When performing a ping on my host machine to Google’s DNS server 8.8.8.8, I get 30-60ms. From within a container pinging 8.8.8.8, I expect the same or slower ping responses.
From within the container, when I ping 8.8.8.9, an IP address that does not exist, the ping should report 100% packet loss.
docker run --rm -it alpine ping -c 1 8.8.8.8 docker run --rm -it alpine ping -c 1 8.8.8.9
Actual behavior
Pings to both 8.8.8.8 and 8.8.8.9 from within the container respond with lightning fast response times (under 1ms).
Information
Host: OS X El Capitan (10.11.4)
Docker: 1.11.1-beta10 (build 6662)
Network: wifi
Container Images tested:
Thanks for the report… we were waiting for someone to notice this
It’s being tracked on issue (#1925) on our tracker and we have a fix in the pipeline.
I’m hoping it will be available in the next 2-3 weeks!
Keep the feedback coming, and thanks for trying Docker for Mac!
any update to this?
running 1.12.0-rc4-beta20 – and the error still persists as of 7/27.
this is particularly an issue as i’m using a toolset that depends on certain pings failing, which never happen.
Same behaviour here, with the final release of Docker for Mac. I can’t use Docker for Mac with tools that rely on ping to detect presence of devices on my network.
I can see the same behaviour with Docker for Windows. When the fix comes is it going to be only for Docker for Mac? (since the github path would suggest that).