1

We have three identical test environments set up. On each of these environments, we have an application that, when you hit a submit button, grabs text from a form and turns it into an XML and then sends it to a WCF to be inserted into our Oracle database (version 11g). Each application has an identical app.config (aside from server-specific config keys: <add key="FolderPath" value="\\[server name]\Folder"/>) that has maxReceivedMessageSize="65536".

Here's the problem, in one environment when I try to submit the XML I get that error everyone gets: Maximum message size quota for incoming messages(65536) has been exceeded, but in the other two environments it works just fine when I submit the exact same XML. I can't figure out for the life of me why this is happening.

Here's what I've tried already in the order I tried it:

  1. Stop and started the appPool for this application's WCF in IIS.
  2. Refreshed the site for this application's WCF in IIS.
  3. Restarted the server that this application's WCF resides on.
  4. Tried submitting various sizes of data in XML form.
  5. Deleted the site from IIS and redeployed the WCF.
  6. Restarted the server again after the deployment.

Now, increasing maxReceivedMessageSize does fix the problem, I'm just wondering why I don't have to do this in the other environments.

Ikeem Wilson
  • 151
  • 1
  • 2
  • 12
  • 1) Where does the exception originate from - server-side or client-side? 2) You only mention the app.configs, have you checked the WCF services' web.configs if they are identical? – nodots Mar 31 '16 at 13:33
  • The error originates on the client. I did check to see that the web.configs were identical across environments and they are. – Ikeem Wilson Mar 31 '16 at 13:42
  • 1
    Enable WCF tracing at transport level and check what's going on the wire. Are the server responses identical on all three environments or is it possible that they have some environment-specific parts, thereby differing in length? – nodots Mar 31 '16 at 13:48
  • I will try enabling WCF tracing, I just looked it up and didn't realize that was an option. The server responses should be identical, but I will probably have a better answer for you after turning on tracing. I'll respond after running through some test cases and seeing what I get in tracing. – Ikeem Wilson Mar 31 '16 at 14:00
  • 1
    Admittedly it's a long shot, but then again you're facing a strange problem. BTW logging has size defaults too, so make sure you set maxSizeOfMessageToLog accordingly or you'll be stumped why you don't see any messages logged. ;) – nodots Mar 31 '16 at 14:04
  • 1
    So, after setting up the tracing on the WCF and setting up test data, I ran through a few test cases. Guess what... the problem fixed itself. We've had this issue for a week and a half. I wish I could tell you what happened, but I'm not sure myself. Thanks for all your help nodots!!! At least I learned about WCF tracing from all this. – Ikeem Wilson Mar 31 '16 at 20:28

0 Answers0