overcloud global response times are slower when 1 controller is offline

Solution In Progress - Updated -

Issue

  • We have just re-installed our overcloud with RHOSP13 Z8.

  • Whenever we take one controller node offline we see a jump in the amount of time it takes to run openstack commands.

  • Whenever we take one controller in the cluster offline using following commands:

# pcs cluster stop
# shutdown now -P
  • The following table explains the situation quite eloquently:
    Action execution time with all controllers online execution time with one controller down
    Volume Create x 1 real 0m2.911s user 0m1.488s sys 0m0.316s real 0m2.214s user 0m1.401s sys 0m0.336s
    Volume List real 0m2.732s user 0m1.581s sys 0m0.380s real 0m2.393s user 0m1.530s sys 0m0.354s
    Instance Create x 1 real 0m2.794s user 0m1.396s sys 0m0.249s real 0m33.635s user 0m1.467s sys 0m0.313s

Environment

  • Red Hat OpenStack Platform 13.0 (RHOSP)

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase of over 48,000 articles and solutions.

Current Customers and Partners

Log in for full access

Log In