I'll try and make sure this is covered with some work with iterators I am
doing right now. It would help if you make sure there is a Bugzilla bug
posted about this... you can go ahead and assign it to me.
-scott
"Lee, William"
<William.Lee@C To: "'[EMAIL PROTECTED]'"
<[EMAIL PROTECTED]>
ognos.COM> cc:
Subject: FW: Iterator pool problem in
Xalan-J 2.2?
01/24/2002
01:40 PM
Please respond
to xalan-dev
Unfortunately, I can not send those attachments since it is > 100K...
Instead, if you looking for a test case, you can find it from the xalan's
user list...
Actually, any xsl with a <xsl:for-each> will reproduce the problem.
Thanks.
William.
-----Original Message-----
From: Lee, William
Sent: Thursday, January 24, 2002 11:38 AM
To: [EMAIL PROTECTED]
Subject: FW: Iterator pool problem in Xalan-J 2.2?
Hi All,
I tried this on the user list, but didn't got any feedback or workaround...
maybe you guys can help me out?
Please refer to the following message for more details...
By the way, it seems like the resource are hanging by the axes's iterator
pool, and never be free until the iterator is reused.
So, if 10 requests comes at the same time which create, says 40 iterators,
then most of these will just sitting there and not freeing the resource,
until there are another 10 requests comes at the same time.....
I understand why Xalan want to use the iterator pool for performance, but
keeping the resources associate with the iterators (i.e., the
transformerImpl object, which then hold onto the ResultStream...) may be a
little bit too much...?
Anyway, thanks in advance for all your help.
William.
-----Original Message-----
From: Lee, William [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, January 23, 2002 10:17 AM
To: [EMAIL PROTECTED]
Subject: Memory leak or threading problem in Xalan-J 2.2?
Hi All,
We hit a show-stopper, and hope someone can shed some lights on it. Thanks
in advance for all you help!
Our application is a server app that serves multiple requests. We are
currently using Xalan 2.2 via JAXP API.
However, we encountered a memory "leak" problem that the memory usage keep
going up after each request, and eventually "out of memory exception" is
thrown by the JVM.
By using some memory monitor, we notices that, for each request, the
StreamResult's streams (about 2M each) are hanging on by multiple instances
of TransformerImpl objects, even after request threads are destroyed by the
JVM.
This only happens when multiple requests comes in at the same time (i.e.,
multiple threads execution)... The problem do not exist for single request.
Did anyone have any idea? The attached files are the trim down version of a
XML, XSL, and the Java test program.
Thanks again.
William.
<<x.xsl>> <<x.xml>> <<TestXalan.java>>
This message may contain privileged and/or confidential information. If
you
have received this e-mail in error or are not the intended recipient, you
may not use, copy, disseminate or distribute it; do not open any
attachments, delete it immediately from your system and notify the sender
promptly by e-mail that you have done so. Thank you.
This message may contain privileged and/or confidential information. If
you
have received this e-mail in error or are not the intended recipient, you
may not use, copy, disseminate or distribute it; do not open any
attachments, delete it immediately from your system and notify the sender
promptly by e-mail that you have done so. Thank you.
