You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Feb 11, 2022. It is now read-only.
As mentioned in #385 , a previous commit has removed the SyncedFolders method on provision, which has caused some stress.
We use a forked version of this repo using spot instances for our development to save some money. I have a chef script that uses the current server's IP address to assign an A record in route 53 (because load balancing at $0.028ph is too expensive for our development needs)
ip_address = open('http://169.254.169.254/latest/meta-data/public-ipv4', options = { proxy: false }, &:gets)
unless node['load_balancer'].nil? || node['load_balancer'].empty?
aws_elastic_lb node['load_balancer'] do
aws_access_key aws_access_key_id
aws_secret_access_key aws_secret_access_key
name node['load_balancer']
action :register
end
end
unless node['route_53_domain'].nil? || node['route_53_domain'].empty?
node['route_53_domain'].each do |domain,subdomain|
# Create a zone for example.com
amazon_dns_zone domain do
aws_access_key_id aws_access_key_id
aws_secret_access_key aws_secret_access_key
end
amazon_dns_record subdomain do
domain domain
value ip_address
type "A"
ttl 60
aws_access_key_id aws_access_key_id
aws_secret_access_key aws_secret_access_key
end
end
end
This needs to be run on the AWS server to test, but without syncing the changes on provision, and the fact that a spot instance can't be stopped and started again, I'm having to scrap the server every time to test this.
The text was updated successfully, but these errors were encountered:
3vcloud
changed the title
Vagra
Vagrant provision no longer updating chef cookbooks
Nov 23, 2015
As you've probably worked out this is a long running issue and actually impacts multiple plug-ins for vagrant, not just AWS.
There's been little movement to get the problem fixed, although it needs to be fixed upstream (something like hashicorp/vagrant#5850).
My suggestion in the meantime (and maybe permanently because it works well) would be to move your test over to Test Kitchen. Test Kitchen + kitchen-ec2 has worked flawless since the last few versions for me.
Thanks for the feedback @yoshiwaan . I'm also getting feedback from the team about issues with the AWS machines losing reference to the synced folders altogether - the fix for this normally being to delete synced_folders and re-provision - but would still give a big issue if we had to reload a production server in this scenario.
Either way, definitely looks like a bug with vagrant itself and not this plugin, so I'll close this one off and subscribe to your other issue - thanks again 👍
Hi,
As mentioned in #385 , a previous commit has removed the SyncedFolders method on provision, which has caused some stress.
We use a forked version of this repo using spot instances for our development to save some money. I have a chef script that uses the current server's IP address to assign an A record in route 53 (because load balancing at $0.028ph is too expensive for our development needs)
This needs to be run on the AWS server to test, but without syncing the changes on provision, and the fact that a spot instance can't be stopped and started again, I'm having to scrap the server every time to test this.
The text was updated successfully, but these errors were encountered: