[ http://jira.qos.ch/browse/LBCLASSIC-153?page=com.atlassian.jira.plugin.syste... ] Ralph Goers edited comment on LBCLASSIC-153 at 9/24/09 10:06 AM: ----------------------------------------------------------------- On the top you will see that 4% of the overall CPU time is spent while holding the lock for LoggerContext (61,895 ms). The bottom window just shows how it was called. A way to solve this issue as well as LBCLASSIC-154 would be to change the TurboFilter to have a Timer and TimerTask that periodically wakes up and checks the file timestamp. If it is newer then it calls reconfigure while holding a write lock. The turbo filter would only acquire the read lock and then immediately release it so that logging will block when a reconfigure occurs. While no reconfiguration is taking place threads would not block at all on the read lock. was (Author: rgoers@apache.org): On the top you will see that 4% of the overall CPU time is spent while holding the lock for LoggerContext (61,895 ms). The bottom window just shows how it was called.
Excessive synchronization in ReconfigureOnChangeFilter.decide -------------------------------------------------------------
Key: LBCLASSIC-153 URL: http://jira.qos.ch/browse/LBCLASSIC-153 Project: logback-classic Issue Type: Bug Affects Versions: 0.917 Reporter: Anders Wallgren Assignee: Ceki Gulcu Attachments: screenshot-1.jpg
The synchronization in ReconfigureOnChangeFilter.decide hurts scalability. It seems unnecessary to serialize the code in changeDetected -- it should be possible to use atomic updates of nextCheck and lastModified and only serialize the actual reconfiguration.
-- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.qos.ch/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira