330 karma · joined May 4, 2012
I'm very impressed that you were able to build in 3 weeks time a low-latency multi-data center application with master/slave database failover, robust fault tolerant load balancing, and backups that can be restored in minutes with an API to control all of those services. That would normally take a senior team of engineers several months to accomplish and have it be of the quality and reliability of the services provided by AWS.
More likely is that you had a use case or a mindset that did not suit AWS very well and was easy to implement on your own. That's awesome and I'm glad you were able to find better value elsewhere. AWS is not for everyone, and is definitely quite expensive on the pocketbook.
For some businesses, the huge AWS feature set (RDS, EBS, ELB, security groups, VPC, ELB, EIPs, etc, etc, etc) is more valuable than the bottom $$ line. For others, those features aren't worth the added cost, but hand-wavey "just use DO" or "Just rent physical servers from SoftLayer/Rackspace" is disingenous.
TL;DR AWS has value above and beyond simply hour-to-hour elasticity.
Unless they have dramatically improved their offering in the last couple years, an hour or two from "I need a new server" to delivery is 1) not an accurate timeframe for physical servers from Rackspace and 2) even if it was realistic, that's an eternity when you are trying to iterate quickly. I can have a new server in 30 seconds with AWS and in the course of an hour could have tested my automation tools half a dozen times or more vs reimaging a server over and over again.
I'm not saying it's always the right choice or that it's cheaper, but that flexibility combined with some of the pre-canned tools (ELB, RDS, CloudWatch, SQS, SNS) has tremendous value even when you aren't autoscaling.
I agree that long-term contract-to-hire is a reasonable way to go for may employers and employees, particularly folks that already are already contractors, but the premise of "you've got a couple weeks to prove yourself" is a non-starter. I'm not going to quit my day job and risk my family's financial security for a company that is unwilling to take a similar risk on me. In an ideal world, it is the best way to evaluate a candidate. However, if it were practical, it would have already become an industry practice. The fact that it seems to be rare indicates to me that it does not work very well.
As a Systems Engineer, a practice that I would like to see more of (and may try to implement at the company I work for) is a brief troubleshooting exercise with a shared tmux session. Set up a server with some problem and ask the candidate to identify the issue and try to resolve it.
I like your idea about contributing to some open source project (paid), though even in that case, you will likely run into qualified people who are unwilling (or legally unable) to do this and as an interviewer, you need to be careful not to exclude those individuals.
Ok. Let's see the code to do that.
I needed to spin up 5 servers, a load balancer and 2 groups of 2 backend servers. I needed to provision them with Chef, have full name resolution, and be able to access the servers from my host machine.
Here's my Vagrantfile
Vagrant.require_plugin "vagrant-chef-zero"
Vagrant.require_plugin "vagrant-berkshelf"
Vagrant.require_plugin "landrush"
last_ip_octet = 10
DOMAIN = "vagrant.dev"
Vagrant.configure("2") do |config|
config.landrush.enable
config.chef_zero.chef_repo_path = "./"
config.omnibus.chef_version = '11.8.0'
config.vm.box = "opscode_ubuntu-12.04_provisionerless"
config.vm.box_url = "https://opscode-vm.s3.amazonaws.com/vagrant/opscode_ubuntu-12.04_provisionerless.box"
%w{default metrics}.each do |backend|
(1..2).each do |i|
config.vm.define "be_#{backend}_#{i}".to_sym do |cfg|
last_ip_octet += 1
cfg.vm.hostname = "be-#{backend}-#{i}.#{DOMAIN}"
cfg.vm.network :private_network, :ip => "192.168.33."+last_ip_octet.to_s
cfg.vm.provision :chef_client do |chef|
chef.node_name = "be-#{backend}-#{i}"
chef.add_role "backend_#{backend}"
end
end
end
end
config.vm.provision :shell, :inline => "/usr/bin/apt-get update --fix-missing"
config.vm.define :lb do |cfg|
last_ip_octet += 1
lb_ip = "192.168.33."+last_ip_octet.to_s
cfg.vm.hostname = "lb.#{DOMAIN}"
cfg.vm.network :private_network, :ip => lb_ip
cfg.vm.provision :chef_client do |chef|
chef.node_name = "lb"
chef.add_role "lb"
end
end
end
This utilizes a few vagrant plugins, vagrant-chef-zero, vagrant-berkshelf, and landrush which are one liners each to install. After that, I can type vagrant up # build my stack from scratch
vagrant destroy # destroy my stack
vagrant ssh lb # ssh into the lb instance
vagrant provision # execute chef (puppet, ansible, bash, etc)
Can you show me equivalent code that fulfills all my requirements using VirtualBox directly and will allow me to change a couple lines to spin this same infrastructure up in AWS or using VMWare?I'm in the market for 3 new thermostats and was hoping this article would help me with my decision. Presently, I'm leaning toward an Ecobees (https://www.ecobee.com/) thermostat which has a lot more features than the Nest and even has an API. It seems to be the Android of the thermostat world where Nest is the iPhone. Features vs polish.
# create a key file
dd if=/dev/urandom of=/etc/mykeyfile.key bs=512 count=1
# create the luks container
cryptsetup luksFormat /dev/sdf1 /etc/mykeyfile.key
# open the luks container
cryptsetup luksOpen /dev/sdf1 myluksvolume --key-file /etc/mykeyfile.key
# format the volume
mkfs.xfs /dev/mapper/myluksvolume
# mount the disk
mount /dev/mapper/myluksvolume /mnt/mymountpoint for tbl in $(mysql -BNe "SELECT DISTINCT TABLE_NAME FROM INFORMATION_SCHEMA.COLUMNS WHERE EXTRA = 'auto_increment' AND TABLE_SCHEMA = 'mydb'"); do
mysql -e "ALTER TABLE ${tbl} AUTO_INCREMENT=1" mydb
done
Run that just after importing your new schema.Note: That code will require a bit of modification if you make the mistake of having spaces in your table name.
What, specifically, "sucks" about Puppet and Chef and what is so much "simpler" about Salt or Ansible? As an Ops guy who has been running Puppet since 2008 (and Chef most recently) against hundred of servers, I don't see the simplicity reflected in the documentation, nor do I find Puppet or Chef particularly complicated.
(Ok, Chef's attributes system is a bit confusing at first, but it is hugely powerful.)
I can assure you, you can exhale upon ascent when inverted. (source: me, who used to log plenty of hours below 150fsw breathing mixed gasses).
Here's a link that provides a number of references: http://www.thediveforum.com/dive-medicine/3706-long-term-phy...