scale in failed in community version

[veezhang]

Please check the FAQ documentation before raising an issue

The operator give the following error message:

E1016 07:54:50.505234       1 nebula_cluster_controller.go:181] NebulaCluster [e2e-nc-0-133ae97fd4f0d0ca44770b8/e2e-nc-0-cfca5c00e440791c9fe2243] reconcile failed: metad client response code -2016 name <UNSET>

And I find that the metad keeps printing the following error every 5 seconds:

I20231016 09:21:51.784233   117 ListHostsProcessor.cpp:354] List Hosts Failed, error E_LEADER_CHANGED
I20231016 09:21:51.824179   117 ListSpacesProcessor.cpp:17] List spaces failed, error E_LEADER_CHANGED

It may need to be optimized. If the leader changes, try other metad in the future instead of trying the problematic one every time?

After locating, I found that the community version does not support it by default. refs

The cluster scaling function has not been officially released in the community edition. The operations involving SUBMIT JOB BALANCE DATA REMOVE and SUBMIT JOB BALANCE DATA are experimental features in the community edition and the functionality is not stable. Before using it in the community edition, make sure to back up your data first and set enable_experimental_feature and enable_data_balance to true in the Graph configuration file.

So, should the operator prohibit the relevant operations of the community version if the corresponding features are not enabled?

Describe the bug (required)

Your Environments (required)

• OS: uname -a
• Commit id (e.g. a3ffc7d8)

How To Reproduce(required)

Steps to reproduce the behavior:

1. Step 1
2. Step 2
3. Step 3

Expected behavior

Additional context

[wey-gu]

Maybe make a judgment on whether balance is enabled via configuration before deciding whether to balance?