlEncoded/HtmlEncoded or UrlDecoded correctly. Perhaps there is character subset within the viewstate that some browsers are replacing with some encoded value. I'm not sure if that even makes sense..

Answer 4

I've seen problems like this when the Viewstate is too large. I've seen it happen becaue of the problem Freddy describes.

I generally dislike the idea of using Viewstate. Can you turn Viewstate off altogether?

Answer 5

I guess you are using ASP.NET AJAX. I am having the same problem. Sporadically I would find this exception in my Event Log, and the requested path is ALWAYS ScriptResource.axd.

Using fixed validationKey and decryptionKey in machineKey did not fix the problem for me.

Based on what I was able to gather, I tend to believe that this error has nothing to do with the ViewState whatsoever; my theory is that for some reason, certain UAs somehow mess up the "d" parameter of the ScriptResource.axd. The problem is easily replicable by requesting the offending path manually. This gives an "Invalid ViewState" exception, even though ViewState doesn't even apply here.

Digging through my logs, I found for example:

This request is served OK (200): /ScriptResource.axd?d=oFCAB7_vUyp7Hhe9lxZBz37lpoAxhfbWwwdfFy3Zd3z41W_33Y_9Dq6i10g9Q1NRCY1n0_DNg1nE6-DDbsD6r4EiuwoeDzp9mjDDfBNLb1k1&t=41df03cc

This slightly different request is also served OK (200): /ScriptResource.axd?d=oFCAB7_vUyp7Hhe9lxZBz37lpoAxhfbWwwdfFy3Zd3z41W_33Y_9Dq6i10g9Q1NR5ijsxQts4AfbJdACRwmQ8sHt6UAzui3spEnooPneTz01&t=41df03cc

This request fails with a 500 response and the Invalid ViewState exception: /ScriptResource.axd?d=oFCAB7_vUyp7Hhe9lxZBz37lpoAxhfbWwwdfFy3Zd3z41W_3products$ctl00$AddToCart1$id

If you look closely, the first few characters on all three request are the same, but the last few characters of the last request (in bold) clearly is Control ID "products$ctl00$AddToCart1$id" (I have a controls named products and AddToCart). I don't know how this ID got there, but in my case this is what is causing all these Invalid ViewState exceptions.

I'm not sure whether this is the same case as the OP or not, but I notice Martin's Request URL ends in "html", which is a bit of a coincidence for a parameter that is supposed to be a key...

I already have a headache thanks to this problem. And so far, the most insightful post I came across is this one http://bytes.com/topic/asp-net/answers/861764-invalid-viewstate-system-string-decryptstringwithiv

Any insights?

Answer 6

This appears to be the same IE8 issue that many people have been experiencing. What appears to happen is that somehow IE8 (in both IE8 rendering mode and IE7 compatibility mode) will lose 4096 bytes out of the middle of the HTML document and this missing data causes this exception (you usually see this in a ScriptResource or WebResource call).

Here is a Microsoft bug report on the issue: https://connect.microsoft.com/VisualStudio/feedback/ViewFeedback.aspx?FeedbackID=434997

Also there are plenty of forum, blog etc posts on this issue:

• Invalid Webresource.axd parameters being generated
• IE 8 dropping memory pages?
• http://forums.asp.net/t/1373410.aspx?PageIndex=1
• http://forums.asp.net/p/1409964/3085329.aspx

---

Microsoft has responded to this issue:

Note is a bug in Internet Explorer 8. The Internet Explorer team has been investigating this issue.

Impact: Thus far, we believe the problem has no impact on the end-user's experience with the web application; the only negative effect is the spurious/malformed requests sent by the JavaScript speculative-download engine. When the script is actually needed by the parser, it will properly be downloaded and used at that time.

Circumstances: The spurious-request appears to occur only in certain timing situations, only when a META HTTP-EQUIV tag containing a Content-Type with a CHARSET directive appears in the document, and only when a JavaScript SRC URL spans the 4096th byte of the HTTP response body.

Workaround: Hence, we currently believe this issue can be mitigated by declaring the CHARSET of the page using the HTTP Content-Type header rather than specifying it within the page.

So, rather than putting

<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=utf-8">

In your head tag, instead, send the following HTTP response header:

Content-Type: text/html; charset=utf-8

Note that specification of the charset in the HTTP header results in improved performance in all browsers, because the browser's parsers need not restart parsing from the beginning upon encountering the character set declaration. Furthermore, using the HTTP header helps mitigate certain XSS attack vectors.

NOTE: There have been reports that this problem still happens when the META HTTP-EQUIV is not on the page. We will update this comment when we have more investigation.

Posted by Microsoft on 6/30/2009 at 12:25 PM.

Edit: I still see this exception occasionally, but this bug is reported as being fixed: http://blogs.msdn.com/b/ieinternals/archive/2010/04/01/ie8-lookahead-downloader-fixed.aspx