-
Notifications
You must be signed in to change notification settings - Fork 51
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Document double nat scenario for OpenStack deployments #806
Changes from 1 commit
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -356,9 +356,29 @@ NAT, or if other applications on the internal network were already using the 450 | |
ports, the remote ends would not be able to contact gateway C or D over the expected ports. | ||
{{% /notice %}} | ||
|
||
### Double NAT Traversal | ||
### Double NAT Traversal (scenario 1) | ||
|
||
In this case case, A & B cluster gateways have direct reachability over their private | ||
In this case, clusters C & D are neither reachable on their private IPs (192.168.0.4 and 192.168.0.4) nor | ||
on the public IP. However, they are reachable over the private floating IPs (10.2.0.1 and 10.2.0.2). | ||
Submariner cannot detect these private floating IPs. To get the connectivity working, you can annotate | ||
the Gateway node with the private floating IP as shown below. | ||
|
||
![on-premises](/images/natt/double_nat_scenario1.png) | ||
|
||
```bash | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Q : the annotated public IP (e.g: 10.2.0.1) will be used as EP's public IP for all connections (including to clusters C,D) , can you elaborate how the connection from cluster A to C will work ? There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. There is no Cluster A in this topology. Please look at the next scenario and it is currently not supported in Submariner. There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Ack, 10x |
||
kubectl annotate node $GWC gateway.submariner.io/public-ip=ipv4:10.2.0.1 | ||
kubectl annotate node $GWD gateway.submariner.io/public-ip=ipv4:10.2.0.2 | ||
|
||
# restart the gateways to pick up the new setting | ||
for cluster in C D; | ||
do | ||
kubectl delete pod -n submariner-operator -l app=submariner-gateway --kubeconfig $cluster | ||
done | ||
``` | ||
|
||
### Double NAT Traversal (scenario 2) | ||
sridhargaddam marked this conversation as resolved.
Show resolved
Hide resolved
|
||
|
||
In this case, A & B cluster gateways have direct reachability over their private | ||
IPs (10.0.0.1 and 10.1.0.1) possibly with large MTU capabilities, while between cluster | ||
C and D (192.168.0.4 and 192.168.0.4 too), reachability over the private IPs is not possible | ||
sridhargaddam marked this conversation as resolved.
Show resolved
Hide resolved
|
||
but it would be possible over the private floating IPs 10.2.0.1 and 10.2.0.2. However Submariner | ||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Shouldn't it be 192.168.128.4 ?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why? The diagram uses 192.168.0.4
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Private IPs are the node-ips. As per the diagram its 192.168.0.4 (deliberately shown the same IPs as its possible in such deployments) and the private floating-ips are 10.2.0.1 and 10.2.0.2 respectively. The public-ip auto-discovery for such clusters would be the same for both (2.1.1.1).
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ooops, I missed the diagram