Hi Fatih, a bit more information about the state of the underlying
services at the time you see these errors would be useful. For example
exactly which API is it that is unreachable at the time you see these
errors? Is it possible that the endpoint services are being
simultaneously restarted this causing them to become temporarily
unavailable?
Also I assume that you have raised this bug on haproxy because you think
that is somehow to blame for apis becoming in accessible. Could you
elaborate on how you came to that conclusion?
** Changed in: haproxy (Ubuntu)
Status: New => Incomplete
--
You received this bug notification because you are a member of Ubuntu
High Availability Team, which is subscribed to haproxy in Ubuntu.
https://bugs.launchpad.net/bugs/1676072
Title:
Paused Openstack Application Unit still gets Api-Requests
Status in haproxy package in Ubuntu:
Incomplete
Bug description:
While performing in-service HA Openstack Upgrade (xenial-mitaka to
xenial-newton), ha-clustured applications (neutron-api, glance,
cinder, nova-cloud-controller) being upgraded one by one as
application and inside the cluster uprade we do perform unit pause ->
unit upgrade -> unit resume as seen in following upgrade script :
https://github.com/openstack-charmers/openstack-charms-
demos/blob/master/nyc/do-upgrade.py
Despite we perform unit pause , still paused unit gets api requests
from horizon and that causes horizon gui malfunctioning.
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/haproxy/+bug/1676072/+subscriptions
_______________________________________________
Mailing list: https://launchpad.net/~ubuntu-ha
Post to : [email protected]
Unsubscribe : https://launchpad.net/~ubuntu-ha
More help : https://help.launchpad.net/ListHelp
