Patent Grant
7411914
The invention relates to a data ring having two or more nodes that can be disconnected and to which appliances can be connected. The nodes are provided for communication with one another. A device for fault handling is provided in the nodes. A monitoring apparatus monitors and drives the nodes.
Data rings such as these are known, for example, as a HUB. One particularly powerful variant of them operates in accordance with the fiber channel standard. The appliances that are connected to the nodes of the data ring can communicate flexibly with one another. At the same time, data is introduced into the data ring from one appliance and is read by another appliance where this is identified as being the destination appliance.
During the process of setting up a data ring such as this, initialization processes are carried out, inter alia. In the initialization processes, each appliance is registered in the data ring and a check is at the same time carried out to determine which other appliances are present with which it can communicate.
In the case of data rings of this type, the nodes have a device for fault handling which can check the data which is supplied from the appliance and can identify when faulty data signals occur. This makes it possible to identify when an appliance is faulty. When an appliance fault occurs, the device for fault handling in the node switch on an so-called bypass, which bridges that appliance. This is also then referred to as switching off that node. This ensures that only appliances that are operating correctly are included in the data ring and that the data communication always takes place correctly.
In the case of faults that could be of a temporary nature or relate to the entire data ring, there should however, often be no immediate reaction. If an appliance is disconnected in the manner described above and is removed from the data ring, there should, however, often be no immediate reaction. If an appliance is disconnected in the manner described above and is removed from the data ring by switching off a bypass, this is a very short fault for the other nodes, and generally does not lead to fault handling. It is thus possible for the other appliances that are included in the data ring not to be aware of the fact that a node has been disconnected, and subsequently to introduce data into the data ring which cannot be accessed by a another appliance. Correct operation is then no longer possible.
On the other hand, a situation occurs in which an additional appliance is connected to a node that has been switched off until then and the node is then connected. Since this appliance had not yet been connected during the process of initialization of the data ring and thus when the data ring was set up, it is still unknown to the other appliance which are connected to the other nodes and, in many cases, cannot be used correctly.
One known possible way to solve the problem is for the monitoring apparatus to monitor the nodes to determine whether any appliances have been removed from or connected to the configuration, for example whether any nodes have been connected or disconnected. The other nodes are informed via suitable fast algorithms. The solution such as this is, however, inflexible and does not operate, at least when a further data ring is connected instead of an appliance. With a concatenated data ring configuration such as this, the monitoring apparatus has no connection to the nodes in the additionally connected data ring, so that point monitoring, originating from standard logic, or driving of the appliance that are connected there is impossible.
It is accordingly an object of the invention to provide a data ring and a method for operating data rings that overcome the hereinafore-mentioned disadvantages of the heretofore-known devices and methods of this general type and that, when appliances are added and removed, the other nodes and/or appliances are reliably made aware of this. In which case, a solution such as this is intended to be flexible and must also have the capability to be used with concatenated data rings.
With the foregoing and other objects in view, there is provided, in accordance with the invention, a data ring that includes at least two disconnectable nodes, a monitoring apparatus, and an additional node. The at least two disconnectable nodes are adapted to connect to appliances and communicate with one another. In addition, the disconnectable nodes have a device for fault handling. The monitoring apparatus monitors and drives the nodes. The additional node is connected to the monitoring apparatus and allows a fault state to be produced deliberately in the other nodes.
In accordance with a further object of the invention, the data ring is distinguished by the ability to add an additional node, which can be connected by the monitoring apparatus and that allows a fault state to be initiated deliberately in other nodes that are provided in the data ring.
The advantage of the data ring according to the invention is that even short faults on one of the nodes lead to the other nodes identifying the fault and, in consequence, having the capability to carry out a reinitialization process or some other fault handling routine since the additional node that can be connected allows a fault state to be produced whose duration is sufficient in order to initiate a fault handling routine in the other nodes, when this routine identified the disconnection or the connection of nodes.
A fault state is thus produced deliberately in order to initiate fault handling routines that are provided as standard and which make the node aware of the failure of another faulty node, or of the addition of a further new node.
In one preferred embodiment, this is a fiber channel data ring, that is to say a data ring in which serial data traffic takes place using a protocol in accordance with which synchronization signals are provided. A fault state can then be produced in the other nodes by the additional node interfering with the transmission of the synchronization signals and thus initiating a synchronization fault in the other nodes. In particular, this fault state that has been produced propagates in concatenated data rings.
With the objects of the invention in view, there is also provided a method for operating a data ring according to the invention that has the advantage that optimum operation of a data ring is ensured in a small number of simple method steps.
Other features which are considered as characteristic for the invention are set forth in the appended claims.
Although the invention is illustrated and described herein as embodied in a data ring and a method for operating data rings, it is nevertheless not intended to be limited to the details shown, since various modifications and structural changes may be made therein without departing from the spirit of the invention and within the scope and range of equivalents of the claims.
The construction and method of operation of the invention, however, together with additional objects and advantages thereof will be best understood from the following description of specific embodiments when read in connection with the accompanying drawings.
Referring now to the figures of the drawings in detail and first, particularly to
If no appliance is connected to one node, the connections for an appliance are bridged by a bypass, as is likewise indicated by a dashed line. The node is thus disconnected. The same is true when the data signals that are received from an appliance by a node are faulty. The device for fault handling in the node identify this, and connect a bypass to the appliance, so that the node is disconnected. The node, N0, N1, and N2 may in this case equipped such that the switching of a bypass is carried out automatically, controlled by a device for fault handling, or else the faulty behavior of the appliance is first of all signaled to the monitoring device 5, which then passes the command for disconnection to the appropriate node.
When setting up a data ring such as this, an initialization procedure, the so-called Loop Initialization Procedure (LIP), is carried out by the individual nodes N0, N1, and N2. The individual nodes N0, N1, and N2 therefore know which other nodes are connected to the data ring 3 and which appliances are connected to them. If, in the example shown in
If the appliance FC03 fails, a problem now arises in that the node N2 switches a bypass, and the data will thus circulate further in the data ring 3. Without any further measures, this would result in ever more data being introduced into the data ring 3, without this data being removed from the data ring 3 again as well. It is therefore necessary for the other appliances FC01 and FC02 to be informed that the appliance FC03 has failed. In a small configuration, such as that illustrated in
If a faulty signal is now found in one of the other nodes two by the appliance 4 which is connected to it then, as already described, the bypass is closed and the node is disconnected, together with the appliance 4. The monitoring apparatus 5 now knows that that node has been disconnected. In a first phase of the fault handling process, the additional node 6 is connected. In consequence, the data traffic in the data ring 3 runs via the additional circuit configuration 7. In a second phase of the fault handing process, the other nodes 2 are now forced not to be disconnected, but are held in the data ring 3. At the same time, the circuit configuration 7 produces a faulty signal, that is to say a signal that does not conform with the fiber channel, and that is fed into the data ring 3. If the fiber channel protocol is used, as is described here, the synchronization signals are interfered with. In the intact nodes 2, this leads to the so-called “loss-of-sync” fault being identified. According to the appliance specifications for node modules, a loss-of-sync fault which is present for a specific minimum time initiates a reinitialization process, a so-called “Loop Initialization Procedure” (LIP). Since the fault has occurred in all of the nodes, as when the data ring was set up, all the intact nodes 2 confirm which other appliances are currently held in the data ring 3. This process identifies which of the appliances 4 and which of the associated nodes 2 are deflective and can thus no longer be addressed by the appliance.
Once this fault handling phase has been completed, the faulty node is finally removed from the data ring 3 in a third fault handling phase, and correct operation of the data ring is subsequently possible once again.
The procedure when a fault occurs is illustrated once again the in form of a diagram in
The upper part of
In the case of a refinement of the data ring 3 for the HUB 1, according to the invention, as has been described with reference to
| Number | Date | Country | Kind |
|---|---|---|---|
| 101 27 286 | Jun 2001 | DE | national |
This application is a continuation of copending International Application No. PCT/DE02/01944, filed May 27, 2002, which designated the United States and was not published in English.
| Number | Name | Date | Kind |
|---|---|---|---|
| 4498082 | Aldridge et al. | Feb 1985 | A |
| 4633468 | Skatrud et al. | Dec 1986 | A |
| 4710915 | Kitahara | Dec 1987 | A |
| 4887256 | Nakayashiki et al. | Dec 1989 | A |
| 5301185 | Cherry | Apr 1994 | A |
| 5317198 | Husbands | May 1994 | A |
| 5355362 | Gorshe et al. | Oct 1994 | A |
| 5363366 | Wisdom et al. | Nov 1994 | A |
| 5425017 | Copley et al. | Jun 1995 | A |
| 5461628 | Nakamura | Oct 1995 | A |
| 5491696 | Nishimura | Feb 1996 | A |
| 5508998 | Sha et al. | Apr 1996 | A |
| 5737370 | Hetzel | Apr 1998 | A |
| 6088141 | Merli et al. | Jul 2000 | A |
| 6175553 | Luk et al. | Jan 2001 | B1 |
| 6426962 | Cabezas et al. | Jul 2002 | B1 |
| 6574192 | Egnell | Jun 2003 | B1 |
| 6731597 | Batchellor et al. | May 2004 | B1 |
| 7003705 | Yip et al. | Feb 2006 | B1 |
| 7016430 | Grivna et al. | Mar 2006 | B1 |
| 7167444 | Afferton | Jan 2007 | B1 |
| 7171224 | Sarkkinen et al. | Jan 2007 | B2 |
| 7184663 | Kinoshita et al. | Feb 2007 | B2 |
| 20020009058 | Kelly et al. | Jan 2002 | A1 |
| 20030031126 | Mayweather et al. | Feb 2003 | A1 |
| 20040264365 | Johnson et al. | Dec 2004 | A1 |
| Number | Date | Country |
|---|---|---|
| 195 03 214 | Oct 1996 | DE |
| 199 26 569 | Dec 2000 | DE |
| 0 806 853 | Nov 1997 | EP |
| Number | Date | Country | |
|---|---|---|---|
| 20040114524 A1 | Jun 2004 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/DE02/01944 | May 2002 | US |
| Child | 10728387 | US |
