[Q&A] Client unable to connect to server under real-world mode #2183
-
Python version (
|
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 6 replies
-
|
I think the issue is because your machine where the client (site-1) could not resolve the server DNS name "server1". You can manually add this item to your "/etc/hosts" on the client machine. You can use command "ping server1" to check if your client machine can reach the server from the network layer. Or you can use the CLI command to directly launch server and clients. The CLI command will automatically take care of the network setup between the server and clients. |
Beta Was this translation helpful? Give feedback.
-
|
Could you specify the CLI command to directly launch the server and clients? Thanks. |
Beta Was this translation helpful? Give feedback.
-
|
I tried renaming server1 in project.yml to my workstation hostname, which I can ping. I then recreated the startup kit with the updated project.yml. The server folder was now named after the hostname. And sever was started as before. However when starting site-1, it still tried to connect server1 and give same error. I noticed in site-1/startup/fed_client.json, there is a line Likewise, in admin@email.com/startup/fed_admin.json there is a line I also tried changing the same line in site-1/startup/fed_client.json, and starting site-1. I then got warning about "insecure fed_client.json", and site-1 still won't start. |
Beta Was this translation helpful? Give feedback.
-
|
@jingyunc thank you for your reply and interest. You will need to modify one more place in the project.yml. The "sp_end_point: server1:8002:8003" Let me illustrate below, please find " api_version: 3
name: example_project
description: NVIDIA FLARE sample project yaml file
participants:
# change example.com to the FQDN of the server
- name: ***server1***
type: server
org: nvidia
fed_learn_port: 8002
admin_port: 8003
- name: site-1
type: client
org: nvidia
# listening_host will enable creating one pair of cert/private key for this client
# so it can behave like a server for client api. The value must be a hostname that
# client api can reach via network.
# listening_host: site-1-lh
- name: site-2
type: client
org: nvidia
- name: admin@nvidia.com
type: admin
org: nvidia
role: project_admin
# The same methods in all builders are called in their order defined in builders section
builders:
- path: nvflare.lighter.impl.workspace.WorkspaceBuilder
args:
template_file: master_template.yml
- path: nvflare.lighter.impl.template.TemplateBuilder
- path: nvflare.lighter.impl.static_file.StaticFileBuilder
args:
# config_folder can be set to inform NVIDIA FLARE where to get configuration
config_folder: config
# scheme for communication driver (currently supporting the default, grpc, only).
# scheme: grpc
# app_validator is used to verify if uploaded app has proper structures
# if not set, no app_validator is included in fed_server.json
# app_validator: PATH_TO_YOUR_OWN_APP_VALIDATOR
# when docker_image is set to a docker image name, docker.sh will be generated on server/client/admin
# docker_image:
# download_job_url is set to http://download.server.com/ as default in fed_server.json. You can override this
# to different url.
# download_job_url: http://download.server.com/
overseer_agent:
path: nvflare.ha.dummy_overseer_agent.DummyOverseerAgent
# if overseer_exists is true, args here are ignored. Provisioning
# tool will fill role, name and other local parameters automatically.
# if overseer_exists is false, args in this section will be used and the sp_end_point
# must match the server defined above in the format of SERVER_NAME:FL_PORT:ADMIN_PORT
#
overseer_exists: false
args:
sp_end_point: ***server1***:8002:8003
- path: nvflare.lighter.impl.cert.CertBuilder
- path: nvflare.lighter.impl.signature.SignatureBuilder
|
Beta Was this translation helpful? Give feedback.
-
|
@chesterxgchen @IsaacYangSLA ^ this is a valid use case, some people will be changing this file for their server name, should we introduce a common variable in the yaml file so people just need to modify one place? |
Beta Was this translation helpful? Give feedback.
-
|
Confirming the issue was solved after renaming the 2nd server1 in project.yml. Many thanks :) |
Beta Was this translation helpful? Give feedback.
-
|
Since you are just running on the same host, using the project.yaml, instead manually do this and try to remember change the /etc/host name for the match server etc. you can simply use the new POC command the POC CLI command nvflare poc -i project.yml ( check the help) will create the startup the structure based on the project.yml but the actual host and corresponding certificates are using localhost so you don't need to truggle with this. |
Beta Was this translation helpful? Give feedback.
@jingyunc thank you for your reply and interest.
You will need to modify one more place in the project.yml. The "sp_end_point: server1:8002:8003"
Let me illustrate below, please find "
***server1***" and substitute with your correct server name.You also need to make sure each client site can find server by its name.