updated service upstream definition (#7972)
This commit is contained in:
parent
0122aba44d
commit
77a63c6855
1 changed files with 1 additions and 1 deletions
|
@ -576,7 +576,7 @@ By default the NGINX ingress controller uses a list of all endpoints (Pod IP/por
|
|||
|
||||
The `nginx.ingress.kubernetes.io/service-upstream` annotation disables that behavior and instead uses a single upstream in NGINX, the service's Cluster IP and port.
|
||||
|
||||
This can be desirable for things like zero-downtime deployments as it reduces the need to reload NGINX configuration when Pods come up and down. See issue [#257](https://github.com/kubernetes/ingress-nginx/issues/257).
|
||||
This can be desirable for things like zero-downtime deployments . See issue [#257](https://github.com/kubernetes/ingress-nginx/issues/257).
|
||||
|
||||
#### Known Issues
|
||||
|
||||
|
|
Loading…
Reference in a new issue