Re: [PATCH ha-manager 3/7] manager: auto rebalance: require a minimum absolute imbalance improvement

["Dominik Rusovac" <d.rusovac@proxmox.com>]

On Wed Oct 14, 2026 at 2:23 PM CEST, Daniel Kral wrote:
> On Sun Sep 20, 2026 at 6:22 PM CEST, Michael Ryom wrote:
>> The automatic load balancer only requires a relative improvement of the
>> node imbalance (default 10%). When a single resource dominates the
>> cluster load, the relative margin provides almost no hysteresis: with
>> two nodes, a resource load v and a node base load difference d, the
>> relative improvement of moving the resource is 2d / (v + d), so it
>> exceeds a margin m as soon as d > v * m / (2 - m). For a mid-size VM
>> that is a base load difference of about half a percentage point, well
>> within the normal fluctuation of e.g. the ZFS ARC or Ceph daemons, and
>> each fluctuation then qualifies a motion (and, after it was carried
>> out, its reverse).
>>
>> Additionally require the imbalance improvement to exceed a minimum
>> absolute value of 5 percentage points. This creates a dead band for
>> motions that shuffle load without meaningfully improving the cluster
>> balance. No expected motion of an existing regression test is affected;
>> the smallest occurring improvement there is 9.6 percentage points.
>>
>> Like the backoff and cooldown constants, this could be exposed as a
>> ha-auto-rebalance-* option later.
>
> I think the absolute minimum imbalance improvement is a good idea.
>
> The relative 'margin' was only introduced to compensate for the
> initially unbound imbalance value for different cluster sizes, which was
> changed to be in [0,1] for all cluster sizes only in a later revision.
>
> Though I'd prefer if this can be configured by users from the get-go.
> For simplicity, I'd merge it with the ha-auto-rebalance-margin option,
> and add a flag which can switch the value's semantic meaning between
> 'relative' (the default, because of backwards compatibility) and
> 'absolute'.

+1

>
> Except if there really are use cases which require both a relative and
> an absolute minimum imbalance improvement?

tbh atm I cannot come up with a use case

actually, I'd find it kind of strange to insist that the relative change
in the load imbalance also corresponds to the absolute change, and vice
versa

[snip]