yahoo-eng-team team mailing list archive

[OpenStack Infra <1812576@xxxxxxxxxxxxxxxxxx>]

Reviewed:  https://review.openstack.org/632014
Committed: https://git.openstack.org/cgit/openstack/neutron/commit/?id=6f927cc11970b683305c96412d248e63cc15c12b
Submitter: Zuul
Branch:    master

commit 6f927cc11970b683305c96412d248e63cc15c12b
Author: liuchengqian90 <xieling1990.happy@xxxxxxx>
Date:   Wed Jan 30 17:48:59 2019 +0800

    Clear old rules that have been applied before applying new rules.
    
    To prevent data from being out of sync in the following situations:
    1. Create a policy with two rules bound to the virtual machine
    2. Stop l2-agent
    3. Delete/change/clear policy rule
    4. Start l2-agent (the rule is still there, out-of-sync)
    
    Change-Id: I194c918d859172c31ae5ce1af925fdbb388f9cfb
    Closes-Bug: #1812576


** Changed in: neutron
       Status: In Progress => Fix Released

-- 
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to neutron.
https://bugs.launchpad.net/bugs/1812576

Title:
  L2 agent do not clear old QoS rules after restart

Status in neutron:
  Fix Released

Bug description:
  This fix is optimized for https://bugs.launchpad.net/neutron/+bug/1810025.
  BUG 1810025 only considers the case where the rule is emptied when the agent is stopped, and does not consider the case where the rule is deleted but not emptied when the agent is stopped.

  Think about that:
  1. Create two different types of rules, such as bandwidth_limit and minimum_bandwidth
  2. Stop l2_agent
  3. Delete a type of rule, such as bandwidth_limit
  4. Start agent
  5. Bandwidth_limit still exists

To manage notifications about this bug go to:
https://bugs.launchpad.net/neutron/+bug/1812576/+subscriptions