Gluster users - Aug 2017 - modifying data via fues causes heal problem

hi there

I run off 3.10.5, have 3 peers with vols in replication.
Each time I copy some data on a client(which is a peer too) 
I see something like it:

# for QEMU-VMs:
Gathering count of entries to be healed on volume QEMU-VMs 
has been successful
Brick 
10.5.6.32:/__.aLocalStorages/0/0-GLUSTERs/0GLUSTER-QEMU-VMs
Number of entries: 0
Brick 
10.5.6.49:/__.aLocalStorages/0/0-GLUSTERs/0GLUSTER-QEMU-VMs
Number of entries: 2
Brick 
10.5.6.100:/__.aLocalStorages/0/0-GLUSTERs/0GLUSTER-QEMU-VMs
Number of entries: 1
# end of QEMU-VMs:

which heals(automatically) later ok, but, why would this 
happen in the first place? Is this expected?

Clients(all peers) mount fuse with help of autofs, like 
this(eg, on 10.5.6.49 peer):

QEMU-VMs -fstype=glusterfs,acl 
127.0.0.1,10.5.6.100,10.5.6.32:/QEMU-VMs

Is this some tuning, tweaking problems(latencies, etc)?
Is this autofs mount problem?
Or some other problems?

many thanks, L.