-
|
Current setup I'm working with. I have my manager arcane instance running on a standalone vm. Now trying to setup node-level agents for swarm. By default they're trying to talk to the main manger node. However failing because it dosn't have the node-level api keys. which were generated by the swarm-level node. When I switch to point to the manager URL to the swarm-level node, I get back 404 when trying to create the tunnel. Is what I'm doing even possible? or can agents only home back to the manager instance? or should both the manager and swarm-level agent be sharing a DB? or should the swarm-level db be in edge mode? I have a feeling this falls more into feature request then bug fix. So wanted to ask to help me decide where I want to go from here. |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment
-
|
Running the repo through deepwiki, looks like this is an architectural limitation |
Beta Was this translation helpful? Give feedback.
Running the repo through deepwiki, looks like this is an architectural limitation
https://deepwiki.com/search/for-docker-swarm-you-need-node_bcdc3853-a0fd-4b6f-a8c0-ef2636cf7271?mode=fast