search

hashbangcode / vlad

Vlad - Vagrant LAMP Ansible Drupal
173 stars 53 forks source link

New behavior — provisioning on vagrant resume #329

Closed wizonesolutions closed 8 years ago

wizonesolutions commented 8 years ago

Vlad didn't used to re-provision on vagrant resume, but now it is. Is this a change to the Vagrantfile that I missed? What is the rationale?

dixhuit commented 8 years ago

This isn't intentional to my knowledge.

philipnorton42 commented 8 years ago

It should not provision the box entirely, but it does run the local_up.yml playbook in order to ensure that the installed services are running correctly.

wizonesolutions commented 8 years ago

No, it's running the actual provisioning process. The local_up.yml one shows up in black. The one it's running shows up in color.

On Mon, Jan 18, 2016 at 1:27 PM, Philip Norton notifications@github.com wrote:

It should not provision the box entirely, but it does run the local_up.yml playbook in order to ensure that the installed services are running correctly.

— Reply to this email directly or view it on GitHub https://github.com/hashbangcode/vlad/issues/329#issuecomment-172515450.

wizonesolutions commented 8 years ago

Just to confirm that I saw this again today when resuming, so it wasn't just one-time.

It updates hosts, "configures cache buckets," runs a shell provisioner (presumably to fix the tty thing), "configures cache buckets," and then runs the Ansible provisioner. Pretty sure Vagrant still only runs provisioning once in principle.

And then it configures cache buckets once again at the end :)

zxaos commented 8 years ago

Well, configure cache buckets is vagrant-cachier running. I'd potentially expect to see it twice, since the Vagrant script re-launches itself now to fix the .vagrant location. But three seems odd.

Can you post a full command output of this issue? Just so we can be sure we're seeing the same things?

wizonesolutions commented 8 years ago

When I get around to it. I have to sanitize the VM name, which is always a pain.

philipnorton42 commented 8 years ago

I have removed the resume trigger from the Vagrantfile. I have to say I don't tend to use the suspend/resume workflow much, opting for the halt/up commands instead so I never noticed this issue.

Thanks for pointing it out @wizonesolutions :)

wizonesolutions commented 8 years ago

Yeah, I use suspend/resume a lot. I don't think those were doing much from Vlad's side, since hostsupdater runs anyway without needing a trigger. Thanks.

On Mon, Jan 25, 2016 at 10:02 AM, Philip Norton notifications@github.com wrote:

I have removed the resume trigger from the Vagrantfile. I have to say I don't tend to use the suspend/resume workflow much, opting for the halt/up commands instead so I never noticed this issue.

Thanks for pointing it out @wizonesolutions https://github.com/wizonesolutions :)

— Reply to this email directly or view it on GitHub https://github.com/hashbangcode/vlad/issues/329#issuecomment-174442016.

wizonesolutions commented 8 years ago

@philipnorton42 This doesn't seem to be fixed. I'm up-to-date from dev but still seeing the behavior on two VMs.