DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://issues.apache.org/bugzilla/show_bug.cgi?id=27249>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=27249

disposed ComponentLocator exception (when sitemap has timestamp in the future?)

[EMAIL PROTECTED] changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|Major                       |Critical



------- Additional Comments From [EMAIL PROTECTED]  2004-05-25 12:58 -------
I confirm this on my side. The weird exceptions have gone.

Now, I simply suppose that a modification timestamp in the future triggers continuous 
reloads of the 
subsitemaps, so, I believe that this bug hides something slightly more serious in the 
release/dispose 
cycle of the Excalibur pool implementation.

For sure I've already found a bug that throws a ConcurrentModificationException in the 
sitemap. See the 
attached patch which should solve the problem in the immediate.

In no whatsoever case the patch is a "FINAL" fix for the problem. What I think happens 
is that either the 
released/disposed component is returned by lookup calls while it's being 
released/disposed. Therefore 
the order of the disallocation from the pool is wrong (first the component is 
disposed, and THEN it is 
removed from the pool).

Someone with Excalibur experience should seriously look into this one.

Reply via email to