> From: Leon Rosenberg [mailto:[EMAIL PROTECTED] > Subject: request for enlightening - StandardSession and expiring flag
> Correct me if I'm wrong, but if two threads come in simultaneously > into this method, wouldn't they both bypass the first check for > expiring==true and the synchronized block would be executed twice? Probably, but we can't tell for sure without knowing whether or the expiration process sets manager = null. If it does, then the expired field is simply a mild performance enhancement to avoid blocking during session expiration. However, a quick browse through the code didn't turn up any calls that would clear the manager field directly (e.g., recycle()) during expiration; of course, I could have just missed it. If there is no clearing of the manager field done by something called within the synchronized block, then yes, this is a bug. > Shouldn't a check whether the session is already expired by a previous > call be added after the synchronized(this) statement? Yes; the check for manager == null may be sufficient (but I couldn't prove it), but it would be more straightforward to add a check for isValid inside the block. (Note that the isValid() method is also unsynchronized, which is likely to be another bug.) - Chuck THIS COMMUNICATION MAY CONTAIN CONFIDENTIAL AND/OR OTHERWISE PROPRIETARY MATERIAL and is thus for use only by the intended recipient. If you received this in error, please contact the sender and delete the e-mail and its attachments from all computers. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
