[docs] Fix notes on Corosync and Pacemaker Operations Guide

[Bogdan Dobrelya]

Ops guide should be updated as the following:
1) Notes on Corosync and Pacemaker - downtime while upgrading corosync/pacemaker stack may be avoided by putting the cluster into maintenance mode.
2) Add a troubleshooting note for the corosync/pacemaker cluster - when it stops processing signals for resources actions like start/stop/demote/promote/monitor, it must be put into the maint mode and restarted on each controller node - corosync, then pacemaker. If a service cannot be gracefully stopped on the restart command, then it must be killed with SIGKILL, then started.

[Fuel Devops McRobotson (fuel-devops-robot)]

Fix proposed to branch: master
Change author: Bogdan Dobrelya <email address hidden>
Review: https://review.fuel-infra.org/17325

[Fuel Devops McRobotson (fuel-devops-robot)]

Reviewed: https://review.fuel-infra.org/17325
Submitter: Evgeny Konstantinov <email address hidden>
Branch: master

Commit: b91ee87c22664fa162cd0a50f48b973a842e2243
Author: Bogdan Dobrelya <email address hidden>
Date: Wed Mar 16 11:40:23 2016

Fix notes on Corosync and Pacemaker

Closes-bug: #1548818

Change-Id: Ica2c17ea4686bc80a106eef5ebbcf9f531c0d18b
Signed-off-by: Bogdan Dobrelya <email address hidden>