my host machine is windows.I have vagrant there.now i want to use ansible as a provisioner.how can i use this
what does vagrant-hostsupdater plugin does
what does this plugin do vagrant-hostsupdater??
i only know vagrant-hostmanager (and just ditched it :) )
does it make any good to re-use boxes in multiple VM (like Docker) or not ?
I'm using the latest ubuntu/trusty64 box and provisioning with Puppet Apply. I'm a bit new to puppet and noticed the box has 3.4 installed. Any ideas if puppet baked into the box, or is vagrant installing it for me? There is no description on the boxes Atlas page.
If anyone has any information, it would be really appreciated!
Hi, I have a vagrant file with multiple vm definitions. Can I set specific memory size for each of them (I use virtualbox)?
I found https://www.vagrantup.com
/docs/virtualbox/configuration.html but "config.vm.provider" seems to be a global setting to all VMs?
Ok, I think I should use "node.vm.provider" instead.
is there a quick command to find IP of a machine, and exposed ports ?
something like `docker inspect` for docker...
strk: you mean using vboxmanage? Yes.
vboxmanage list bridgedifs
You could also parse it out of showvminfo
vboxmanage showvminfo themachine | grep <the-known-ip-address> # gives no hits
found <themachine> via vboxmanage list runningvms
and took the first word
VBoxManage guestproperty enumerate <uuid>
This works for me: VBoxManage guestproperty enumerate <uuid> | grep 'Net/0'
uhm, not really mnemonic :P
other question: is it normal that if provisioning fails, next `vagrant up` doesn't show any error ?
I mean, the machine is left in non-fully provisioned state
strk: Did adding debug and verbosity show any errors? Did you try with --no-provision and see if some process left it in an inconsistent state? Some race condition?
setuid: the provisioning is just not run on second "vagrant up" call
Right, but you don't yet know why it's breaking, you have to instrument the process to figure out where it's dying
I know why it's breaking: I gave wrong vault password
but what I'm asking is: it is normal for "vagrant up" to still succeed after the first failure ?
because it sounds wrong to me
the Vault passowrd is always wrong, but on the first run "provisioning" is run and fails, while on the secodn run "provisioning" is not run at all (or this is my impression) so things just succeed
resulting in a silent failure
if you run a vagrantfile including provisioners without --provision, it actually warns about this at startup
I've got an interesting issue I'm trying to work out: I'm bootstrapping 4 VMs in a row, all of them 90% identical, except the final cookbooks that install some openstack-specific kit on each one. Is there an easier way to bring these up, maybe as clones of each other, so I can then just add the last-mile packages as deltas?
IOW, each of the 4 nodes does its own 'apt-get update && apt-get upgrade' within it, but they all start from the same trusty64 box
It seems wasteful and redundant
Any thoughts? Ideas?
I was thinking something with rsync-auto might help
Hi… How’s it going?
I’m looking at the —machine-readable interface and slightly confused
I’m running vagrant global-status —machine-readable, and finding no recognizable type name in the output
anyone awake in here?
Hi… I’m looking at —machine-readable output and seem to have found some undocumented tyops