[Bugs] [Bug 1417027] New: option performance.parallel-readdir should honor cluster.readdir-optimize

bugzilla at redhat.com bugzilla at redhat.com
Fri Jan 27 02:46:15 UTC 2017


https://bugzilla.redhat.com/show_bug.cgi?id=1417027

            Bug ID: 1417027
           Summary: option performance.parallel-readdir should honor
                    cluster.readdir-optimize
           Product: GlusterFS
           Version: 3.10
         Component: unclassified
          Assignee: bugs at gluster.org
          Reporter: rgowdapp at redhat.com
                CC: bugs at gluster.org



Description of problem: With performance.parallel-readdir enabled, there is a
readdir-ahead loaded as a subvolume to DHT. Since readdir-ahead starts pre
fetching dentries in opendir, readdir-ahead is not aware of whether
cluster.readdir-optimize is enabled in DHT (as DHT uses it only in readdir).
This results in fetching all dentries from all subvols irrespective of
cluster.readdir-optimize is enabled or not.

The fix is to retain the current logic of filtering out dentries pointing to
directories from all but one subvolume even when performance.parallel-readdir
is enabled.

Version-Release number of selected component (if applicable):


How reproducible:


Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:


Additional info:

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.


More information about the Bugs mailing list