Add support for restore from backup for FULL cluster (clone) with initial tokens (much faster than sstableloader)

[tarzanek]

Is this a bug report or feature request?

• Feature Request

What should the feature do: support restore of full cluster (clone) by using initial token ranges

What is use case behind this feature: clone of a cluster / migrate of existing operator managed cluster

Additional Information:

This playbook basically describes what is happening : https://github.com/scylladb/scylla-manager/tree/master/ansible/restore

scylla manager supports this, it was performed few times for various customers (on-prem, field cloud, scylla cloud), just operator support is kinda missing

[tarzanek]

expected performance bonus is like 3x when compared to sstable

the drawback is of course downtime (resp. setup of initial token ranges) (it is very possible, that it can be architected also as dual write, since load of sstables can also be done using upload / nodetool refresh)

[rzetelskik]

On one of the Operator meetings we discussed implementing this using Scylla's load and stream feature. Not a high priority for now though.

[mykaul]

I'm not sure that's a 1.9 material. I'd defer.

[scylla-operator-bot[bot]]

The Scylla Operator project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 30d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out

/lifecycle stale