Merge pull request #11367 from davidopp/doc2

Take availability.md doc and
This commit is contained in:
Abhi Shah
2015-07-16 16:38:42 -07:00
5 changed files with 86 additions and 82 deletions

View File

@@ -471,7 +471,7 @@ You will need to run one or more instances of etcd.
- Alternative: run 3 or 5 etcd instances.
- Log can be written to non-durable storage because storage is replicated.
- run a single apiserver which connects to one of the etc nodes.
See [Availability](../admin/availability.md) for more discussion on factors affecting cluster
See [cluster-troubleshooting](../admin/cluster-troubleshooting.md) for more discussion on factors affecting cluster
availability.
To run an etcd instance: