We have an Elmah configured to send error notification emails to us whenever something is wrong. It all worked great for a while but we're noticing that sometimes Elmah would just stop sending error notifications, but still logs the error as XML.
So to verify, we have a test page which generates uncaught 500 server error. If we did an IIS reset and try the page, Elmah would correctly logs and send the notification error. However after a while (we're still unsure of how long or when it stopped working.) Elmah would only log the file without sending email notifications.
So far we have ruled out SMTP server not sending Elmah notifications because when we trigger the 500 server error while Elmah is in a bad state, there's no entry in SMTP server log at all.
In summary, the symptoms are as follows:
When Elmah is in a bad state, it'll stop generating Email notification but continue logging error in XML file.
After IIS Reset, everything goes back to normal. (Though we do have scheduled app recycle everyday, but that doesn't seem to help in this situation.)
The bad state seems to be triggered by whenever we overridethe existing binary to a newer release without restarting IIS. However, the 500 server error test indicates that Elmah sometimes goes into bad state, sometimes don't.
Is anyone observing similar symptoms?
Did you ever get this fixed?
http://groups.google.com/group/elmah/browse_thread/thread/51d24dd6c94e6dc2
Seems to be a similar problem and says:
Got it.... it's the JScript error filter. . If I strip the filter out, it works just
fine.
Related
I am finding we receive this error:
Failed to invoke 'CreateChatRequest' due to an error on the server. HubException: Method does not exist.
at _this.callbacks.<computed> (chat.min.js:2060)
at HubConnection.processIncomingData (chat.min.js:2154)
at WebSocketTransport.HubConnection.connection.onreceive (chat.min.js:1881)
at WebSocket.webSocket.onmessage (chat.min.js:3922)
The method does exist. The code does work, then after a while of testing, it returns this and won't stop.
What seems to temporarily fix the issue is restarting the Azure SigR service, but it comes back. This issue does not occur when targeting local SigR during development; it only happens against Azure SigR.
According to documentation, this seems to be a catchall error when something errors up on the server but, no exceptions are being logged. I've turned on detailed errors, but that didn't change the error coming back from the server. I've also tried catching the exception and sending back a HubException to see what is going on, but that didn't change the error message either.
TIA for any help.
So the answer, for me, was that i'd checked in (to source control) my azure service's connection string, so other devs were also using this and pointing their versions of the API service at that azure service. So when i ran my code sometimes azure would hit my API service with the new method in and it would work, sometimes it would hit someone else's API service and so fail because their code didn't have the new method in.
I'm using Xamarin.Forms with azure and everything been working fine, but today I started getting this weird error - "The request could not be completed. (Internal Server Error)", the only thing that I know of that I've changed lately is the sql server password. So I'm wondering if that could be the issue of if it could be something else and how to fix it. The diagnostics tool is also showing http server errors whenever a request is attempted.
Ok, so I figured it out, sure enough it was the change in my sql server password, to fix it I simply changed it back to what it originally was.
For a while it's been working fine. Now today, all of a sudden, when I try to list messages for a particular email address the API call just hangs for a while and then returns Backend Error (500). It still works sometimes, and sometimes it's just really slow at responding, but for the most part it seems to hang.
Any ideas what the problem could be, or how to diagnose it. I'm using the node, googleapis package, oauth2 tokens, server-side calls.
PS: I've checked the logs and there aren't any 4xx errors.
I had written a mini App in asp classic this week. It worked perfectly on the test server connecting to the test data base. Then yesterday evening I moved it from the test server to the live server updating the connection strings to the live db.
I published it as an application to the default website in the default app pool. Then I tested it and it worked perfectly.
This morning however both myself and another user receive a 500 -internal Server error when we try and save changes to the database(there appears to be no issue reading from the db) yet my two other collogues have no issue at all.
Even more odd is that the same thing is happening on the test server where the code hasn't been changed in weeks. But this morning I cannot commit to the db from there either.
I have attempt to enable more detailed error tracking and logging but the property options for the server are seem unavailable when i tried to set up custom Active Server Pages (ASP) error pages off online tutorials.
The server is used by a lot of people so I was wondering is their a permission issue depending on the user that restricts writting to the database. Or something else that may have changed to allow some users to write data and others to receive the error.
Im very knew to IIS so it may be something glaringly obvious that I haven't considered.
Thanks
This article should help you:
In earlier versions of IIS, error messages from classic ASP scripts
were sent to a Web browser, by default. Because these error messages
might reveal sensitive information to malicious users, IIS 7 and above
disables this feature by default. When your classic ASP scripts
encounter an error in IIS, you receive the following error message by
default:
An error occurred on the server when processing the URL. Please contact the system administrator.
If you are the system administrator please click here to find out more about this error.
I am on IIS 7 with a Classic ASP application using Server.GetLastError() and i added this script that writes the error details to a database as a custom '500.100' error in IIS to execute this URL. So far it's been working on most situations when there is an error in the application. However, I am noticing in the IIS log that many people are getting ASP_0147 | Internal Server Error with a 500 error and no further explanation.
I am trying to catch this error showing up in the IIS Log using the method explained above, however, these errors are not being caught (written to database) using the application i made, whereas other types of errors are being caught, but the only errors that are being caught using the script/database i made, already show up with the entire error details in the IIS log.
So the question is, how can i catch the errors still happening in the IIS Log in the database using custom errors? because the method i'm using obviously isn't catching all the errors that show in the IIS Log, only the ones with an explanation being displayed in the log. Or, if I change the Error 'Feature Settings' to 'Detailed Errors' (because now its set on, Detailed only for Local Requests) will it show up in the IIS log, etc? Why am i only seeing 500 error in the IIS Log on certain things, but not on others? Thanks.
This could be an application stall.
If the server gets a lot of requests on the same time, causing the server to "get busy", it stalls some of the requests and possibly denying them. This could give a false-positive 500 error with the error-description (0x0).