From a customer [feedback](https://gitlab.com/gitlab-org/gitlab/-/issues/410031#note_1587484019) > I also decided to follow the [instructions](https://docs.gitlab.com/ee/user/workspace/configuration.html), but again ended up with strange errors: After I added “remote_development” section to the agent configuration, its log started to show: > > ``` > {"level":"info","time":"2023-10-02T12:02:59.171Z","msg":"starting full sync","mod_name":"remote_development","agent_id":1} > {"level":"error","time":"2023-10-02T12:02:59.347Z","msg":"Remote Dev - full sync cycle ended with error","mod_name":"remote_development","error":"unexpected status code: 403","agent_id":1} > {"level":"info","time":"2023-10-02T12:03:09.171Z","msg":"starting partial update","mod_name":"remote_development","agent_id":1} > {"level":"error","time":"2023-10-02T12:03:09.203Z","msg":"Remote Dev - partial sync cycle ended with error","mod_name":"remote_development","error":"unexpected status code: 403","agent_id":1} > ``` From the related [ZD Ticket](https://gitlab.zendesk.com/agent/tickets/454549) > > {"level":"error","ts":1696249083.5287197,"caller":"sshproxy/server.go:79","msg":"failed to create SSH connection","error":"EOF","stacktrace":"gitlab.com/remote-development/gitlab-workspaces-proxy/pkg/sshproxy.(*sshProxy).handleSSHConnection\n\t/app/pkg/sshproxy/server.go:79"} > > > > {"level":"error","ts":1696249083.5287933,"caller":"sshproxy/server.go:190","msg":"error closing incoming connection","workspace_name":"unknown","error":"close tcp 10.1.32.15:22->10.170.0.10:2958: use of closed network connection","stacktrace":"gitlab.com/remote-development/gitlab-workspaces-proxy/pkg/sshproxy.(*sshProxy).closeConnectionAndLogError\n\t/app/pkg/sshproxy/server.go:190\ngitlab.com/remote-development/gitlab-workspaces-proxy/pkg/sshproxy.(*sshProxy).handleSSHConnection\n\t/app/pkg/sshproxy/server.go:80"} > > I also see a continuous stream of these sshproxy errors in the proxy log. I believe the reference to gitlab.com is just from the stacktrace showing the source code file details. The errors appear to relate to the proxy pod closing connections from the cluster nodes on port 22. I'm not sure why those connections are being attempted but I suspect the errors can be ignored. They don't appear to affect access to the workspace via SSH.