[Bugs] [Bug 1361098] Feature: Entry self-heal performance enhancements using more granular changelogs

[bugzilla at redhat.com]

https://bugzilla.redhat.com/show_bug.cgi?id=1361098

SATHEESARAN <sasundar at redhat.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ON_QA                       |VERIFIED



--- Comment #7 from SATHEESARAN <sasundar at redhat.com> ---
Tested with RHGS 3.2.0 interim build ( glusterfs-3.8.4-8.el7rhgs )

1. Created replica 3 volume and started it
2. Enabled granular entry heal
3. Used gluster volume as the data domain in RHV
4. Created few VMs and started I/O inside the VM, halted the I/O after few
minutes
5. When there were around 10,000 shards under .shards directory, killed a brick
and modified few files inside the VMs, which resulted in modification of less
than 40 shards

Observation is that when the brick was brought up, it just took 12 seconds to
heal, whereas without granular entry heal, it took 40+ seconds to heal.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug https://bugzilla.redhat.com/token.cgi?t=3Bx6w2RQJA&a=cc_unsubscribe