If one AppDomain is going away and one is coming online I suspect it'll be difficult to coordinate them talking to each other. Is this is a common problem?
----- Original Message ---- From: Erich Eichinger <[EMAIL PROTECTED]> To: Log4NET User <[email protected]>; Log4NET User <[email protected]> Sent: Wednesday, February 27, 2008 2:29:17 PM Subject: RE: log4net and IIS problems As AppDomains are completely isolated from each other, it needs some IPC mechanism for this. A common solution on windows is to create a named Mutex or a named Event as named objects are visible across multiple processes on a machine. -Erich From: Ron Grabowski [mailto:[EMAIL PROTECTED] Sent: Wed 2008-02-27 08:53 To: Log4NET User Subject: Re: log4net and IIS problems I wonder if there's any way for the two AppDomains to be aware of each other... ----- Original Message ---- From: Morten Andersen <[EMAIL PROTECTED]> To: Log4NET User <[email protected]> Sent: Tuesday, February 26, 2008 8:14:10 AM Subject: log4net and IIS problems <!-- _filtered {font-family:"Cambria Math";} _filtered {font-family:Calibri;} p.MsoNormal, li.MsoNormal, div.MsoNormal {margin:0cm;margin-bottom:.0001pt;font-size:11.0pt;font-family:"Calibri", "sans-serif";} a:link, span.MsoHyperlink {color:blue;text-decoration:underline;} a:visited, span.MsoHyperlinkFollowed {color:purple;text-decoration:underline;} span.EpostStil17 {font-family:"Calibri", "sans-serif";color:windowtext;} .MsoChpDefault {} _filtered {margin:70.85pt 70.85pt 70.85pt 70.85pt;} div.Section1 {} --> Hi I am using log4net on my asp.net web page hosted at an IIS server. This seems to work just fine until I do some changes in the source. Since asp.net runs as a shadow copy, changes will be compiled on the next request after the change was made. I guess this causes the log manager to duplicate itself, and the second one will not have write access to the log files. Since the old app domain will not die before all the requests are completed, the new log manager must wait for this pool to die before enabling the appenders. An alternative solution can be that the log manager retries to create the appenders if it fails. Maybe it already exist a solution for this problem? Best regards, Morten Andersen
