Fixed in trunk using the 2nd patch (i.e. removing the unused messages).
Transmitting file data ..
Committed revision 51538.
Marina, it's a few hours into SCF. Let me take this as an opportunity to initiate a bug fix approval process
- What is the impact on the customer of the bug?
An alarming message (about a file being corrupted!) is logged during container shutdown if the EJB Timer Service was used. The "file corruption" itself is not serious as the file content turned out not to be used.
It is a regression as GF 2.x did not log this message.
The fix takes out the reading and writing of the file, so actually reduces security and performance concerns.
- What is the cost/risk of fixing the bug?
Fix is simple as it removes code from private methods: the paths leading to the code are clearly visible.
- Is there an impact on documentation or message strings?
Not on documentation. Two messages, which are no longer used anywhere in the code, are removed from the message bundle as part of the fix.
- Which tests should QA (re)run to verify the fix did not destabilize GlassFish?
Tests using the EJB Timer Service and then shut the timer service down.
- Which is the targeted build of 3.1.2 for this fix?
First one in the SCF period, 3.1.2b15 if I am not mistaken.