[FEATURE] Auto-balancing volumes between nodes & disks

[iosifnicolae2]

Is your feature request related to a problem?

• The actual size of volumes is very high and disks are getting low on available disk space.
• the performance is affected because instead of distributing the volumes across multiple disks, one disk is used to store all the volumes (a better option would be to balance volumes based on IOPS usage, but for now this is science fiction :D)

Describe the solution you'd like

It would be awesome if Longhorn would be able to:

1. allocate new volumes based on available disk space and/or based on actual size usage .
2. evict a volume from a disk

Describe alternatives you've considered

1. Set best-effort as data locality and then constrain the deployment on different nodes (this solution does not fix the problem entirely because we can't configure which disk to be used).
2. To evict a volume, increase the replication factor to 2, wait a few hours, delete the old replica, set replication factor to 1

---

#3840 #3887

[derekbit]

Hello @iosifnicolae2,
Could you elaborate evict a volume from a disk more?
What's the purpose of the feature?

[iosifnicolae2]

Hello @iosifnicolae2,
Could you elaborate evict a volume from a disk more?
What's the purpose of the feature?

hi, sure

by evicting a volume from a disk I mean to actually move it to move it to another disk because the current disk is almost full and we have other disks which have plenty of space

[derekbit]

Hello @iosifnicolae2,
Could you elaborate evict a volume from a disk more?
What's the purpose of the feature?

hi, sure

by evicting a volume from a disk I mean to actually move it to move it to another disk because the current disk is almost full and we have other disks which have plenty of space

Got it. More precisely, you hope the replica can automatically move to another disk which has more free space.

[joshimoo]

Epic for tracking replica scheduling improvements #3840
I think we will be adding disk anti affinity rules as part of the volume group feature #2128
The replica eviction request sounds reasonable, since we currently already allow for evicting replica from a node.

[innobead]

cc @c3y1huang

[longhorn-io-github-bot]

Pre Ready-For-Testing Checklist

• Where is the reproduce steps/test steps documented?
  The reproduce steps/test steps are at:

• Is there a workaround for the issue? If so, where is it documented?
  The workaround is at:

• Does the PR include the explanation for the fix or the feature?

• Does the PR include deployment change (YAML/Chart)? If so, where are the PRs for both YAML file and Chart?
  The PR for the YAML change is at:
  The PR for the chart change is at:

• Have the backend code been merged (Manager, Engine, Instance Manager, BackupStore etc) (including backport-needed/*)?
  The PR is at:

  • feat(auto-balance): disk in pressure longhorn-engine#1003
  • feat(auto-balance): disk in pressure longhorn-instance-manager#390
  • feat(auto-balance): disk in pressure longhorn-manager#2547

• Which areas/issues this PR might have potential impacts on?
  Area auto-balance, replica, scheduling
  Issues

• If labeled: require/LEP Has the Longhorn Enhancement Proposal PR submitted?
  The LEP PR is at feat(lep): auto-balance pressured disks #7576

• If labeled: area/ui Has the UI issue filed or ready to be merged (including backport-needed/*)?
  The UI issue/PR is at

• If labeled: require/doc Has the necessary document PR submitted or merged (including backport-needed/*)?
  The documentation issue/PR is at TBU

• If labeled: require/automation-e2e Has the end-to-end test plan been merged? Have QAs agreed on the automation test case? If only test case skeleton w/o implementation, have you created an implementation issue (including backport-needed/*)
  The automation skeleton PR is at
  The automation test case PR is at test(integration): replica auto balance disk in pressure longhorn-tests#1703
  The issue of automation test case implementation is at (please create by the template)

• If labeled: require/automation-engine Has the engine integration test been merged (including backport-needed/*)?
  The engine automation PR is at

• If labeled: require/manual-test-plan Has the manual test plan been documented?
  The updated manual test plan is at

• If the fix introduces the code for backward compatibility Has a separate issue been filed with the label release/obsolete-compatibility?
  The compatibility issue is filed at