Did the info I had given help? Is there any more information you need from me to help troubleshoot the issue?
Hi @vspromsdn — I opened up a ticket on your account to request some more information that would be better handled in a support ticket. Are you able to view that ticket?
We are experiencing this issue as well. Was there any resolution on this?
What agent version are you using and do you have a permalink to the application where you’re running into this issue?
Where do I find the agent version? Thank you.
Thanks for providing the permalink
You can verify which version of the .NET agent you currently have installed by using Windows Programs and Features(Version column) or by using your Linux package manager. Please update us with that information and we would be happy to continue troubleshooting.
The version is: 3.11.296.0
We’ve been experiencing this behavior since installing the New .NET Agent. We have ELMA installed, and custom error logging, and no 500 errors are occurring at the time New Relic indicates. I’m not 100% sure if end users are seeing an error. Any help would be appreciated.
If you have custom error logging, it’s possible errors are being thrown that are visible to the user, but IIS (and subsequently ELMA) is logging them with a 200 status codes because the custom error page was successfully delivered. The 500 error can be seen by IIS and the agent will then catch it, but it may not necessarily look like an error to ELMA. You may want to look into parsing your IIS logs and look for the custom error page used for 500 errors and determine what status code IIS is recording.
Does your site get a significant amount of traffic? Should that be the case, you may want to look into Microsoft LogParser. It’s a great tool that uses T-SQL like queries to extract specific data from the logs (it can be used on all types of logs, not just IIS).
This may or may not account for the discrepancy, but it’s worth looking into.
We’ve done everything we can possibly think of, and we still aren’t seeing the details of what is causing the 500 internal server errors to show up in New Relic. I’ve analyzed several dozen requests in IIS that correspond to the 500 errors in New Relic, and based on the data we receive from our web site users, they are reaching the end of our online survey without any issues. Somehow, New Relic is still reporting a 500 internal server error. There are no details recorded in New Relic, Windows Event log, ELMA, and two additional custom error logs that we’ve implemented. I don’t understand why New Relic would report a 500 but not have any information as to what caused the 500. Since we are receiving all of the data we expect from our web site users, it’s not affecting our business, but it is disturbing to see this error logged without any details. Is there any way to force New Relic to capture more details about the cause of the 500 error?
HI @jmanosh - I have a similar issue for my services. I have discovered that some of the validation exceptions thrown are of the wrong type and get translated to 500 errors. Could be something related to your situation?
Do you have all the information you need? Please let me know if you have any update on this issue. Thank you.
The unfortunate reality of this is the agent can only capture the information it is given. There are no configuration options to tell the agent to get more data. It’s possible the 500 errors are thrown as the result of a request to some background process that isn’t critical to the visitor’s end results. There’s no way for the agent to know this, though. All it knows is that the 500 has been thrown, so it captures and reports that error. About the only thing I can suggest is to use something like Fiddler and navigate to as many pages on the site as possible, then see if it is showing any 500 errors in the background. That may at least identify the source of those errors from the client side.
According to Microsoft’s analysis, the problem is a specific 500.0.64 error, which according to their experts is normal, and can happen when an end-user suddenly disconnects from our application. Based on our assessment, this error is not affecting our end-users. How can we ignore this error in New Relic so it doesn’t affect our metrics?
how do we mask/hide these errors? We perfer to not ignore these errors since it will continue to get mis-reported as a issue.
Hi there @kchow,
It sounds like you’re looking for a way to classify some errors as “expected” rather than ignore them completely, and that is not currently a feature that the .NET agent has. The feature is available for some other agents but we’d need to submit a Feature Request for .NET to consider adding it.
Our Product Managers and the Engineers who plan our product roadmap are always interested in hearing the different ideas about how to make New Relic even better. In order to turn their attention toward your idea, please visit our Feature Ideas for .NET section where you can add your use case for your proposed feature. This will allow Product Managers to discover your idea, and other New Relic users can participate in the conversation by adding use cases, possible workarounds and even their vote if they think it’s a useful feature as well.
We take all feature ideas seriously, so I encourage you to submit as much detail about yours as possible.