Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

Rule 1: Before Deploying Identify Redundant Paths for Management Communications to Each TCX1000 Device

 

This topic describes deployment rule 1 for the TCX1000-RDM20 and TCX1000-ILA, which requires redundant paths for management communications.

Before You Deploy TCX1000 Devices

Before you deploy TCX1000 devices in your optical network, we recommend that you sketch out your network and identity how each TCX1000 device (TCX1000-RDM20 and TCX1000-ILA) in the network will handle redundant management communications.

Note

Every TCX Series device managed by the proNX Optical Director must have redundant links to your DCN HA network.

Your DCN HA design must include a fault tolerant management network design with redundant paths for management communications to each TCX1000 device and the proNX Optical Director. There are several ways to achieve this and it is your responsibility to design your own DCN HA network.

TCX1000-RDM20

Best Practice

The best way to provide redundant management communications to the TCX1000-RDM20 is to connect the TCX1000-RDM20 to your DCN HA is using a Layer 2 switch between the DCN 0 and DCN 1 ports and your DCN HA. In this case the TCX1000-RDM20 has redundant paths directly to your DCN HA with the proNX Optical Director.

If the deployment site has no DCN access, you can manage the TCX1000-RDM20 over its line port if you enable OSC forwarding on the device.

In remotely connected multi-directional ROADM nodes, you can daisy chain the DCN ports on the TCX1000-RDM20s within the node in order to pass along management messages addressed to the DCN port of a co-located TCX1000-RDM20. If the management message is addressed to a TCX1000 device that is further down in the optical network, the TCX1000-RDM20 will pass the message along over its line port as long as OSC forwarding is enabled.

As per our requirement for redundant paths for proNX Optical Director IP management communications to each TCX1000 device in your optical network, you must ensure that your TCX1000 optical network, at minimum, is homed at each end by a TCX1000-RDM20 connected through a Layer 2 switch on the same DCN HA as the proNX Optical Director. Enabling OSC forwarding on the homed TCX1000-RDM20s at each end of your optical network provides the second path for management communications—each homed TCX1000-RDM20 can forward proNX Optical Director management messages over their line ports to remotely connected TCX1000 devices using the OSC signal. The IP address used by the proNX Optical Director is the address of the remotely connected TCX1000 device. If the remotely connected device is a TCX1000-RDM20, it must also have OSC forwarding enabled in order to exchange IP management packets with the local TCX1000-RDM20 using the OSC signal over its line port.

In summary, this is one example of how a TCX1000 device can become unreachable. However, it demonstrates the need for redundant management communications to each TCX1000 device. We understand that in some deployments this may not always be possible. But you should be aware of the risks and take precautions. we recommend in all cases, that you provide redundant connections for management communications to the TCX1000 device.

Caution: Know the Risks When You Cannot Provide Redundant Management Communications to a TCX1000-RDM20

Caution

If there is no way to provide redundant paths for management communications to a remote TCX1000-RDM20, you can manage the device over its Line port using the OSC. In this case, the TCX1000-RDM20 is a 1D terminal and you should be aware that a break in the fiber span or a failure of the OSC signal on the remote TCX1000-RDM20 or the local TCX1000-RDM20 providing remote management connectivity to the remote TCX1000-RDM20, can isolate the remote TCX1000-RDM20 making it unreachable and requiring you to go to the site to repair the problem.

Let us look at an example of this in Figure 1. Specifically, look at the H node. At the H node, we have a single TCX1000-RDM20, which is connected through its Line port to Node D. The TCX1000-RDM20 at Node H receives its management messages over the OSC signal from the TCX1000-RDM20s at Site D. At site D we a 3-degree ROADM node, which is comprised of 3 TCX1000-RDM20s:

  • One TCX1000-RDM20 connects to site H

  • One TCX1000-RDM20 connects to site E

  • One TCX1000-RDM20 connects to site C

These three TCX1000-RDM20s can exchange traffic with each other over their pass-through connections and switch traffic in any direction.

Figure 1: Node Isolation Example
Node Isolation Example

Now let us look at what would happen to the TCX1000-RDM20 at node H if we had a break in the fiber span to site D or a failure of the OSC signal at either end of the span. There are three possible ways that site H can become unreachable and require you to go on site to correct the issue:

  • Line fiber break between sites D and H

    In this case, all traffic would down at site H and the TCX1000-RDM20 would become unreachable. You can try to diagnose the problem from site D but if you find no reason for the issue, you would need to go on site to resolve the problem.

  • OSC failure (fiber or SFP) on the TCX1000-RDM20 at site D that connects to site H

    In this case, all traffic would still be up between sites D and H but the TCX1000-RDM20 at site H would become unreachable by the proNX Optical Director. A device unreachable alarm is raised for the TCX1000-RDM20 at site H after a timeout period. However, the fact that traffic is up between sites D and H confirms the line connection, so we can deduce that they issue is with either the OSC at site D or the OSC at site H. You can try correcting the problem at site D by swapping out the OSC cable or the SFP but if it does not fix the problem, then the issue is with the OSC at site H and you would need to go on site to resolve the problem.

  • OSC failure (fiber or SFP) on the TCX1000-RDM20 at site H

    In this case, all traffic would still be up between sites D and H but the TCX1000-RDM20 at site H would become unreachable by the proNX Optical Director. A device unreachable alarm is raised for the TCX1000-RDM20 at site H after a timeout period. However, the fact that traffic is up between sites D and H confirms the line connection, so we can deduce that they issue is with the OSC between sites D and H. You can try to diagnose the problem from site D but if you find no reason for the issue, you would need to go to site H to resolve the problem.

You could solve the issue for site H by adding another TCX1000-RDM20 to sites D and H and having redundant paths at site H back to site D.

TCX1000-ILA

Note

For redundant management communication paths, you must connect the TCX1000-ILA to your DCN HA using at least two of three ports described in TCX1000 Management Architecture.

Best Practice

The best method for providing redundant management paths for the TCX1000-ILA is to connect TCX1000-ILA device to your DCN HA network is using a Layer 2 switch between the MGMT port and your DCN. In this case the device is directly connected to your DCN HA with the proNX Optical Director.

As the secondary path, the TCX1000-ILA also transmits and receives management messages over its Line ports using the OSCA and OSCB signals. This is not user configurable.