Skip to main content

Will bug 6566201 be fixed in the final 6uN?

3 replies [Last post]
jacobdk
Offline
Joined: 2008-04-21

Hello forum

As some of you might know, the current line of 1.6.0 has a long-standing issue with being extremely slow on loading large JAR files from the JAR cache, an issue which is documented in bug 6566201. This issue didn't exist in 1.5.0 and earlier JRE major versions.

http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6566201

We have one such application deployed, whose JAR file is very large, 35 MB to be exact - Oracle Discoverer, some of you might know it. If I study the above bug link, it claims, that the bug should fixed in the "6u10(b08)" build.

All I can say to this is, that when I test Discoverer with the latest 1.6.0_10 beta build 25, this bug does NOT seem to have been fixed. This bug is a severe showstopper for us regarding moving to 1.6.0 on our clients, because on 1.6.0, the application still runs painfully slow compared to 1.5.0.

For your reference, Oracle MetaLink Note 465234.1 verifies Oracle's own research on the Discoverer problem to be linked to this bug.

Can someone from Sun themselves elaborate a bit on this? What is the current status of this bug?

Thanks in advance,
Jacob

Reply viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
rogerl
Offline
Joined: 2004-11-15

Yes, please let us know how you produce this bug. Using the test case that we have, the bug has been fixed. There may be a subtle difference in our test case and your code is doing. Please do send along a test case.

Thank you,
Roger

jacobdk
Offline
Joined: 2008-04-21

Hi Dimitri and Roger

Thanks for your responses, I have notified my support contact at Oracle. Do you have any contacts in the Sun JDK team at Oracle, with whom you might get more information about this issue, or should I simply pursue an answer from my support case at Oracle?

I might be able to create a test login for our public test Discovery server, where you yourself might verify, that this problem still does exist. Might that be an option?

Thanks,
Jacob

trembovetski
Offline
Joined: 2003-12-31

The bug is fixed and _verified_ in 6u10 b08. If you can still reproduce it either add comments to the bug (the responsible engineer should see the comment) or file a new bug, with your reproducible test case.

Dmitri