PID Judge comes too fast results in premature kill.

I've came to a pretty intriguing bug up here:

At first running containers failed with the following error:

x11docker ERROR: start_container(): Did not receive PID of PID1 in container.
  Maybe the container immediately stopped for unknown reasons.
  Just in case, check if host and image architecture are compatible:
  Host architecture: amd64 (x86_64), image architecture: amd64.

After enabling verbose I found that the script repeatedly checks if any PID is received.

After some attempts I got one running successfully out of pure luck.

I found out AFTER the window appeared and BEFORE anything shows up in the window, if I suspend the x11docker process for some time, during which I guess docker is starting, and I run fg to bring the process back, the container can startup normally.

Is there an option to manually adjust the timeout?

I guess we should use some other mechanisms to determine if a docker container have been started up or not, instead of using PID and a timeout?

There is no option yet to adjust the timeout.
However, you could change the script at

x11docker/x11docker

Line 8345 in cb29a99

for ((Count=1 ; Count<=40 ; Count++)); do

Instead of 40 (4 seconds) change e.g. to 100 (10 seconds).

I guess we should use some other mechanisms to determine if a docker container have been started up or not, instead of using PID and a timeout?

It is not a nice way, but the only one I found so far.

So sad that I got this from a package manager. I'm pretty much unable to write to such scripts manually......

Actually I don't think this issue is even prevalent or whatsoever. I have a decent desktop, wonder why docker started so slow......

Currently I'm doing it the crude way and it worked quite well......