3 navsection: installguide
4 title: Install the cloud dispatcher
7 Copyright (C) The Arvados Authors. All rights reserved.
9 SPDX-License-Identifier: CC-BY-SA-3.0
12 {% include 'notebox_begin_warning' %}
13 @arvados-dispatch-cloud@ is only relevant for cloud installations. Skip this section if you are installing an on premises cluster that will spool jobs to Slurm or LSF.
14 {% include 'notebox_end' %}
16 # "Introduction":#introduction
17 # "Create compute node VM image":#create-image
18 # "Update config.yml":#update-config
19 # "Install arvados-dispatch-cloud":#install-packages
20 # "Start the service":#start-service
21 # "Restart the API server and controller":#restart-api
22 # "Confirm working installation":#confirm-working
24 h2(#introduction). Introduction
26 The cloud dispatch service is for running containers on cloud VMs. It works with Microsoft Azure and Amazon EC2; future versions will also support Google Compute Engine.
28 The cloud dispatch service can run on any node that can connect to the Arvados API service, the cloud provider's API, and the SSH service on cloud VMs. It is not resource-intensive, so you can run it on the API server node.
30 More detail about the internal operation of the dispatcher can be found in the "architecture section":{{site.baseurl}}/architecture/dispatchcloud.html.
32 h2(#update-config). Update config.yml
34 h3. Configure CloudVMs
36 Add or update the following portions of your cluster configuration file, @config.yml@. Refer to "config.defaults.yml":{{site.baseurl}}/admin/config.html for information about additional configuration options. The @DispatchPrivateKey@ should be the *private* key generated in "the previous section":install-compute-node.html#sshkeypair.
42 "http://localhost:9006": {}
45 # BootProbeCommand is a shell command that succeeds when an instance is ready for service
46 BootProbeCommand: "sudo systemctl status docker"
48 <b># --- driver-specific configuration goes here --- see Amazon and Azure examples below ---</b>
51 -----BEGIN RSA PRIVATE KEY-----
52 MIIEpQIBAAKCAQEAqXoCzcOBkFQ7w4dvXf9B++1ctgZRqEbgRYL3SstuMV4oawks
53 ttUuxJycDdsPmeYcHsKo8vsEZpN6iYsX6ZZzhkO5nEayUTU8sBjmg1ZCTo4QqKXr
54 FJ+amZ7oYMDof6QEdwl6KNDfIddL+NfBCLQTVInOAaNss7GRrxLTuTV7HcRaIUUI
55 jYg0Ibg8ZZTzQxCvFXXnjseTgmOcTv7CuuGdt91OVdoq8czG/w8TwOhymEb7mQlt
56 lXuucwQvYgfoUgcnTgpJr7j+hafp75g2wlPozp8gJ6WQ2yBWcfqL2aw7m7Ll88Nd
58 oFyAjVoexx0RBcH6BveTfQtJKbktP1qBO4mXo2dP0cacuZEtlAqW9Eb06Pvaw/D9
59 foktmqOY8MyctzFgXBpGTxPliGjqo8OkrOyQP2g+FL7v+Km31Xs61P8=
60 -----END RSA PRIVATE KEY-----
63 ProviderType: x1.medium
69 ProviderType: x1.large
72 IncludedScratch: 128GB
77 h4. NVIDIA GPU support
79 To specify instance types with NVIDIA GPUs, you must include an additional @CUDA@ section:
82 <pre><code> InstanceTypes:
84 ProviderType: g4dn.xlarge
87 IncludedScratch: 125GB
91 HardwareCapability: "7.5"
96 The @DriverVersion@ is the version of the CUDA toolkit installed in your compute image (in X.Y format, do not include the patchlevel). The @HardwareCapability@ is the CUDA compute capability of the GPUs available for this instance type. The @DeviceCount@ is the number of GPU cores available for this instance type.
98 h4. Minimal configuration example for Amazon EC2
100 The <span class="userinput">ImageID</span> value is the compute node image that was built in "the previous section":install-compute-node.html#aws.
103 <pre><code> Containers:
105 ImageID: <span class="userinput">ami-01234567890abcdef</span>
108 # If you are not using an IAM role for authentication, specify access
109 # credentials here. Otherwise, omit or set AccessKeyID and
110 # SecretAccessKey to an empty value.
111 AccessKeyID: XXXXXXXXXXXXXXXXXXXX
112 SecretAccessKey: YYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYY
116 SubnetID: subnet-0123abcd
119 AdminUsername: arvados
123 Example policy for the IAM role used by the cloud dispatcher:
128 "Version": "2012-10-17",
129 "Id": "arvados-dispatch-cloud policy",
135 "ec2:DescribeKeyPairs",
138 "ec2:DescribeInstances",
140 "ec2:TerminateInstances"
149 h4. Minimal configuration example for Azure
153 The <span class="userinput">ImageID</span> value is the compute node image that was built in "the previous section":install-compute-node.html#azure.
156 <pre><code> Containers:
158 ImageID: <span class="userinput">"zzzzz-compute-v1597349873"</span>
160 # (azure) managed disks: set MaxConcurrentInstanceCreateOps to 20 to avoid timeouts, cf
161 # https://docs.microsoft.com/en-us/azure/virtual-machines/linux/capture-image
162 MaxConcurrentInstanceCreateOps: 20
165 SubscriptionID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
166 ClientID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
167 ClientSecret: XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
168 TenantID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
170 # Data center where VMs will be allocated
173 # The resource group where the VM and virtual NIC will be
176 NetworkResourceGroup: yyyyy # only if different from ResourceGroup
178 Subnet: xxxxx-subnet-private
180 # The resource group where the disk image is stored, only needs to
181 # be specified if it is different from ResourceGroup
182 ImageResourceGroup: aaaaa
187 Azure recommends using managed images. If you plan to start more than 20 VMs simultaneously, Azure recommends using a shared image gallery instead to avoid slowdowns and timeouts during the creation of the VMs.
189 Using an image from a shared image gallery:
192 <pre><code> Containers:
194 ImageID: <span class="userinput">"shared_image_gallery_image_definition_name"</span>
198 SubscriptionID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
199 ClientID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
200 ClientSecret: XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
201 TenantID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
203 # Data center where VMs will be allocated
206 # The resource group where the VM and virtual NIC will be
209 NetworkResourceGroup: yyyyy # only if different from ResourceGroup
211 Subnet: xxxxx-subnet-private
213 # The resource group where the disk image is stored, only needs to
214 # be specified if it is different from ResourceGroup
215 ImageResourceGroup: aaaaa
217 # (azure) shared image gallery: the name of the gallery
218 SharedImageGalleryName: "shared_image_gallery_1"
219 # (azure) shared image gallery: the version of the image definition
220 SharedImageGalleryImageVersion: "0.0.1"
225 Using unmanaged disks (deprecated):
227 The <span class="userinput">ImageID</span> value is the compute node image that was built in "the previous section":install-compute-node.html#azure.
230 <pre><code> Containers:
232 ImageID: <span class="userinput">"https://zzzzzzzz.blob.core.windows.net/system/Microsoft.Compute/Images/images/zzzzz-compute-osDisk.55555555-5555-5555-5555-555555555555.vhd"</span>
236 SubscriptionID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
237 ClientID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
238 ClientSecret: XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
239 TenantID: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX
241 # Data center where VMs will be allocated
244 # The resource group where the VM and virtual NIC will be
247 NetworkResourceGroup: yyyyy # only if different from ResourceGroup
249 Subnet: xxxxx-subnet-private
251 # Where to store the VM VHD blobs
252 StorageAccount: example
258 Get the @SubscriptionID@ and @TenantID@:
264 "cloudName": "AzureCloud",
265 "id": "XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXX",
267 "name": "Your Subscription",
269 "tenantId": "YYYYYYYY-YYYY-YYYY-YYYYYYYY",
271 "name": "you@example.com",
278 You will need to create a "service principal" to use as a delegated authority for API access.
280 <notextile><pre><code>$ az ad app create --display-name "Arvados Dispatch Cloud (<span class="userinput">ClusterID</span>)" --homepage "https://arvados.org" --identifier-uris "https://<span class="userinput">ClusterID.example.com</span>" --end-date 2299-12-31 --password <span class="userinput">Your_Password</span>
281 $ az ad sp create "<span class="userinput">appId</span>"
282 (appId is part of the response of the previous command)
283 $ az role assignment create --assignee "<span class="userinput">objectId</span>" --role Owner --scope /subscriptions/{subscriptionId}/
284 (objectId is part of the response of the previous command)
285 </code></pre></notextile>
287 Now update your @config.yml@ file:
289 @ClientID@ is the 'appId' value.
291 @ClientSecret@ is what was provided as <span class="userinput">Your_Password</span>.
293 h3. Test your configuration
295 Run the @cloudtest@ tool to verify that your configuration works. This creates a new cloud VM, confirms that it boots correctly and accepts your configured SSH private key, and shuts it down.
298 <pre><code>~$ <span class="userinput">arvados-server cloudtest && echo "OK!"</span>
302 Refer to the "cloudtest tool documentation":../../admin/cloudtest.html for more information.
304 {% assign arvados_component = 'arvados-dispatch-cloud' %}
306 {% include 'install_packages' %}
308 {% include 'start_service' %}
310 {% include 'restart_api' %}
312 h2(#confirm-working). Confirm working installation
314 On the dispatch node, start monitoring the arvados-dispatch-cloud logs:
317 <pre><code>~$ <span class="userinput">sudo journalctl -o cat -fu arvados-dispatch-cloud.service</span>
321 "Make sure to install the arvados/jobs image.":../install-jobs-image.html
323 Submit a simple container request:
326 <pre><code>shell:~$ <span class="userinput">arv container_request create --container-request '{
328 "state": "Committed",
330 "container_image": "arvados/jobs:latest",
331 "command": ["echo", "Hello, Crunch!"],
332 "output_path": "/out",
339 "runtime_constraints": {
347 This command should return a record with a @container_uuid@ field. Once @arvados-dispatch-cloud@ polls the API server for new containers to run, you should see it dispatch that same container.
349 The @arvados-dispatch-cloud@ API provides a list of queued and running jobs and cloud instances. Use your @ManagementToken@ to test the dispatcher's endpoint. For example, when one container is running:
352 <pre><code>~$ <span class="userinput">curl -sH "Authorization: Bearer $token" http://localhost:9006/arvados/v1/dispatch/containers</span>
357 "uuid": "zzzzz-dz642-hdp2vpu9nq14tx0",
360 "scheduling_parameters": {
362 "preemptible": false,
366 "runtime_status": null,
371 "Name": "Standard_D2s_v3",
372 "ProviderType": "Standard_D2s_v3",
375 "Scratch": 16000000000,
376 "IncludedScratch": 16000000000,
387 A similar request can be made to the @http://localhost:9006/arvados/v1/dispatch/instances@ endpoint.
389 When the container finishes, the dispatcher will log it.
391 After the container finishes, you can get the container record by UUID *from a shell server* to see its results:
394 <pre><code>shell:~$ <span class="userinput">arv get <b>zzzzz-dz642-hdp2vpu9nq14tx0</b></span>
398 "log":"a01df2f7e5bc1c2ad59c60a837e90dc6+166",
399 "output":"d41d8cd98f00b204e9800998ecf8427e+0",
406 You can use standard Keep tools to view the container's output and logs from their corresponding fields. For example, to see the logs from the collection referenced in the @log@ field:
409 <pre><code>~$ <span class="userinput">arv keep ls <b>a01df2f7e5bc1c2ad59c60a837e90dc6+166</b></span>
413 ~$ <span class="userinput">arv-get <b>a01df2f7e5bc1c2ad59c60a837e90dc6+166</b>/stdout.txt</span>
414 2016-08-05T13:53:06.201011Z Hello, Crunch!
418 If the container does not dispatch successfully, refer to the @arvados-dispatch-cloud@ logs for information about why it failed.