Comment 6 for bug 1717579

Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix merged to charm-rabbitmq-server (stable/17.08)

Reviewed: https://review.openstack.org/506763
Committed: https://git.openstack.org/cgit/openstack/charm-rabbitmq-server/commit/?id=18ced509825bcdd249322d574a90d431ab9d088f
Submitter: Jenkins
Branch: stable/17.08

commit 18ced509825bcdd249322d574a90d431ab9d088f
Author: Jorge Niedbalski <email address hidden>
Date: Fri Sep 15 17:17:58 2017 -0300

    Remove update_clients() call from update-status.

    The commit [0] added a explicit call to update_clients
    which is executed on every hook run (including update-status),
    the update_clients function iterates over all the units
    related through the amqp relation and makes a call
    to the amqp_changed function.

    As the update_status hook interval is set to run every 5 minutes,
    isn't convenient to do a call to amqp_changed for every related unit
    particularly in large scale deployments, as this might cause
    an unwanted workload overhead on the cloud at every tick
    of the update-status hook.

    Further work needs to be done to improve the overall
    handling performed via the update_clients function but
    this commit alleviates the problem of leaving the update-status
    hook stuck for ~1 hour every time.

    [0] 2472e1ca9fbd68cdd50640eadd071177ddc5968e

    Change-Id: If9d6ed03d0b8a79fa0210fba57a3d54c2646214d
    Closes-Bug: #1717579
    Signed-off-by: Jorge Niedbalski <email address hidden>
    (cherry picked from commit f9b2eb310e78cdc5a8ae9461a5bf7124c5bb95d7)