A separate thread is used for fsync() calls to avoid it blocking the main thread. This can be more expensive than perform the fsync() directly when the file size is small.
I would like to work on it!
Work is already done, just needs to be submitted to gluster upstream.
Hi Peter, Can you please update the bug if the work is submitted to master along with the review request / patch number ? Thanks, Chetan Risbud.
We'd have to wade through the old PDQ branch in order to find this code. Sorry, that was a long time ago. It is not clear to me that on GlusterFS the DiskFile backend needs to issue an fsync in order to ensure the data is persisted, because Gluster already does that.
The version that this bug has been reported against, does not get any updates from the Gluster Community anymore. Please verify if this report is still valid against a current (3.4, 3.5 or 3.6) release and update the version, or close this bug. If there has been no update before 9 December 2014, this bug will get automatocally closed.