Bug 1596726

Based on RFE - https://bugzilla.redhat.com/show_bug.cgi?id=1575529



Description of problem:
++++++++++++++++++++++++++
A fresh setup was created for OCP 3.10(v3.10.0-0.67.0) and CNS 3.10 using Ansible advanced deployment playbook( deploy_cluster.yaml).

Starting from gluster-block version- gluster-block-0.2.1-19.el7rhgs , load-balancing feature is enabled and to use this benefit, on the initiator side we will have to set 'prio alua' instead of prio "const" in the /etc/multipath.conf file. 

Since multipath is configured as part of a sub-task in deploy_cluster.yaml deployment, requesting to make the below changes to the multipath configuration file which is added/created in the OCP+CNS nodes.

prio "const"

A document bug is already raised to address changes in multipath conf file.
Bug https://bugzilla.redhat.com/show_bug.cgi?id=1595684

Steps to be performed when we change the conf file manually:
---------------------------

1. Add prio "alua" in /etc/multipath.conf
2. restart multipathd service

 
Sample multipath.conf file to include prio "alua" instead of prio "const"
 # cat /etc/multipath.conf
[...]
# LIO iSCSI
devices {
        device {
                vendor "LIO-ORG"
                user_friendly_names "yes" # names like mpatha
                path_grouping_policy "failover" # one path per group
                path_selector "round-robin 0"
                failback immediate
                path_checker "tur"
                prio "alua"
                no_path_retry 120
                rr_weight "uniform"
        }
}


Versions 
===========
tcmu-runner = tcmu-runner-1.2.0-20.el7rhgs 
gluster-block version- gluster-block-0.2.1-19.el7rhgs and beyond
OCP 3.10 latest

Actual results:
+++++++++++++++++
in current advanced deployment playbook, multipath.conf file of OCP/CNS hosts have [prio "const"] pre-configured. To take advantage of load balancing at gluster-block, we have to manually edit the /etc/multipath.conf file on each of the initiator nodes.


Expected results:
++++++++++++++++

Starting from OCP 3.10 deployment, tcmu-runner = tcmu-runner-1.2.0-20.el7rhgs and  gluster-block version- gluster-block-0.2.1-19.el7rhgs,  Ansible playbook should ,by default, configure prio "alua" and no manual change must be required.

Thanks in advance.