Bee node stuck unhealthy due to storage radius mismatch (10 vs neighborhood 9), redistribution never resumes

[ablehobo]

Context

• Bee version: 2.7.1-rc2
• Running in Docker
• Full node (full-node: true)
• Mainnet enabled
• Swap + storage incentives enabled
• Stable RPC backend (HAProxy in front of multiple Gnosis RPC providers)
• Node has been running continuously for several days

---

Summary

Node enters a persistent state where:

• isFullySynced = true
• hasSufficientFunds = true
• isFrozen = false

but:

• isHealthy = false
• node does not participate in redistribution (lastPlayedRound remains unchanged)

Logs indicate a storage radius discrepancy:

node is unhealthy due to storage radius discrepancy
self_radius=10 network_radius=9

The condition persists indefinitely and does not self-correct over time.

---

Expected behavior

Once the node is:

• fully synced
• well connected
• sufficiently funded
• not frozen

…it should eventually:

• align its storage radius with the neighborhood
• transition to isHealthy = true
• resume participation in redistribution rounds

---

Actual behavior

The node remains stuck in the following state:

• isFullySynced = true
• hasSufficientFunds = true
• isFrozen = false
• isHealthy = false

Redistribution does not resume:

{
"lastPlayedRound": 283212,
"round": 2995xx
}

State remains unchanged for multiple days.

---

Steps to reproduce

Not fully deterministic, but observed sequence:

1. Node operates normally and participates in redistribution
2. Node experiences RPC instability (temporary backend connectivity issues)
3. After RPC is stabilized, node resumes syncing and connectivity
4. Node reaches:
   • isFullySynced = true
   • healthy peer count (~120+)
5. Node enters persistent state:
   • storageRadius = 10
   • neighborhood peers predominantly at storageRadius = 9
   • isHealthy = false
   • redistribution no longer progresses

Restarting the node:

• temporarily sets isHealthy = true while isFullySynced = false
• once fully synced, state reverts to unhealthy

---

Possible solution

Unknown.

From observation, node does not automatically converge its storageRadius to match neighborhood consensus once mismatch occurs.

Possible areas to investigate:

• radius convergence logic after sync recovery
• reserve state recalculation after RPC disruption
• conditions under which storageRadius is allowed to decrease

---

Additional diagnostics

/status

{
"storageRadius": 10,
"connectedPeers": 135,
"isReachable": true
}

/reservestate

{
"radius": 11,
"storageRadius": 10
}

/redistributionstate

{
"isHealthy": false,
"isFullySynced": true,
"lastPlayedRound": 283212
}

Peer observation

Nearby peers (via /status/peers) are predominantly:

storageRadius: 9

---

AI Disclosure

• This issue contains suggestions and text generated by an LLM.
• I have reviewed the AI generated content thoroughly.
• I possess the technical expertise to responsibly review the AI generated content mentioned in this issue.