diff --git a/docs/source/design/hadr/decisions/near-term-target.md b/docs/source/design/hadr/decisions/near-term-target.md index 919a8ea8aa..92b7fd1147 100644 --- a/docs/source/design/hadr/decisions/near-term-target.md +++ b/docs/source/design/hadr/decisions/near-term-target.md @@ -27,15 +27,15 @@ Designing for high availability is a complex task which can only be delivered ov #### Advantages -1. Flushes out lots of basic deployment issues that will be ofbenefit later. +1. Flushes out lots of basic deployment issues that will be of benefit later. 2. If stuff slips we at least have a backstop position with hot-cold. -3. For now, the only DR story we have is essentially a continuationof this mode +3. For now, the only DR story we have is essentially a continuation of this mode 4. The intent of decisions such as using a loadbalancer is to minimise code changes #### Disadvantages 1. Distracts from the work for more complete forms of HA. -2. Involves creating a few components that are notmuch use later, for instance the mutual exclusion lock. +2. Involves creating a few components that are not much use later, for instance the mutual exclusion lock. ## Recommendation and justification @@ -43,4 +43,4 @@ Proceed with Option 2: Hot-cold. ## Decision taken -**[DRB meeting, 16/11/2017:](./drb-meeting-20171116.md)** Adopt option 2: Near-term target: Hot Cold (RGB, JC, MH agreed) \ No newline at end of file +**[DRB meeting, 16/11/2017:](./drb-meeting-20171116.md)** Adopt option 2: Near-term target: Hot Cold (RGB, JC, MH agreed)