Hello,
Not sure if this is what you are after?
static:multipush kikitux$ vagrant push db
static:multipush kikitux$ vagrant push web
static:multipush kikitux$ vagrant push
The Vagrantfile defines more than one 'push' strategy. Please specify a
strategy. Defined strategy names are:
["web", "db"]
the scripts are run locally on the box, so you can call whatever
command you require.. here I am just doing date > file
Vagrant.configure("2") do |config|
#next will run in all instances, since is config.*
config.vm.provision "shell", inline: "echo Hello"
config.vm.define "web" do |web|
web.vm.box = "hashicorp/precise64"
#next will run only on web, since is web.*
web.vm.provision "shell", inline: "echo Hello from $HOSTNAME"
end
config.vm.define "db" do |db|
#next will run only on db, since is db.*
db.vm.provision "shell", inline: "echo Hello from $HOSTNAME"
db.vm.box = "hashicorp/precise64"
end
config.push.define "web", strategy: "local-exec" do |push|
push.inline = <<-SCRIPT
date > web
SCRIPT
end
config.push.define "db", strategy: "local-exec" do |push|
push.inline = <<-SCRIPT
date > db
SCRIPT
end
end
On Sun, Mar 8, 2015 at 5:51 PM, Noah Gibbs <[email protected]> wrote:
> Hey, folks! That's much to all the Hashicorp folks and Vagrant contributors.
> I've been working more with Vagrant lately, and I'm repeatedly surprised at
> how well the API works. Appreciate it!
>
> This may be in the documentation. If so, I couldn't find it, and mea culpa.
>
> Let's say you're using multiple VMs. For example, say there are two
> (different) app servers and a DB server. We want these to be compatible with
> both VirtualBox and AWS provisioners, so we won't always have an IP address
> of 127.0.0.1 that works.
>
> When you're writing a top-level config.vm.provision or config.push block in
> your Vagrantfile, is there any way to tell which VM it's being invoked for?
> For instance, it would be nice to be able to write a top-level Chef
> provisioner that loaded the config out of a JSON file, but used a different
> JSON file for each app server.
>
> But then it would need to know which VM it was being called for, to load the
> right JSON file. Should it just be multiple provisioner blocks, one inside
> each config.vm block?
>
> Similarly for the pusher -- say you had a top-level config.push block that
> knew how to parse "vagrant ssh-info" and call Capistrano -- it would still
> need to know which app server it was being called for so that it could call
> ssh-info for the correct app server.
>
> Any hints? Or is the right answer to just put the Provisioner and Pusher
> blocks inside the config.vm blocks and maybe wind up with a bit of duplicate
> code?
>
> --
> You received this message because you are subscribed to the Google Groups
> "Vagrant" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google Groups
"Vagrant" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
