Open hi-ko opened 4 years ago
These errors are caused by the client closing the socket rather than shutting down the session cleanly, which seems to happen more often with newer clients.
You will see these messages if you have the 'Socket' sessionDebug flag set. The simplest option is to switch off that debug flag, via the smb.sessionDebug property in alfresco-global.properties.
I'll have a look into filtering out that exception. That code has been the same for a long time, it's in the original Alfresco-JLAN code, there is a catch all in the NIO based request handler, you just don't see it with Alfresco-JLAN because the debug flag is off by default.
The ERROR should be a warning then since it is not an ERROR. Checking client by client pointed out that these users didn't reboot since migration to fileServersNG. Windows seems not to support switching the smb version on reconnect. After rebooting all these clients reduced at least these ERRORS from thousands to some hundreds a day
I've put a filter into the code now so you should not see those errors again, it should be in the next fileServersNG release.
In production we see all the time following error pattern. How could we track this down and how to fix this?