This is only necessary for manual instllations. setup.sh script takes care of this for you.
Add the following in your .bashrc, .zshrc, etc. file to automate the environment variable export:
# Docker (default for Vagrant based boxes)
export DOCKER_HOST=tcp://localhost:2375
If you also have $(boot2docker shellinit)
there, then make sure those lines go BEFORE it, e.g.:
# Docker (default for Vagrant based boxes)
export DOCKER_HOST=tcp://localhost:2375
# boot2docker shellinit
$(boot2docker shellinit)
This way if boot2docker is NOT running, your DOCKER_HOST
will default to tcp://localhost:2375
.
Otherwise $(boot2docker shellinit)
will overwrite the variables and set DOCKER_HOST
to point to the boot2docker VM.
Vagrant can be controlled (e.g. vagrant up
, vagrant ssh
, vagrant reload
, etc.) from the root directory of the Vagrantfile as well as from any subdirectory. This is very usefull when working with multiple projects in subdirectories.
Add this to your user settings (Sublime Text > Preferences > Settings - User):
{
"atomic_save": false
}
ST3 does not update the ctime extended file attribute when saving a file. This leads to NFS not seeing the changes in a file unless the file size changes as well (i.e. changing a single symbol in a file with ST3 will not be visible over NFS). The setting above fixes that.