Telepresence fails to kill the user's container if the container has a name set by the user

[davidmichaelkarr]

I am starting telepresence with the following sample command line:

telepresence --verbose --swap-deployment cartms-blue --docker-run -p 5005:5005 -p 8080:8080 --name=cartms -e xdebug_enable_env=t -t dockercentral.it.att.com:5100/com.att.idp/cartms:$LOGNAME-latest

This starts up fine and does what it does. At some point, I will see something like this in the output:

T: Background process (SSH port forward (socks and proxy poll)) exited with return code 255. Command was:
T:   ssh -N -oServerAliveInterval=1 -oServerAliveCountMax=10 -F /dev/null -oStrictHostKeyChecking=no -oUserKnownHostsFile=/dev/null -vv -p 46216 telepresence@127.0.0.1 -L127.0.0.1:39127:127.0.0.1:9050 -R9055:127.0.0.1:39230

T: Recent output was:
T:   debug1: channel 1: free: ::1, nchannels 2
T:   debug1: client_input_global_request: rtype keepalive@openssh.com want_reply 1
...
T:   Timeout, server 127.0.0.1 not responding.


T: Proxy to Kubernetes exited. This is typically due to a lost connection.

T: Exit cleanup in progress

However, after this, I find that there are still background processes writing to stdout. I looked at running processes, and I saw:
<uid> 19458 0.0 0.6 434812 51800 pts/14 Sl 11:43 0:00 docker run --name=telepresence-1556304181-5809424-18840 ...

After I manually killed that process, the output to stdout stopped.

[davidmichaelkarr]

telepresence.log

[ark3]

Thank you for the issue. As an experiment, could you please try your command again without passing any --name argument to Docker? I suspect Telepresence is doing something dumb with respect to named containers.

[davidmichaelkarr]

Acknowledged. I'll run some tests today. Even when it's failing this way, it sometimes takes quite a long time after starting up the process that this happens.

[davidmichaelkarr]

I haven't seen it happen yet. Is there a separate issue to address why it bungles named containers?

[ark3]

This issue will be the one.

[ark3]

To fix this:

• Always capture the container ID that gets output by docker run -d
• Pass the IDs to docker kill commands instead of names.

[donnyyung]

I believe this is no longer an issue in Telepresence 2, since you can pass docker commands to --docker-run so adding --rm should fix this problem. Here are the docs on how to install Telepresence (https://www.telepresence.io/docs/latest/install/), please re-open if you still see this issue in our latest version!