Initial commit

.gitignore

@@ -0,0 +1,15 @@
+*.qcow2
+*.tfstate*
+
+/*.tfvars
+
+*.local
+*local.tf
+*local.sh
+
+# files from profiles
+/profile-*.tf
+
+*.crt
+*.key
+*.rpm

Makefile

@@ -0,0 +1,296 @@
+#
+# Author(s): Alvaro Saurin <[email protected]>
+#
+# Copyright (c) 2017 SUSE LINUX GmbH, Nuernberg, Germany.
+#
+# All modifications and additions to the file contributed by third parties
+# remain the property of their copyright owners, unless otherwise agreed
+# upon. The license for this file, and modifications and additions to the
+# file, is the same license as for the pristine package itself (unless the
+# license for the pristine package is not an Open Source License, in which
+# case the license is the MIT License). An "Open Source License" is a
+# license that conforms to the Open Source Definition (Version 1.9)
+# published by the Open Source Initiative.
+#
+
+# Makefile FOR DEVELOPERS
+# if you are not a developer, just do "terraform apply"
+#
+# Usage:
+# * install "sshpass", "wget"...
+# * customize some of these vars (for example, putting them in a Makefile.local)
+# * do "make dev-apply" (or some othe target)
+#
+
+# Note: you can overwrite these vars from command line with something like:
+# make dev-apply CHECKOUTS_DIR=~/dev
+
+# prefix for all the resources'
+PREFIX            = caasp
+
+# pool used for images in libvirt
+LIBVIRT_POOL_NAME = personal
+LIBVIRT_POOL_DIR  = ~/.libvirt/images
+
+# some dis
+CHECKOUTS_DIR     = ~/Development/SUSE
+SALT_DIR          = $(CHECKOUTS_DIR)/k8s-salt
+SALT_VM_DIR       = /usr/share/salt/kubernetes
+E2E_TESTS_RUNNER  = $(CHECKOUTS_DIR)/automation/k8s-e2e-tests/e2e-tests
+
+#####################################################################
+
+# the directory with resources that will be copied to the VMs
+RESOURCES_DIR    = resources
+
+VARS_STAGING_A   = profiles/devel/images-staging-a.tfvars
+VARS_STAGING_B   = profiles/devel/images-staging-b.tfvars
+
+# VMs we use
+VMS_NODES        = node_0 node_1
+VMS_ALL          = admin $(VMS_NODES)
+VMS_NUM_MINIONS  = 2
+
+CAASPCTL         = /tmp/caasp/caaspctl
+RUN_CAASPCTL     = bash $(CAASPCTL)
+
+# environment variables we always pass to Terraform
+TERRAFORM_VARS   = TF_VAR_prefix=$(PREFIX) \
+	               TF_VAR_img_pool=$(LIBVIRT_POOL_NAME) \
+	               TF_VAR_salt_dir=$(SALT_DIR)
+
+# common options for ssh, rsync, etc
+SSH              = sshpass -p "linux" ssh
+SCP              = sshpass -p "linux" scp
+SSH_OPTS         = -oStrictHostKeyChecking=no \
+                   -oUserKnownHostsFile=/dev/null
+EXCLUDE_ARGS     = --exclude='*.tfstate*' \
+                   --exclude='.git*' \
+                   --exclude='Makefile' \
+                   --exclude='README.md' \
+                   --exclude='*.sublime-*' \
+                   --exclude='.idea' \
+                   --exclude='*.tgz'
+RSYNC_OPTS       = $(EXCLUDE_ARGS) -e '$(SSH) $(SSH_OPTS)' --delete
+
+# stuff for building distributions
+DIST_CONT        = *.profile  k8s-setup terraform
+DIST_TAR         = kubernetes-terraform.tgz
+TAR_ARGS         = $(EXCLUDE_ARGS) -zcvf
+
+# a kubeconfig we will download from the master
+KUBECONFIG       = kubeconfig
+
+# files to remove after a "destroy"
+CLEANUP_FILES    = terraform.tfstate* \
+                   admin.{tar,crt,key} ca.crt \
+                   $(KUBECONFIG) \
+		           $(LIBVIRT_POOL_DIR)/$(PREFIX)_*
+
+# the dashboard and node IPs (might need a "terraform refresh" after a while)
+ADMIN_IP         = `support/mk/get-admin-ip.sh`
+NODES_IPS        = `support/mk/get-node-ip.sh`
+
+# you can customimze vars in a local Makefile
+-include Makefile.local
+
+####################################################################
+# CAASP
+####################################################################
+
+all:
+	@echo "Makefile FOR DEVELOPERS"
+	@echo "For regular users please do: terraform apply"
+
+dev-apply-with-args:
+	@echo ">>> Applying Terraform..."
+	@env $(TERRAFORM_VARS) terraform apply $(ARGS)
+	@sleep 10 && make dev-snapshot
+
+dev-apply:
+	make dev-apply-with-args
+dev-apply-staging-a:
+	make dev-apply-with-args ARGS="-var-file=$(VARS_STAGING_A) $(ARGS)"
+dev-apply-staging-b:
+	make dev-apply-with-args ARGS="-var-file=$(VARS_STAGING_B) $(ARGS)"
+
+dev-destroy:
+	-@for h in $(VMS_ALL) ; do \
+		echo ">>> Destroying snapshots for $(PREFIX)_$$h" ; \
+		while sudo virsh snapshot-delete $(PREFIX)_$$h --current &>/dev/null ; do echo ; done ; \
+	done
+	-terraform destroy -force
+	-rm -f $(CLEANUP_FILES)
+	-@notify-send "k8s: cluster destruction finished" &>/dev/null
+
+dev-copy:
+	@echo ">>> Copying the VM's resources to /tmp/caasp"
+	@rsync -avz $(RSYNC_OPTS) $(RESOURCES_DIR)/common/ root@$(ADMIN_IP):/tmp/caasp/
+	@rsync -avz $(RSYNC_OPTS) $(RESOURCES_DIR)/admin/  root@$(ADMIN_IP):/tmp/caasp/admin/
+	@echo ">>> Making fs RW-able"
+	@make dev-ssh CMD='$(RUN_CAASPCTL) rw on'
+	@echo ">>> Copying the Salt scripts/pillar"
+	@rsync -avz $(RSYNC_OPTS) $(SALT_DIR)/  root@$(ADMIN_IP):$(SALT_VM_DIR)/
+	@echo ">>> Synchronizing Salt stuff"
+	@make dev-ssh CMD='$(RUN_CAASPCTL) salt sync'
+
+dev-orch: dev-copy
+	@make dev-ssh CMD='$(RUN_CAASPCTL) orchestrate'
+	@rm -f $(KUBECONFIG)
+	-@notify-send "k8s: cluster orchestration finished" &>/dev/null
+
+dev-reorch: dev-rollback _wait-20s dev-orch
+dev:        dev-apply dev-orch
+
+dev-restart-master: dev-copy
+	@make dev-ssh CMD='$(RUN_CAASPCTL) salt restart-master'
+dev-restart-api: dev-copy
+	@make dev-ssh CMD='$(RUN_CAASPCTL) salt restart-api'
+
+# some times we might need to refresh data
+# and readjust some stuff (ie, when IPs change)
+dev-refresh:
+	support/mk/refresh-vms.sh
+
+dev-kubeconfig: $(KUBECONFIG)
+$(KUBECONFIG):
+	@echo ">>> Generating a kubeconfig"
+	@make dev-ssh CMD='$(RUN_CAASPCTL) kubeconfig gen'
+	@echo ">>> Getting the kubeconfig"
+	@$(SCP) -q $(SSH_OPTS) root@$(ADMIN_IP):.kube/config $(KUBECONFIG)
+	@echo ">>> done."
+
+# run the e2e tests
+dev-e2e: $(KUBECONFIG)
+	@echo ">>> Running the kubernetes e2e tests"
+	$(E2E_TESTS_RUNNER) --kubeconfig $(KUBECONFIG) $(E2E_ARGS)
+	-@notify-send "k8s: e2e tests finished" &>/dev/null
+
+####################################################################
+# profiles
+
+dev-profile-apply: dev-profile-clean
+	@echo ">>> Applying development profile"
+	@for i in profiles/devel/profile-devel*.tf ; do ln -sf $$i ; done
+
+dev-profile-clean:
+	@echo ">>> Cleaning profile files"
+	@for i in *.tf ; do \
+		if [ -L $$i ] ; then \
+			l=`readlink $$i` ; \
+			[[ $$l == profiles/* ]] && rm -f $$i ; \
+		fi ; \
+	done
+
+####################################################################
+# updates & reboots
+
+dev-nodes-dist-update:
+	@make dev-ssh-nodes CMD='$(RUN_CAASPCTL) zypper update'
+	-@notify-send "k8s: cluster updates downloaded... would need a reboot" &>/dev/null
+
+dev-nodes-reboot:
+	@make dev-ssh-nodes CMD='$(RUN_CAASPCTL) reboot'
+	-@notify-send "k8s: nodes rebooted" &>/dev/null
+
+####################################################################
+# some ssh convencience targets
+
+dev-ssh: dev-ssh-admin
+dev-ssh-admin:
+	@$(SSH) -q $(SSH_OPTS) root@$(ADMIN_IP) '$(CMD)'
+
+dev-ssh-nodes:
+	@for node in $(NODES_IPS) ; do \
+		$(SSH) -q $(SSH_OPTS) root@$$node '$(CMD)' ; \
+	done
+
+dev-ssh-node-0:
+	@$(SSH) $(SSH_OPTS) root@`support/mk/get-node-ip.sh 0` '$(CMD)'
+dev-ssh-node-1:
+	@$(SSH) $(SSH_OPTS) root@`support/mk/get-node-ip.sh 1` '$(CMD)'
+
+dev-ssh-salt-master:
+	@make dev-ssh CMD='$(RUN_CAASPCTL) salt $(CMD)'
+
+####################################################################
+# some logging utilities
+dev-logs-salt-master:
+	@echo ">>> Dumping logs from the Salt master"
+	@make dev-ssh CMD='$(RUN_CAASPCTL) salt logs'
+
+dev-logs-events:
+	@echo ">>> Dumping Salt events at the master"
+	@make dev-ssh CMD='$(RUN_CAASPCTL) salt events'
+
+####################################################################
+# VMs management
+
+dev-snapshot:
+	@echo ">>> Creating snapshots..."
+	@for h in $(VMS_ALL) ; do \
+		sudo virsh snapshot-create --atomic $(PREFIX)_$$h ; \
+	done
+	-@notify-send "k8s: cluster creation finished" &>/dev/null
+
+dev-suspend:
+	@echo ">>> Suspending VMs before rolling back"
+	-@for h in $(VMS_ALL) ; do \
+		sudo virsh suspend $(PREFIX)_$$h ; \
+	done
+	-@notify-send "k8s: all VMs suspended" &>/dev/null
+
+dev-resume:
+	@echo ">>> Resuming VMs before rolling back"
+	-@for h in $(VMS_ALL) ; do \
+		sudo virsh resume $(PREFIX)_$$h ; \
+	done
+	-@notify-send "k8s: all VMs suspended" &>/dev/null
+
+# NOTE: we need to "echo 1 > /proc/sys/vm/overcommit_memory"
+#       or qemu-kvm will kill our machine...
+dev-rollback: dev-suspend
+	@for h in $(VMS_ALL) ; do \
+		echo ">>> Rolling back $(PREFIX)_$$h" ; \
+		sudo virsh snapshot-revert --current --running $(PREFIX)_$$h ; \
+	done
+	@echo ">>> Refreshing Terraform data"
+	-@make dev-refresh
+	-@notify-send "k8s: all VMs rolled back" &>/dev/null
+
+####################################################################
+# packages installation
+
+_install-rpms-on:
+	@echo "Copying RPMs to $(NODE)"
+	@$(SSH) -q $(SSH_OPTS) root@$(NODE) 'rm -rf /tmp/rpms && mkdir -p /tmp/rpms'
+	@$(SCP) -q $(SSH_OPTS) rpms/* root@$(NODE):/tmp/rpms/
+	@$(SSH) -q $(SSH_OPTS) root@$(NODE) 'ls -lisah /tmp/rpms/*'
+	@echo "Importing keys"
+	@$(SSH) -q $(SSH_OPTS) root@$(NODE) 'caaspctl rw 1'
+	@$(SSH) -q $(SSH_OPTS) root@$(NODE) 'rpm --import /tmp/rpms/*.key /tmp/rpms/*.pub || /bin/true'
+	@echo "Installing packages"
+	@$(SSH) -q $(SSH_OPTS) root@$(NODE) 'caaspctl zypper in -y /tmp/rpms/*.rpm'
+	@echo "Rebooting $(NODE)"
+	@-$(SSH) -q $(SSH_OPTS) root@$(NODE) 'reboot'
+	-@notify-send "k8s: packages installed in $(NODE)" &>/dev/null
+
+dev-install-rpms-nodes:
+	@for node in $(NODES_IPS) ; do \
+		make _install-rpms-on NODE=$$node ; \
+	done
+
+####################################################################
+# aux
+
+.PHONY: _wait-20s
+_wait-20s:
+	@echo ">>> Waiting some time..."
+	@sleep 20
+
+####################################################################
+# distribution
+
+dist:
+	@echo "Creating distribution package"
+	tar $(TAR_ARGS) $(DIST_TAR) $(DIST_CONT)

README.md

@@ -0,0 +1,71 @@
+
+# Requirements
+
+* `libvirt`
+* [`terraform`](https://github.com/hashicorp/terraform)
+* the [`terraform-provider-libvirt`](https://github.com/dmacvicar/terraform-provider-libvirt) plugin
+* Some olter tools like `wget`, `sshpass`...
+
+# Cluster configuration
+
+The deployment can by tuned uses some terraform variables. All of them
+are defined at the top of the `terraform.tf` file. Each variable has also a
+description field that explains its purpose.
+
+These are the most important ones:
+
+  * `libvirt_uri`: by default this points to localhost, however it's possible
+    to perform the deployment on a different libvirt machine. More on that later.
+  * `img_src`: this is the URL of **a directory** where the CaaSP image 
+    can be found for creating the whole cluster. Note: the
+    **latest version** of the image will be **automatically obtained**
+    unless the `refresh` variables is set fo `false`.
+  * `nodes_count`: number of non-admin nodes to be created.
+
+The easiest way to set these values is by creating a `terraform.tfvars`. The
+project comes with an example file named `terraform.tfvars.example`.
+
+## cloud-init
+
+The project comes with two cloud-init files: one for the admin node, the other
+for the generic nodes.
+
+Note well: the system is going to have a `root` and a `qa` users with password
+`linux` (specified on the Terraform variable `passwprd`).
+
+# Cluster architecture
+
+The cluster is made by 1 admin node and the number of generic nodes chosen by
+the user.
+
+All the nodes are based on the same CaaSP image and will have the same amount of
+memory.
+
+All of them have a cloud-init ISO attached to them to inject the cloud-init
+configuration.
+
+All the nodes are attached to the `default` network of libvirt. This is a network
+that satisfies CaaSP's basic network requirement: there's a DHCP and a DNS
+enabled but the DNS server is not able to resolve the names of the nodes inside
+of that network.
+
+# Creating the cluster
+
+Steps to perform:
+
+  * Configure the cluster the way you want (see above section).
+  * Execute: `terraform apply`
+
+At the end of the deployment you will see the IP address of the admin server.
+Use the velum instance running inside of this node to deploy the CaaSP cluster.
+
+## Using specific cluster profiles
+
+There are some specific profiles that can be useful for adding some extra
+behaviour for developers, QA people, etc.
+Take a look at these profiles (and read the instructions on how to use them):
+
+* the [development](profiles/devel) profile
+
+These files could be used as templates for your own `terraform-local.tf` recipes...
+