You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The clients should be connected to the consul server cluster. With the current code, if you ssh to a server and run consul members, you only see the servers, and if you ssh to a client and run consul members, you only see the clients.
This means the clients and apps that talk to them would not have access to the servers.
A workaround for this is to modify the run-consul script to have the following at line 201:
and to also change the user_data_client template_file data source at line 138 to have scale_set_name = "${var.cluster_name}-server" so that both the client and server scripts point at the server VM scale set.
After making these changes, rebuilding my packer image, re-provisioning the virtual machine scale sets, ssh-ing to the server or client VMs, and running consul members, I see both servers and clients:
consuladmin@consul000000:~$ consul members
Node Address Status Type Build Protocol DC Segment
6151e229-8c48-49b7-b9d3-5cbf778c8ad3 10.0.10.6:8301 alive server 1.0.3 2 eastus <all>
7b861f7d-0598-4402-b4dc-d5306328ee18 10.0.10.5:8301 alive server 1.0.3 2 eastus <all>
d3f6866f-f3f0-40e8-bd5a-77227bba118a 10.0.10.4:8301 alive server 1.0.3 2 eastus <all>
3a59a069-a755-4b85-b634-204e70168f08 10.0.10.8:8301 alive client 1.0.3 2 eastus <default>
aacd535b-a82a-4822-a367-ecd128ee5ea9 10.0.10.7:8301 alive client 1.0.3 2 eastus <default>
The text was updated successfully, but these errors were encountered:
cori
added a commit
to cori/terraform-azurerm-consul
that referenced
this issue
Feb 28, 2018
The clients should be connected to the consul server cluster. With the current code, if you ssh to a server and run
consul members
, you only see the servers, and if you ssh to a client and runconsul members
, you only see the clients.This means the clients and apps that talk to them would not have access to the servers.
A workaround for this is to modify the run-consul script to have the following at line 201:
and to also change the user_data_client template_file data source at line 138 to have
scale_set_name = "${var.cluster_name}-server"
so that both the client and server scripts point at the server VM scale set.Note that the first change uses Consul's cloud auto-joining feature for Azure: https://www.consul.io/docs/agent/options.html#microsoft-azure.
After making these changes, rebuilding my packer image, re-provisioning the virtual machine scale sets, ssh-ing to the server or client VMs, and running
consul members
, I see both servers and clients:The text was updated successfully, but these errors were encountered: