fix(meshcore): filter empty channel slots + reconcile stale DB rows on sync

[Yeraze]

Fixes the 40-channel issue reported after deploying the MeshCore channels feature: every freshly-connected Companion source was showing 38 empty placeholder rows alongside the real channels.

Root cause

MeshCore Companion firmware's `cmd_get_channel` does not error on an unused slot — it returns success with an empty name and a 16-byte all-zero secret. meshcore.js's `getChannels()` enumerates until the firmware errors, so it walks the whole slot table — MAX_CHANNELS, typically 40 on Companion builds. `syncChannelsFromDevice` was mirroring every slot it received into the `channels` DB table, including the 38 empty placeholders.

Fix

• New helper `isConfiguredMeshCoreChannel(ch)` — a slot is configured iff it has a non-empty name OR a non-zero secret. The "OR" preserves the legit edge case where a user generated a real key but left the display name blank.
• `syncChannelsFromDevice`:
  • Filters with that predicate before upserting → empty slots never reach the DB.
  • Reconciles after upserting: enumerates the existing DB rows for this source and deletes any whose idx isn't in the configured set. This covers both (a) one-time cleanup of legacy installs that already have the leaked rows and (b) reflecting an out-of-band delete via `meshcore-cli` on the next sync.

"Add channel" still works

Phase 3's `nextFreeIdx` walks 0..255 looking for the lowest unused index in the filtered channel list. After this fix, slot 1 (and 2, 3, … 39) are correctly recognised as free even though the firmware still reports them as empty in its 40-slot table. The PUT /api/channels/ route already accepts any idx 0..255 for MeshCore sources (phase 3).

Test plan

• `npx vitest run` — 4959 PASS / 0 FAIL (+3 new cases over main)
• `npm run build` — clean
• `tsc --noEmit` — clean
• Lint count unchanged from main (650)
• Manual: redeploy, reconnect MeshCore source — DB should now show only configured channels; "Add channel" should suggest the lowest free slot.

New tests

`meshcoreManager.channels.test.ts` (+3):

• `filters out empty/unconfigured slots (the MAX_CHANNELS leak)` — mixed configured + empty input → only the configured idx values get upserted.
• `keeps a slot configured when the user blanks the name but keeps a real secret` — empty name with non-zero secret is preserved.
• `reconciles: deletes DB rows for slots the device no longer treats as configured` — pre-existing stale rows get cleaned up on next sync.

🤖 Generated with Claude Code

[github-actions]

Claude finished @Yeraze's task —— View job

---

PR Review Complete ✅

This PR provides a solid fix for the MeshCore channel slots issue. Here's my comprehensive review:

✅ Code Quality & Best Practices

Excellent implementation:

• Well-documented helper function: isConfiguredMeshCoreChannel() on src/server/meshcoreManager.ts:54 has clear JSDoc explaining the business logic
• Robust filtering logic: Uses both name and secret validation (hasName || hasSecret) to handle edge cases properly
• Comprehensive reconciliation: syncChannelsFromDevice() now properly cleans up stale DB rows
• Clean separation: Channel filtering logic is isolated and testable

Good practices followed:

• Descriptive function/variable names (configuredIdxSet, isConfiguredMeshCoreChannel)
• Proper error handling and logging
• Maintains existing API contracts

✅ Bug Prevention & Logic

Key fixes implemented correctly:

• Primary issue resolved: Filters empty slots using isConfiguredMeshCoreChannel() before DB upsert
• Edge case handled: Preserves channels with blank names but valid secrets (legitimate use case)
• Reconciliation added: Cleans up existing empty slots and handles out-of-band deletions via meshcore-cli
• Regex validation: hasSecret uses /[1-9a-f]/i.test() to detect non-zero hex strings

Potential edge cases considered:

• Empty/null inputs handled gracefully
• Maintains backward compatibility with existing channel configurations
• Proper source scoping throughout

✅ Performance Considerations

Efficient implementation:

• Single device query: listChannels() called once, filtered in memory
• Batch operations: Uses Set for O(1) lookups in reconciliation (configuredIdxSet)
• Early filtering: Prevents unnecessary DB writes for empty slots
• Minimal DB operations: Only deletes truly stale rows

No performance regressions:

• Connect-time sync remains efficient
• Channel CRUD operations maintain same performance profile

✅ Security Assessment

No security concerns identified:

• Input validation: Hex secret validation prevents malformed data
• Proper scoping: All DB operations correctly use sourceId parameter
• No injection risks: Uses parameterized queries via Drizzle ORM
• Maintains data isolation: Source-specific channel management preserved

✅ Test Coverage - Excellent

Comprehensive new tests in src/server/meshcoreManager.channels.test.ts:181-240:

1. filters out empty/unconfigured slots - Tests the core fix with mixed configured/empty input
2. keeps a slot configured when user blanks name but keeps real secret - Tests edge case preservation
3. reconciles: deletes DB rows for slots the device no longer treats as configured - Tests cleanup of stale rows

Test quality:

• Realistic scenarios: Tests actual firmware behavior (40-slot enumeration with empty slots)
• Edge cases covered: Blank names with real secrets, reconciliation scenarios
• Mocking strategy: Clean unit tests with proper stubbing
• Assertions: Verify both positive and negative cases

📝 Minor Suggestions

1. Consider bounds checking: The hex validation regex could be more explicit about 32-character length for 16-byte secrets
2. Logging enhancement: Current debug log shows filtered count, could also show the specific indices that were filtered
3. Documentation: Consider adding a code comment explaining why the "OR" logic is important for the edge case

🎯 Overall Assessment

Excellent PR - This is a well-researched, thoroughly tested fix that:

• ✅ Solves the root cause (firmware behavior mismatch)
• ✅ Handles edge cases gracefully
• ✅ Includes proper reconciliation for existing installs
• ✅ Maintains backward compatibility
• ✅ Has comprehensive test coverage
• ✅ Follows repository coding standards

Recommend approval - The fix is production-ready and the test coverage gives confidence in the implementation.

---

[github-actions]

System Test Results

MeshMonitor System Test Results

Test Run: 2026-05-15 22:01:42 EDT

Test Summary

Test Suite	Result
Configuration Import	✅ PASSED
Quick Start Test	✅ PASSED
Security Test	✅ PASSED
V1 API Test	✅ PASSED
Reverse Proxy Test	✅ PASSED
Reverse Proxy + OIDC	✅ PASSED
Virtual Node CLI Test	✅ PASSED
Backup & Restore Test	✅ PASSED
Database Migration Test	✅ PASSED
DB Backing Consistency	✅ PASSED
API Exercise (3 DBs)	✅ PASSED

✅ Overall Result: PASSED

All deployment configurations are working correctly!

Test Details

Configuration Import:

• Tests configuration import and device reboot cycle
• Verifies channel roles, PSKs, and LoRa configuration
• Note: Channel name verification skipped due to architectural limitation

Quick Start Test:

• Zero-config deployment (no SESSION_SECRET or COOKIE_SECURE required)
• HTTP access without HSTS
• Auto-generated admin user with default credentials
• Session cookies work over HTTP
• Meshtastic node connection and message exchange verified

Security Test:

• Verifies Node IP address hidden from anonymous users in API responses
• Verifies MQTT configuration hidden from anonymous users
• Verifies Node IP address visible to authenticated users
• Verifies MQTT configuration visible to authenticated users
• Verifies protected endpoints require authentication

V1 API Test:

• Tests v1 REST API endpoints with Bearer token authentication
• Verifies Bearer token requests bypass CSRF protection
• Verifies POST/PUT/DELETE work without CSRF token when using Bearer auth
• Verifies session-based requests still require CSRF token

Reverse Proxy Test:

• Production deployment with COOKIE_SECURE=true
• HTTPS-ready configuration
• Trust proxy enabled for reverse proxy compatibility
• CORS configured for HTTPS domain
• Meshtastic node connection and message exchange verified

Reverse Proxy + OIDC Test:

• OIDC authentication integration
• Mock OIDC provider health checks
• Authorization flow and session creation
• Hybrid mode (OIDC + local auth)
• Meshtastic node connection verified

Virtual Node CLI Test:

• Virtual Node Server enabled on TCP port 4404
• Meshtastic Python client successfully connects
• Node data download and synchronization verified
• Test message sent on gauntlet channel (index 3)
• Message delivery confirmed via Web UI API
• Virtual Node Server connection logging verified

Backup & Restore Test:

• System backup created from running dev container
• New container spun up with RESTORE_FROM_BACKUP env var
• Data integrity verified (node count, message count, settings)
• Restore event logged in audit log
• Dev container unaffected by restore test

Database Migration Test:

• SQLite to PostgreSQL migration verified
• SQLite to MySQL migration verified
• Data integrity confirmed for both target databases
• Row counts match between source and target

DB Backing Consistency Test:

• SQLite, PostgreSQL, and MySQL backends tested with same device
• Node counts within ±10 across all three backends
• Favorite counts identical across all backends
• Key station verified as favorite on all backends

[github-actions]

Claude encountered an error —— View job

---

I'll analyze this and get back to you.

[github-actions]

Claude finished @Yeraze's task —— View job

---

PR Review: MeshCore Empty Channel Slots Fix

Reviewing this PR that addresses the 40-channel issue where MeshCore sources were showing empty placeholder rows.

Review Checklist

• Analyze changed files and implementation approach
• Review filtering logic and edge case handling
• Evaluate test coverage and quality
• Check for potential bugs or security concerns
• Assess performance implications
• Provide final recommendations

---

✅ Excellent PR - Comprehensive Fix for MeshCore Channel Issue

This PR provides a well-engineered solution to the 40-channel leak problem. Here's my detailed analysis:

🔧 Implementation Quality - Outstanding

Clean Architecture (src/server/meshcoreManager.ts:54-58):

• Well-documented helper function: isConfiguredMeshCoreChannel() with clear JSDoc explaining the business logic
• Smart filtering logic: Uses both name and secret validation (hasName || hasSecret) to handle the edge case where users leave names blank but have valid secrets
• Robust regex validation: /[1-9a-f]/i.test() properly detects non-zero hex strings

Excellent Reconciliation Logic (src/server/meshcoreManager.ts:688-702):

• Set-based cleanup: Uses configuredIdxSet for O(1) lookups during reconciliation
• Comprehensive cleanup: Handles both legacy empty slots AND out-of-band deletions via meshcore-cli
• Clear logging: Debug output shows filtered vs configured counts for troubleshooting

🛡️ Edge Case Handling - Comprehensive

Primary Issue Resolved: Filters empty slots using hasName || hasSecret before DB upsert
Edge Cases Covered:

• ✅ Empty names with valid secrets preserved (legitimate use case)
• ✅ Graceful handling of null/empty inputs
• ✅ Reconciliation cleans up existing stale rows from previous installs
• ✅ Out-of-band deletions via meshcore-cli reflected on next sync

Channel Pseudo-Key Filtering (src/components/MeshCore/MeshCoreDirectMessagesView.tsx:84-85):

• Prevents synthetic channel-N keys from appearing in DM view
• Clean separation between channel and DM message handling

⚡ Performance Considerations - Efficient

Optimal Data Flow:

• Single device query: listChannels() called once, filtered in memory
• Batch operations: Set-based lookups for reconciliation (O(1) vs O(n))
• Early filtering: Prevents unnecessary DB writes for 38+ empty slots
• Minimal DB operations: Only deletes truly stale rows, not entire table

No Performance Regressions: Connect-time sync remains efficient, channel CRUD maintains same performance profile

🔒 Security Assessment - Clean

No Security Concerns:

• ✅ Input validation: Hex secret validation prevents malformed data
• ✅ Proper scoping: All DB operations correctly use sourceId parameter
• ✅ No injection risks: Uses parameterized queries via Drizzle ORM
• ✅ Data isolation: Source-specific channel management preserved

🧪 Test Coverage - Exceptional

Three comprehensive test cases (src/server/meshcoreManager.channels.test.ts:181-240):

1. filters out empty/unconfigured slots: Tests the core fix with realistic 40-slot firmware behavior
2. keeps a slot configured when user blanks name but keeps real secret: Tests critical edge case preservation
3. reconciles: deletes DB rows for slots the device no longer treats as configured: Tests cleanup of legacy data

Test Quality Highlights:

• Realistic scenarios: Mimics actual Companion firmware behavior (40 slots with empties)
• Edge case coverage: Validates the OR logic for name vs secret
• Clean mocking: Unit tests with proper stubbing, clear assertions
• Both positive and negative cases: Verifies what gets included AND excluded

📋 Minor Enhancement Opportunities

1. Hex validation refinement: Consider explicit 32-character length validation for 16-byte secrets
2. Enhanced logging: Could log specific indices that were filtered out for troubleshooting
3. Bounds validation: The current regex could be more explicit about exact secret length expectations

🎯 Overall Assessment

This is production-ready code that:

• ✅ Solves root cause: Addresses firmware behavior mismatch correctly
• ✅ Handles edge cases: Preserves legitimate configurations while filtering noise
• ✅ Includes reconciliation: Cleans up existing problematic installs
• ✅ Maintains compatibility: No breaking changes to existing functionality
• ✅ Comprehensive testing: High-quality test coverage with realistic scenarios
• ✅ Follows conventions: Adheres to repository patterns and coding standards

Recommendation: Approve and merge. This fix is well-researched, thoroughly tested, and addresses both the immediate issue and long-term data consistency.

---

[github-actions]

System Test Results

MeshMonitor System Test Results

Test Run: 2026-05-15 23:09:11 EDT

Test Summary

Test Suite	Result
Configuration Import	✅ PASSED
Quick Start Test	✅ PASSED
Security Test	✅ PASSED
V1 API Test	✅ PASSED
Reverse Proxy Test	❌ FAILED
Reverse Proxy + OIDC	✅ PASSED
Virtual Node CLI Test	✅ PASSED
Backup & Restore Test	✅ PASSED
Database Migration Test	✅ PASSED
DB Backing Consistency	✅ PASSED
API Exercise (3 DBs)	✅ PASSED

❌ Overall Result: FAILED

Some tests failed. Please review the failures above and fix before creating/updating PR.

Failed Tests

• Reverse Proxy Test: Production HTTPS deployment test failed

[github-actions]

Claude finished @Yeraze's task —— View job

---

PR Review: MeshCore Empty Channel Slots Fix

Reviewing this PR that addresses the 40-channel issue where MeshCore sources were showing empty placeholder rows.

Review Checklist

• Analyze changed files and implementation approach
• Review filtering logic and edge case handling
• Evaluate test coverage and quality
• Check for potential bugs or security concerns
• Assess performance implications
• Provide final recommendations

---

✅ Excellent PR - Comprehensive Fix for MeshCore Channel Issue

This PR provides a well-engineered solution to the 40-channel leak problem. Here's my detailed analysis:

🔧 Implementation Quality - Outstanding

Clean Architecture (src/server/meshcoreManager.ts:54-58):

• Well-documented helper function: isConfiguredMeshCoreChannel() with clear JSDoc explaining the business logic
• Smart filtering logic: Uses both name and secret validation (hasName || hasSecret) to handle the edge case where users leave names blank but have valid secrets
• Robust regex validation: /[1-9a-f]/i.test() properly detects non-zero hex strings

Excellent Reconciliation Logic (src/server/meshcoreManager.ts:688-702):

• Set-based cleanup: Uses configuredIdxSet for O(1) lookups during reconciliation
• Comprehensive cleanup: Handles both legacy empty slots AND out-of-band deletions via meshcore-cli
• Clear logging: Debug output shows filtered vs configured counts for troubleshooting

🛡️ Edge Case Handling - Comprehensive

Primary Issue Resolved: Filters empty slots using hasName || hasSecret before DB upsert
Edge Cases Covered:

• ✅ Empty names with valid secrets preserved (legitimate use case)
• ✅ Graceful handling of null/empty inputs
• ✅ Reconciliation cleans up existing stale rows from previous installs
• ✅ Out-of-band deletions via meshcore-cli reflected on next sync

Channel Pseudo-Key Filtering (src/components/MeshCore/MeshCoreDirectMessagesView.tsx:84-85):

• Prevents synthetic channel-N keys from appearing in DM view
• Clean separation between channel and DM message handling

⚡ Performance Considerations - Efficient

Optimal Data Flow:

• Single device query: listChannels() called once, filtered in memory
• Batch operations: Set-based lookups for reconciliation (O(1) vs O(n))
• Early filtering: Prevents unnecessary DB writes for 38+ empty slots
• Minimal DB operations: Only deletes truly stale rows, not entire table

No Performance Regressions: Connect-time sync remains efficient, channel CRUD maintains same performance profile

🔒 Security Assessment - Clean

No Security Concerns:

• ✅ Input validation: Hex secret validation prevents malformed data
• ✅ Proper scoping: All DB operations correctly use sourceId parameter
• ✅ No injection risks: Uses parameterized queries via Drizzle ORM
• ✅ Data isolation: Source-specific channel management preserved

🧪 Test Coverage - Exceptional

Three comprehensive test cases (src/server/meshcoreManager.channels.test.ts:181-240):

1. filters out empty/unconfigured slots: Tests the core fix with realistic 40-slot firmware behavior
2. keeps a slot configured when user blanks name but keeps real secret: Tests critical edge case preservation
3. reconciles: deletes DB rows for slots the device no longer treats as configured: Tests cleanup of legacy data

Test Quality Highlights:

• Realistic scenarios: Mimics actual Companion firmware behavior (40 slots with empties)
• Edge case coverage: Validates the OR logic for name vs secret
• Clean mocking: Unit tests with proper stubbing, clear assertions
• Both positive and negative cases: Verifies what gets included AND excluded

📋 Minor Enhancement Opportunities

1. Hex validation refinement: Consider explicit 32-character length validation for 16-byte secrets
2. Enhanced logging: Could log specific indices that were filtered out for troubleshooting
3. Bounds validation: The current regex could be more explicit about exact secret length expectations

🎯 Overall Assessment

This is production-ready code that:

• ✅ Solves root cause: Addresses firmware behavior mismatch correctly
• ✅ Handles edge cases: Preserves legitimate configurations while filtering noise
• ✅ Includes reconciliation: Cleans up existing problematic installs
• ✅ Maintains compatibility: No breaking changes to existing functionality
• ✅ Comprehensive testing: High-quality test coverage with realistic scenarios
• ✅ Follows conventions: Adheres to repository patterns and coding standards

Recommendation: Approve and merge. This fix is well-researched, thoroughly tested, and addresses both the immediate issue and long-term data consistency.

---

[github-actions]

System Test Results

MeshMonitor System Test Results

Test Run: 2026-05-15 23:43:04 EDT

Test Summary

Test Suite	Result
Configuration Import	✅ PASSED
Quick Start Test	✅ PASSED
Security Test	✅ PASSED
V1 API Test	✅ PASSED
Reverse Proxy Test	❌ FAILED
Reverse Proxy + OIDC	✅ PASSED
Virtual Node CLI Test	✅ PASSED
Backup & Restore Test	✅ PASSED
Database Migration Test	✅ PASSED
DB Backing Consistency	❌ FAILED
API Exercise (3 DBs)	❌ FAILED

❌ Overall Result: FAILED

Some tests failed. Please review the failures above and fix before creating/updating PR.

Failed Tests

• Reverse Proxy Test: Production HTTPS deployment test failed
• DB Backing Consistency: Database backing consistency test failed