Unable to update configuration when the cluster is stopped

[BusyJay]

Bug Report

Please answer these questions before submitting your issue. Thanks!

1. What did you do?
   I deploy a cluster that uses patched version TiKV that adds new configurations. So I:

• deploy the cluster with default packages;
• patch the cluster with my packages offline;
• edit-config to adjust new configurations
• reload the cluster to make the configuration takes affect.

2. What did you expect to see?
   The cluster uses new configuration to start.

3. What did you see instead?
   I can't reload the cluster as it's not started. I have to start it and then reload it. To make it worse,
   the new configuration must be changed before the cluster is initialized, so I can't start it first.

Or, I should be able to deploy the cluster with the patched version in the very beginning.

4. What version of TiUP are you using (tiup --version)?

1.5.3 tiup
Go Version: go1.16.5
Git Ref: v1.5.3
GitHash: 222bd4d460eaa3fb6feec6ddeb8612fe0870ffd3

[AstroProfundis]

You may try reload with --skip-restart, and maybe also --ignore-config-check

[BusyJay]

You may try reload with --skip-restart

It doesn't work.

{"level":"warn","ts":"2021-08-05T16:52:45.206+0800","logger":"etcd-client","caller":"v3@v3.5.0/retry_interceptor.go:62","msg":"retrying of unary invoker failed","target":"etcd-endpoints://0xc000178000/#initially=[127.0.0.1:2379]","attempt":0,"error":"rpc error: code = DeadlineExceeded desc = latest balancer error: last connection error: connection error: desc = \"transport: Error while dialing dial tcp 127.0.0.1:2379: connect: connection refused\""}

Error: context deadline exceed