History | View | Annotate | Download (6.7 kB)
Change design of algorithm for computing rolling reboots
Instead of computing a coloring for one condition first and then refiningit for the other condition, we can construct a graph with edges for allconditions that prevent simultaneous reboots. This will not only result...
HRoller design updates
- Specify that there will be options for selecting nodes by at least nodegroups and tags, rather than just individually.- Specify a better handling for non-redundant instances (eg. plain or file) which today are simply ignored...
Hroller design doc
Hroller in 2.7 was implemented (as an experiment) but never explicitlydesigned. This design mentions the current functionality, and explainsthe changes that will happen for the next release to support livemaintenance scheduling (rolling reboot)....