===== git log ====
commit 88dc7324f2e862921abbf901e341a69375fcc13a
Author: Shwetha Acharya <sacharya@redhat.com>
Date:   Thu Aug 19 21:29:05 2021 +0530

    doc: Added release 8.6 notes (#2718)
    
    * doc: Added release 8.6 notes
    
    Change-Id: I4efcb4351433fa785ae5766c9289871a06d64a32
    Updates: #2702
    Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
    
    * Release notes: remove blank line
    
    * Remove blank line again

commit 263b262d836a07d56a2f04d287ca4b9fad80bd8d
Author: Xavi Hernandez <xhernandez@redhat.com>
Date:   Tue Aug 17 09:17:57 2021 +0200

    geo-rep: Improve handling of gfid mismatches (#2422)
    
    In some circumstances geo-replication can detect mismatching gfids
    between primary and secondary. These entries are fixed in an iterative
    way, assuming that after a fix, a previously failing entry could
    succeed.
    
    Previous code was trying to fix them in a loop that can be executed
    up to 10 times. If some entry cannot be fixed after 10 attempts, it's
    discarded. These fixes are very slow, so trying to do them many times
    causes geo-replication to get out of sync.
    
    To minimize the number of iterations done, this patch checks if the
    number of entries and failures remains constant after each iteration.
    If they are constant, it means that nothing else can be fixed, so it
    makes no sense to do more iterations. This reduces the number of
    iterations to 2 or 3 in most of the cases, improving geo-replication
    performance.
    
    Fixes: #2388
    Change-Id: I6d9a623a60045694e1a832195e1dc1fb9e88ae54
    Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>

commit 462721efddbd6e437031a1fecf0f6713fbbb1675
Author: Nikhil Ladha <nladha@redhat.com>
Date:   Tue Aug 10 11:52:27 2021 +0530

    glusterd: reset mgmt_v3_lock_timeout after it be used (#2689)
    

More commit messages for this ChangeLog can be found at
https://forge.gluster.org/glusterfs-core/glusterfs/commits/v8.6
