Machine: Virtualbox machine fails with `tls: DialWithDialer timed out` when checking certificates

[lox]

I get the following error after docker-machine start mymachine after it's been stopped some time (hard to pin down exact cause, can't force reproducing it with a machine sleep or network change). Subsequent runs do not have this error:

Starting "99dev"...
(99dev) Check network to re-create if needed...
(99dev) Waiting for an IP...
Machine "99dev" was started.
Waiting for SSH to be available...
Detecting the provisioner...
Started machines may have new IP addresses. You may need to re-run the `docker-machine env` command.
Error checking TLS connection: Error checking and/or regenerating the certs: There was an error validating certificates for host "192.168.99.100:2376": tls: DialWithDialer timed out
You can attempt to regenerate them using 'docker-machine regenerate-certs [name]'.
Be advised that this will trigger a Docker daemon restart which will stop running containers.

At a guess, is it possible that there is a race condition occurring where the docker daemon is slow to start but the machine is reporting it's ip as ready?

Version

docker-machine version 0.6.0, build e27fb87`

VirtualBox Version

5.0.14r105127

[barnabyc]

I'm experiencing the same issue, same versions.

Following the advice of regenerating the certs also has no effect.

I have not been able to start my instances since upgrading with DockerToolbox from 1.9.1g to 1.10.1 last night...

[barnabyc]

Update: After trying the same steps I've tried several times, it now works. I did not machine any other changes, all Id id was: stop default, ensure it was stopped via status, start default, then check env - this time no error and I'm able to connect.

Race condition? Resource issues? No clue.

[alexander-lazarov]

This happened to me too, after upgrading boot2docker image. The workaround is the following:

• open VirtualBox and stop the machine (send ACPI Shutdown)
• run docker-machine start machine-name in a new shell session. Old ones won't work (I guess there is something left in the environment that breaks even docker-machine env)

[beorn]

I have the same problem, but it doesn't work even if I try to do it in a separate shell after a ACPI shutdown.

[wmcg]

After experiencing this problem and trying ACPI shutdown, separate shell, reupgrading to no avail. I removed and recreated the VM which, unsurprisingly, fixed the issue.

[tobilarscheid]

@alexander-lazarov: this worked for me!
I am running Docker version 1.11.0, build 4dc5990 on my Mac.

[r4j4h]

I had to recreate things as @wmcg described to get things going from 1.9.1 => 1.11.1 on my Mac running El Capitan v10.11.23

[MehrCurry]

I had success with doing a docker-machine kill default and then starting normally again.

[rogaha]

It happened with me too.

docker-machine version 0.8.2, build e18a919

[ldshi]

I am encountering this also.

docker-machine version 0.8.1, build 41b3b25

[omnific9]

I'm encountering this too.

docker-machine version 0.9.0, build 15fd4c7

I've since killed my docker-machine, uninstalled docker, reinstalled docker, reinstalled virtualBox, restarted my computer, and tried "docker-machine regenerate-certs" multiple times. I've also tried using different IP addresses and moved my laptop from one network to another. This continues to happen. It's not random either. It happens every time.

Please help. I'm desperate.

[anmolnagpal]

Getting the same issue.

docker-machine version 0.10.0, build 76ed2a6