Hi, It seems there is a regression introduced on v2.6.30-rc1 while using DEFER_SETUP on a RFCOMM socket. It happens when the user reject the authorization: 2009-05-27 13:38:14.048925 < ACL data: handle 42 flags 0x02 dlen 12 L2CAP(s): Disconn req: dcid 0x0042 scid 0x0040 2009-05-27 13:38:14.052864 > HCI Event: Number of Completed Packets (0x13) plen 5 handle 42 packets 1 2009-05-27 13:38:14.174975 > ACL data: handle 42 flags 0x02 dlen 12 L2CAP(s): Disconn rsp: dcid 0x0042 scid 0x0040 No hci disconnect command is sent so the ACL link will stay up forever, I suspect some timeout got removed in rc1 since with v2.6.29 after some period the ACL is disconnected, but for my surprise it seems HCI_IDLE_TIMEOUT is not being used. -- Luiz Augusto von Dentz Engenheiro de Computação -- To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html