What causes "Failed to get schema: {str(e)}" in getredash/redash?

Read the suffix of the message — it contains the underlying _query error; fix that first (token, IDs, network)

getredash/redash · error · Exception

Failed to get schema: {str(e)}

Error message

Failed to get schema: {str(e)}

What it means

D1 runner's get_schema wraps its schema-building loop; any exception while listing tables or reading per-table column info is re-raised prefixed 'Failed to get schema: '. Usually the inner error is one of errors [12]-[14] bubbling up from _query during introspection.

Source

Thrown at redash/query_runner/d1.py:141

            # Get all tables from sqlite_master
            tables = self._query(
                "SELECT name FROM sqlite_master WHERE type='table' AND name NOT LIKE 'sqlite_%' AND name NOT LIKE '_cf_KV'"
            )

            for table in tables:
                table_name = table["name"]
                # Get column information for each table
                columns = self._query(f"PRAGMA table_info({table_name})")

                # Extract detailed column information including data types
                column_info = []
                for col in columns:
                    column_info.append({"name": col["name"], "type": col["type"]})

                schema.append({"name": table_name, "columns": column_info})

        except Exception as e:
            raise Exception(f"Failed to get schema: {str(e)}")

        return schema

    def test_connection(self):
        query = "SELECT 1 as test"
        _, error = self.run_query(query, None)
        if error:
            raise Exception(error)


register(D1QueryRunner)

View on GitHub (pinned to ca79fe988d)

Solutions

Read the suffix of the message — it contains the underlying _query error; fix that first (token, IDs, network)
Use test_connection (SELECT 1) to verify basic connectivity before refreshing schema
Confirm the token's account scope covers every database being introspected
Retry after fixing the inner error; the wrapper itself needs no separate action

Defensive patterns

Strategy: fallback

Validate before calling

_, err = d1_runner.run_query('SELECT 1', None)
assert err is None, f'fix base connectivity first: {err}'

Try / catch

try:
    schema = d1_runner.get_schema()
except Exception as e:
    if str(e).startswith('Failed to get schema:'):
        log(inner_error := str(e).split(':', 1)[1]); schema = []  # degrade gracefully

Prevention

Run test_connection before schema refresh in automation
Treat the inner errors (12-14) as the root cause; the wrapper adds no new information
Cache the last-good schema to survive transient API failures

When it happens

Trigger: Clicking 'Refresh schema' or opening the schema panel for a D1 source when the introspection queries (table list or column info) fail with a connection, auth, or parse error inside _query.

Common situations: Expired API token discovered only during schema refresh, renamed/deleted database, or transient Cloudflare API failures.

Related errors

AI-assisted analysis of getredash/redash@ca79fe988d (2026-08-28). Data as JSON: /api/errors/2ce2745cf4aeb1da. Report an issue: GitHub.