Network Error viewing large pdf

Question

I'm running into problems trying to use the Adobe Reader plugin for IE to view some large PDFs in the browser window. Apparently, the PDF fails to completely download to the browser. I've seen two slightly different behaviors:
1) When IE tries to view one of these PDFs, it will display a Loading progress bar. Usually, the progress bar will just stop before getting too far. For example: "Loading 832.5 KB/16.91MB". Attempting to view the same file again will result in similar behavior - stopping at a different place on the progress bar. Repeating this will eventually result in IE displaying a page that indicates a connection problem: "Internet Explorer cannot display the webpage".
2) Viewing the PDF in the browser will display the Loading progress bar, but before it completes, this message is displayed by Adobe Reader: "A network error occurred while accessing this document on the Internet. Would you like to close the document or reload it?"
Clicking the reload button then yields this error: "An error exists on this page. Acrobat may not display the page correctly. Please contact the person who created the PDF document to correct the problem.
Repeating the view attempt a few times eventually gives the same general IE error page described in scenario 1.
Scenario 1 seems to occur when the request is sent to a server that is sitting behind firewalls, load balancers, etc. The second condition happens when I'm using servers in the lab where there is nothing between my client and the server, other than a switch.
This only occurs when using HTTPS. If configure the server to use HTTP, then the PDF will always be viewed. This has restricted my ability to watch the traffic between the client and the server with a packet analyzer.
This only occurs with the IE plugin. Using Chrome and Firefox with the Adobe Reader plugin works every time.
It only seems to be occurring on large PDF files: the files I've been working with are all 10MB or larger.
I tried disabling the "fast web view" option in Adobe Reader, but this didn't seem to have any effect on the problem.
I disabled the option to display a PDF file in the browser window, and then the PDF file will be downloaded and opened properly in the Adobe Reader application. So it only happens with the PDF is displayed in the browser window.
I've been running my tests using IE 9 (9.0.8112.16421) and Reader plugin 10.1.3, but I've had reports from customers of similar behavior from earlier versions of the browser and reader. The web server is IIS 7.
I found several PDFs on the internet with which I've duplicated the problem. Here's an example PDF file that shows this problem.
http://www.nbb.be/DOC/BA/PDF7MB/2010/201009400082_2.PDF
There are many other similar sized files at this site:
http://www.nbb.be/pub/03_00_00_00_00/03_02_00_00_00/03_02_04_00_00.htm?l=en
many of which will also reproduce the error. However, because the problem only happens when going over HTTPS, I had to download the files and import them into my system which uses HTTPS to see the problem.
Any help would be appreciated. I'm not sure what to look at next.

Answer 1

There are several resources that show how this is done. See here for an example: http://msdn.microsoft.com/en-us/library/46c5ddfy(v=vs.100).aspx
I've added an entry into web.config to identify the presence of a custom handler:
<system.webServer>
  <handlers>
    <add name="PDFHandler" verb="*" path="*.pdf" type="PDFHttpHandler" />
  </handlers>
</system.webServer>
Then in our application code I defined a new class :
Public Class PDFHttpHandler
Implements IHttpHandler
Public ReadOnly Property IsReusable() As Boolean Implements System.Web.IHttpHandler.IsReusable
  Get
    Return False
  End Get
End Property
Public Sub ProcessRequest(ByVal context As HttpContext) Implements System.Web.IHttpHandler.ProcessRequest
  Select Case context.Request.HttpMethod
    Case "GET"
      If file.exists(context.Server.MapPath(context.Request.FilePath)) Then
        Dim requestedFile As String = context.Server.MapPath(context.Request.FilePath)
        SendContentTypeAndFile(context, requestedFile)
      Else
        context.response.statuscode = 404
      End If
    Case Else
      context.response.statuscode = 404
  End Select
End Sub
Private Function SendContentTypeAndFile(ByVal context As HttpContext, ByVal strFile As String) As HttpContext
          context.Response.ContentType = "application/pdf"
          context.Response.AppendHeader("Accept-Ranges","none")
          ' The web.config file sets the "x-ua-compatible" header to "ie=edge". This has an unfortunate side
          ' effect of casuing problems when we use IE to view PDF files.
          ' So for PDF files, we will try to override that header value.
          context.Response.AppendHeader("X-UA-Compatible", "")
  context.Response.TransmitFile(strFile)
  context.Response.end()
  Return context
End Function
End Class