pinging: driver: bad connection

[achettyiitr]

func TestDatabricks(t *testing.T) {
    const (
        host      = "xxx"
        port      = 443
        path      = "xxx"
        token     = "xxx"
        userAgent = "xxx"
        catalog   = ""
    )

    connector, err := dbsql.NewConnector(
        dbsql.WithServerHostname(host),
        dbsql.WithPort(port),
        dbsql.WithHTTPPath(path),
        dbsql.WithAccessToken(token),
        dbsql.WithSessionParams(map[string]string{
            "ansi_mode": "false",
        }),
        dbsql.WithUserAgentEntry(userAgent),
        dbsql.WithInitialNamespace(
            catalog,
            "",
        ),
        dbsql.WithTimeout(time.Second*15),
        dbsql.WithRetries(10, 1*time.Second, 300*time.Second),
    )
    require.NoError(t, err)

    db := sql.OpenDB(connector)
    defer func() { _ = db.Close() }()

    ctx, cancel := context.WithTimeout(context.Background(), time.Second*15)
    defer cancel()

    err = db.PingContext(ctx)
    require.NoError(t, err)
}

Issue: Getting driver bad connection, while calling Ping with ContextWithTimeout. Although, it's not reproducible. There is some edge case happening when the cluster is starting and we are executing Query. Driver Version: v1.1.0 Expectation: To return the error instead of DriverBadConnection.

[rcypher-databricks]

Thanks for raising this issue. We will look into a change.

[rcypher-databricks]

In version 1.5.1 we are now wrapping the causative error in a bad connection error. You can use errors.Is/As to check for timeout, cancellation, etc.

[rcypher-databricks]

Has the fix in 1.5.1 resolved this issue for you?