hermes - 💡(How to fix) Fix Kanban migration crashes on duplicate column: ALTER TABLE ADD COLUMN not idempotent

The kanban dispatcher fails on every tick with:

ERROR gateway.run: kanban dispatcher: tick failed on board default Traceback (most recent call last): File "/opt/hermes/gateway/run.py", line 3878, in _tick_once_for_board conn = _kb.connect(board=slug) File "/opt/hermes/hermes_cli/kanban_db.py", line 928, in connect _migrate_add_optional_columns(conn) File "/opt/hermes/hermes_cli/kanban_db.py", line 1034, in _migrate_add_optional_columns conn.execute("ALTER TABLE tasks ADD COLUMN max_retries INTEGER") sqlite3.OperationalError: duplicate column name: max_retries

Root cause:

_migrate_add_optional_columns() reads the column list once into a cols set via PRAGMA table_info(tasks), then uses if "col" not in cols guards before each ALTER TABLE ADD COLUMN.

However, this snapshot can be stale when:

1. Two gateway processes run the migration concurrently (e.g. during a restart)
2. A previous version's migration partially completed and already added the column

In both cases, the column exists in the actual schema but the cols snapshot doesn't contain it, so the guard passes and the ALTER TABLE fails with duplicate column name.

All 14 ALTER TABLE ADD COLUMN statements in _migrate_add_optional_columns() are vulnerable. The first failure crashes the entire migration, preventing the dispatcher from starting.

Suggested fix:

Wrap each ALTER TABLE ADD COLUMN in a try/except for sqlite3.OperationalError to make the migration idempotent. SQLite lacks IF NOT EXISTS for columns, so this is the standard approach:

python def _safe_add_column(conn, table, column, typedef): try: conn.execute(f"ALTER TABLE {table} ADD COLUMN {column} {typedef}") except sqlite3.OperationalError: pass # column already exists

Environment: latest version, Docker container, WhatsApp gateway.