[Mono-bugs] [Bug 80230][Wis] Changed - Stoping a background MonoWebServer.ApplicationServer leads to ObjectDisposedException

[bugzilla-daemon at bugzilla.ximian.com]

Please do not reply to this email- if you want to comment on the bug, go to the
URL shown below and enter your comments there.

Changed by jfburdet at humanweb.ch.

http://bugzilla.ximian.com/show_bug.cgi?id=80230

--- shadow/80230	2006-12-23 03:13:15.000000000 -0500
+++ shadow/80230.tmp.2957	2006-12-23 09:12:28.000000000 -0500
@@ -33,6 +33,55 @@
 Thanks, 
 JF
 
 ------- Additional Comments From gonzalo at ximian.com  2006-12-23 03:13 -------
 Can you run xsp with debug information and provide the stacktrace you
 get? Thanks.
+
+------- Additional Comments From jfburdet at humanweb.ch  2006-12-23 09:12 -------
+Hi,
+
+I'm getting confused because :
+
+1. Since my bug report, I was using the Mono.WebServer.dll I build 
+from 1.2.1 source code + patch mentionned above. Everything was fine.
+
+2. 
+- To reproduce the problem, I switched back to the original 1.2.1 (I 
+tried from source and using the Windows build dll from the official 
+web site).
+- The problem is that, somehow, the problem does NOT appens when 
+running from the debugger ! It only appens when running outside.
+- Also now the message I get is a "This process crashed, sending 
+information to Microsoft ?" modal Windows, containing :
+
+EventType : clr20r3     P1 : revelate.server.http.exe     P2 : 
+1.0.0.0     
+P3 : 458d2403     P4 : system     P5 : 2.0.0.0     P6 : 4333ae87     
+P7 : 2c09
+P8 : 34     P9 : system.objectdisposedexception  
+
+Then I can't attach the process from the debugger, because all 
+threads finished.
+
+I can't remember how I did manage to reproduce the problem to get it 
+from the debugger, since my code has changed a lot since.  
+
+3. 
+
+BUT I do confirm you, that if I'm applying the patch sugested above, 
+for example, having RealStop() looking like 
+
+void RealStop ()
+{
+   started = false;
+   runner.Abort ();
+   listen_socket.Close ();
+   UnloadAll ();
+   // started = false;
+   Thread.Sleep (1000);
+}
+
+the problem does not show anymore : no problem running from the 
+debugger, no problem running from outside the debugger.
+
+Weird isn'it ?