-
Notifications
You must be signed in to change notification settings - Fork 3
/
master.tf
256 lines (216 loc) · 7.59 KB
/
master.tf
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
# Reattach the internal ELBs to the master if they change
resource "aws_elb_attachment" "internal-master-elb" {
count = "${var.num_of_masters}"
elb = "${aws_elb.internal-master-elb.id}"
instance = "${aws_instance.master.*.id[count.index]}"
}
# Internal Load Balancer Access
# Mesos Master, Zookeeper, Exhibitor, Adminrouter, Marathon
resource "aws_elb" "internal-master-elb" {
name = "${data.template_file.cluster-name.rendered}-int-master-elb"
internal = "true"
subnets = ["${aws_subnet.public.id}"]
security_groups = ["${aws_security_group.master.id}","${aws_security_group.public_slave.id}", "${aws_security_group.private_slave.id}"]
instances = ["${aws_instance.master.*.id}"]
listener {
lb_port = 5050
instance_port = 5050
lb_protocol = "http"
instance_protocol = "http"
}
listener {
lb_port = 2181
instance_port = 2181
lb_protocol = "tcp"
instance_protocol = "tcp"
}
listener {
lb_port = 8181
instance_port = 8181
lb_protocol = "http"
instance_protocol = "http"
}
listener {
lb_port = 80
instance_port = 80
lb_protocol = "tcp"
instance_protocol = "tcp"
}
listener {
lb_port = 443
instance_port = 443
lb_protocol = "tcp"
instance_protocol = "tcp"
}
listener {
lb_port = 8080
instance_port = 8080
lb_protocol = "http"
instance_protocol = "http"
}
health_check {
healthy_threshold = 2
unhealthy_threshold = 2
timeout = 5
target = "TCP:5050"
interval = 30
}
lifecycle {
ignore_changes = ["name"]
}
}
# Reattach the public ELBs to the master if they change
resource "aws_elb_attachment" "public-master-elb" {
count = "${var.num_of_masters}"
elb = "${aws_elb.public-master-elb.id}"
instance = "${aws_instance.master.*.id[count.index]}"
}
# Public Master Load Balancer Access
# Adminrouter Only
resource "aws_elb" "public-master-elb" {
name = "${data.template_file.cluster-name.rendered}-pub-mas-elb"
subnets = ["${aws_subnet.public.id}"]
security_groups = ["${aws_security_group.http-https.id}", "${aws_security_group.master.id}", "${aws_security_group.internet-outbound.id}"]
instances = ["${aws_instance.master.*.id}"]
listener {
lb_port = 80
instance_port = 80
lb_protocol = "tcp"
instance_protocol = "tcp"
}
listener {
lb_port = 443
instance_port = 443
lb_protocol = "tcp"
instance_protocol = "tcp"
}
health_check {
healthy_threshold = 2
unhealthy_threshold = 2
timeout = 5
target = "TCP:5050"
interval = 30
}
lifecycle {
ignore_changes = ["name"]
}
}
resource "aws_instance" "master" {
# The connection block tells our provisioner how to
# communicate with the resource (instance)
connection {
# The default username for our AMI
user = "${module.aws-tested-oses.user}"
private_key = "${local.private_key}"
agent = "${local.agent}"
# The connection will use the local SSH agent for authentication.
}
root_block_device {
volume_size = "${var.aws_master_instance_disk_size}"
}
count = "${var.num_of_masters}"
instance_type = "${var.aws_master_instance_type}"
iam_instance_profile = "${aws_iam_instance_profile.master.name}"
ebs_optimized = "true"
tags {
owner = "${coalesce(var.owner, data.external.whoami.result["owner"])}"
expiration = "${var.expiration}"
Name = "${data.template_file.cluster-name.rendered}-master-${count.index + 1}"
cluster = "${data.template_file.cluster-name.rendered}"
}
# Lookup the correct AMI based on the region
# we specified
ami = "${module.aws-tested-oses.aws_ami}"
# The name of our SSH keypair we created above.
key_name = "${var.ssh_key_name}"
# Our Security group to allow http, SSH, and outbound internet access only for pulling containers from the web
vpc_security_group_ids = ["${aws_security_group.http-https.id}", "${aws_security_group.any_access_internal.id}", "${aws_security_group.ssh.id}", "${aws_security_group.internet-outbound.id}"]
# OS init script
provisioner "file" {
content = "${module.aws-tested-oses.os-setup}"
destination = "/tmp/os-setup.sh"
}
# We're going to launch into the same subnet as our ELB. In a production
# environment it's more common to have a separate private subnet for
# backend instances.
subnet_id = "${aws_subnet.public.id}"
# We run a remote provisioner on the instance after creating it.
# In this case, we just install nginx and start it. By default,
# this should be on port 80
provisioner "remote-exec" {
inline = [
"sudo chmod +x /tmp/os-setup.sh",
"sudo bash /tmp/os-setup.sh",
]
}
lifecycle {
ignore_changes = ["tags.Name", "tags.cluster"]
}
}
# Create DCOS Mesos Master Scripts to execute
module "dcos-mesos-master" {
source = "github.com/dcos/tf_dcos_core"
bootstrap_private_ip = "${aws_instance.bootstrap.private_ip}"
dcos_bootstrap_port = "${var.custom_dcos_bootstrap_port}"
dcos_type = "${var.dcos_type}"
# Only allow upgrade and install as installation mode
dcos_install_mode = "${var.state == "upgrade" ? "upgrade" : "install"}"
dcos_version = "${var.dcos_version}"
role = "dcos-mesos-master"
}
resource "null_resource" "master" {
# If state is set to none do not install DC/OS
count = "${var.state == "none" ? 0 : var.num_of_masters}"
# Changes to any instance of the cluster requires re-provisioning
triggers {
cluster_instance_ids = "${null_resource.bootstrap.id}"
current_ec2_instance_id = "${aws_instance.master.*.id[count.index]}"
}
# Bootstrap script can run on any instance of the cluster
# So we just choose the first in this case
connection {
host = "${element(aws_instance.master.*.public_ip, count.index)}"
user = "${module.aws-tested-oses.user}"
private_key = "${local.private_key}"
agent = "${local.agent}"
}
count = "${var.num_of_masters}"
# Generate and upload Master script to node
provisioner "file" {
content = "${module.dcos-mesos-master.script}"
destination = "run.sh"
}
# Wait for bootstrapnode to be ready
provisioner "remote-exec" {
inline = [
"until $(curl --output /dev/null --silent --head --fail http://${aws_instance.bootstrap.private_ip}:${var.custom_dcos_bootstrap_port}/dcos_install.sh); do printf 'waiting for bootstrap node to serve...'; sleep 20; done"
]
}
# Install Master Script
provisioner "remote-exec" {
inline = [
"sudo chmod +x run.sh",
"sudo ./run.sh",
]
}
# Watch Master Nodes Start
provisioner "remote-exec" {
inline = [
"until $(curl --output /dev/null --silent --head --fail http://${element(aws_instance.master.*.private_ip, count.index)}/); do printf 'loading DC/OS...'; sleep 10; done"
]
}
# Mesos poststart check workaround. Engineering JIRA filed to Mesosphere team to fix.
provisioner "remote-exec" {
inline = [
"sudo sed -i.bak '131 s/1s/10s/' /opt/mesosphere/packages/dcos-config--setup*/etc/dcos-diagnostics-runner-config.json &> /dev/null || true",
"sudo sed -i.bak '140 s/1s/10s/' /opt/mesosphere/packages/dcos-config--setup*/etc/dcos-check-config.json &> /dev/null || true",
"sudo sed -i.bak '162 s/1s/10s/' /opt/mesosphere/packages/dcos-config--setup*/etc/dcos-diagnostics-runner-config.json &> /dev/null || true"
]
}
}
output "Master ELB Public IP" {
value = "${aws_elb.public-master-elb.dns_name}"
}
output "Master Public IPs" {
value = ["${aws_instance.master.*.public_ip}"]
}