Keepalived role for ansible deployment
The last one is very sad. I am now thinking of a different approach, using lxd or systemd-nspawn. I am tired of the regular breaking changes that plague the role testing.
Full Changelog: https://github.com/evrardjp/ansible-keepalived/compare/7.0.0...8.0.0
A lot of things happened for this release. Thank you to all the contributors!
Amongst the major changes:
See more details below, or go to the full changelog (link below).
Full Changelog: https://github.com/evrardjp/ansible-keepalived/compare/6.1.0...7.0.0
Full Changelog: https://github.com/evrardjp/ansible-keepalived/compare/6.0.0...6.1.0
Full Changelog: https://github.com/evrardjp/ansible-keepalived/compare/5.1.0...6.0.0
Full Changelog: https://github.com/evrardjp/ansible-keepalived/compare/5.0.1...5.1.0
Fix incomplete removal of notify_master_rx_lower_pri.
Bug fixes, features, and important changes.
keepalived_sysctl_tcp_retries
, that automatically configures sysctl tcp_retries2. It is useful for faster failover. I used to set it on my machines, but it's not the first time I hear people wanting to have it in the role (to understand the context of the change), so here it is. By default, this will set the sysctl to a more appropriate default (8) than the current linux default of "15". If you have other roles defining your sysctls running on the same nodes as the ones running keepalived, keep that sysctl in mind! This is a "breaking" change, as behaviour of systems is effectively changed by this. We are bumping this role to a new X version.I hope you enjoyed reading those release notes, drop me a message if you like this role!
Exposed in the role the keepalived feature notify_master_rx_lower_pri
: This script is executed if a master receives an advert with priority lower than the master's advert.
Thanks @eifelmicha for this!
As this is a new feature for the role, we bump the Y version.
smtp notifications now possible, thanks to @eifelmicha !