startx inconsistent behavior

I have a SSH server that is responsible for running dwm binary through X forwarding, on my client computer, i have a shell script that replaces the dwm binary on /usr/local/bin, inside that script, i simply make a call to the server requesting it to run the original dwm:

ssh -q -tt user@172.17.0.2 dwm $@

With only a single -t, i get the following error:

Pseudo-terminal will not be allocated because stdin is not a terminal

--- On the SSH server side, i'm making use of ForceCommand to pass the parameters received through a container_runner script, it looks like this:

#!/bin/sh
/usr/local/bin/$SSH_ORIGINAL_COMMAND

The reason for the ForceCommand here is to limit the possibility of running anything else other than a set of Docker container initialization scripts (located on /usr/local/bin) Here's an example of one of those scripts, the one below runs the dwm Docker image:

#!/bin/sh
docker run \
    --pull=never \
    --rm \
    -v container-scripts:/container-scripts \
    -v ssh-keys:/home/dwm/.ssh \
    -v x11-shared:/tmp/.X11-unix \
    -it \
    -e DISPLAY=:1 \
    -e XAUTHORITY=/tmp/.X11-unix/container-cookie \
    dwm \
    $@ 2>/dev/null

- container-scripts - volume containing a set of scripts that will allow running others sibling containers inside containers - ssh-keys - the SSH keys in order to be able to run containers from inside other containers - x11-shared - X11 shared data (X11 socket and xauth cookie) --- My ssh_config looks like this:

StrictHostKeyChecking no
UserKnownHostsFile /dev/null
ForwardX11 yes
ForwardX11Trusted yes
PreferredAuthentications=publickey

Things started to get inconsistent after i disabled SSH multiplexing from my ssh_config, previously with the settings below, it worked fine all the time (which makes no sense, as far as i know, SSH multiplexing shouldn't interfere in that), settings below:

ControlPath /tmp/%r@%h:%p
ControlMaster auto
ControlPersist yes

What happens is that whenever i try to run startx passing as argument my dwm script that makes a request to the SSH server to run the actual dwm, i just get a black screen, and nothing happens, it just stays like that, example below:

startx /usr/local/bin/dwm -- :1

But, if i run the same command above, slightly differently, using shell command substitution alongside with the shell noop operator, it works fine (just takes a few more seconds to show dwm than usual, probably due to the ugly hack)

: $(startx /usr/local/bin/dwm -- :1)