What is the Wait-to-Restore Period within a Protection-Switching System?
The purpose of this post is to describe and define the Wait-to-Restore period within a Revertive Protection-Switching system.
All Protection-Groups will perform Protection-Switching, to route the Normal Traffic Signal around a defective Working Transport entity, anytime it is declaring a service-affecting defect with that Working Transport entity.
In other words, the Protection-Group will route the Normal Traffic Signal through the Protection Transport entity, for the duration that it is declaring this defect condition.
Whenever a Revertive Protection-Group clears that defect, then it will switch the Normal Traffic Signal back to flowing through the Working Transport entity.
We call this second switching procedure (to return the Protection-Group to its NORMAL, pre-protection-switching state), revertive switching.
In contrast, a Non-Revertive Protection-Group will NOT perform this revertive switch, and the Normal Traffic Signal will continue to flow through the Protection Transport entity for an indefinite period.
When the Tail-End Node clears the Service-Affecting Defect
Protection-Switching events are very disruptive to the Normal Traffic Signal. Each time we perform a protection-switching procedure, we are inducing a glitch (or a burst of bit-errors) within the Normal Traffic Signal
Therefore, Protection-Switching events should not be a common occurrence within any network.
To minimize the number of protection-switching events (occurring within a network), the Protection-Group will usually force the Tail-End Node to go through a Wait-to-Restore period, after it clears the service-affecting defect (which caused the Protection-Switching event in the first place) before it can proceed on to the next step and revert the protection-switching (and traffic).
In other words, the Tail-End Node (within a Protection-Group) will execute the following set of steps, each time it clears a service-affecting defect, which caused a protection-switching event.
- It clears the defect condition.
- The Tail-End circuit will then start a Wait-to-Restore Timer, and will wait until this timer expires before it proceeds to the next step.
- If the Tail-End circuit declares another service-affecting defect, while it is waiting for this Wait-to-Restore timer to expire, then it will reset this timer back to zero and will continue to wait.
- Once the Wait-to-Restore timer expires, then the Tail-End circuit will proceed to revert the protection-switched configuration into the NORMAL configuration.
I show these same steps within the Revertive Procedure Flow-Chart below.
Figure 1, Flow-Chart of the Revertive Protection-Switching Procedure – after the Service-Affecting defect clears
What is the purpose of using this Wait-to-Restore Period?
There are two main reasons what we use the Wait-to-Restore period in a Protection-Switching system.
- To make sure that the condition of the Working Transport entity has stabilized and is not declaring intermittent defects before we start to pass the Normal Traffic signal through it again.
- And, to reduce the number of protection-switching events within a protection-group.
How Long should the Wait-to-Restore period be?
ITU-T G.808.1 recommends that this period be between 5 and 12 minutes.
All revertive protection-switching systems must wait through a Wait-to-Restore period (after they have cleared the defect condition) before executing the actual revertive switch.
The purpose of waiting through this Wait-to-Restore period is to prevent the occurrence of multiple Protection-Switching events, due to the intermittent occurrence of defects within the Working Transport entity.
ITU-T G.808.1 recommends that this Wait-to-Restore period be between 5 and 12 minutes.
All of this means that the Tail-End circuit must go through this Wait-to-Restore period, and declare no defects for the entire 5 to 12 minute period, before it can move on to revert its protection-switching.