Deploying to gh-pages from @ dstackai/dstack@cd07427 🚀

Author: jvstme

docs/concepts/backends/index.html

@@ -850,6 +850,19 @@
     </span>
   </a>
 
+</li>
+      
+        <li class="md-nav__item">
+  <a href="#hot-aisle" class="md-nav__link">
+    <span class="md-ellipsis">
+      
+        <span class="md-typeset">
+          Hot Aisle
+        </span>
+      
+    </span>
+  </a>
+  
 </li>
 
         <li class="md-nav__item">
@@ -3855,6 +3868,19 @@
     </span>
   </a>
 
+</li>
+      
+        <li class="md-nav__item">
+  <a href="#hot-aisle" class="md-nav__link">
+    <span class="md-ellipsis">
+      
+        <span class="md-typeset">
+          Hot Aisle
+        </span>
+      
+    </span>
+  </a>
+  
 </li>
 
         <li class="md-nav__item">
@@ -4593,6 +4619,32 @@ <h2 id="gcp">GCP<a class="headerlink" href="#gcp" title="Permanent link">&para;<
 <p>Using private subnets assumes that both the <code>dstack</code> server and users can access the configured VPC's private subnets.
 Additionally, <a href="https://cloud.google.com/nat/docs/overview">Cloud NAT</a> must be configured to provide access to external resources for provisioned instances.</p>
 </details>
+<h2 id="hot-aisle">Hot Aisle<a class="headerlink" href="#hot-aisle" title="Permanent link">&para;</a></h2>
+<p>Log in to the SSH TUI as described in the <a href="https://hotaisle.xyz/quick-start/">Hot Aisle Quick Start <span class="twemoji external"><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="m11.93 5 2.83 2.83L5 17.59 6.42 19l9.76-9.75L19 12.07V5z"/></svg></span></a>.
+Create a new team and generate an API key for the member in the team.</p>
+<p>Then, go ahead and configure the backend:</p>
+<div editor-title="~/.dstack/server/config.yml">
+
+<div class="highlight"><pre><span></span><code><span class="nt">projects</span><span class="p">:</span>
+<span class="p p-Indicator">-</span><span class="w"> </span><span class="nt">name</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">main</span>
+<span class="w">  </span><span class="nt">backends</span><span class="p">:</span>
+<span class="w">    </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="nt">type</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">hotaisle</span>
+<span class="w">      </span><span class="nt">team_handle</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">hotaisle-team-handle</span>
+<span class="w">      </span><span class="nt">creds</span><span class="p">:</span>
+<span class="w">        </span><span class="nt">type</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">api_key</span>
+<span class="w">        </span><span class="nt">api_key</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">9c27a4bb7a8e472fae12ab34.3f2e3c1db75b9a0187fd2196c6b3e56d2b912e1c439ba08d89e7b6fcd4ef1d3f</span>
+</code></pre></div>
+
+</div>
+
+<details class="info">
+<summary>Required permissions</summary>
+<p>The API key must have the following roles assigned:</p>
+<ul>
+<li><strong>Owner role for the user</strong> - Required for creating and managing SSH keys</li>
+<li><strong>Operator role for the team</strong> - Required for managing virtual machines within the team</li>
+</ul>
+</details>
 <h2 id="lambda">Lambda<a class="headerlink" href="#lambda" title="Permanent link">&para;</a></h2>
 <p>Log into your <a href="https://lambdalabs.com/service/gpu-cloud">Lambda Cloud <span class="twemoji external"><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="m11.93 5 2.83 2.83L5 17.59 6.42 19l9.76-9.75L19 12.07V5z"/></svg></span></a> account, click API keys in the sidebar, and then click the <code>Generate API key</code>
 button to create a new API key.</p>

docs/concepts/services/index.html

@@ -4453,7 +4453,31 @@ <h3 id="probes">Probes<a class="headerlink" href="#probes" title="Permanent link
 </ul>
 <p>If multiple probes are configured for the service, their statuses are displayed in the order in which the probes appear in the configuration.</p>
 </details>
-<p>Probes are executed for each service replica while the replica is <code>running</code>. Probe statuses do not affect how <code>dstack</code> handles replicas, except during <a href="#rolling-deployment">rolling deployments</a>.</p>
+<p>Probes are executed for each service replica while the replica is <code>running</code>. A probe execution is considered successful if the replica responds with a <code>2xx</code> status code. Probe statuses do not affect how <code>dstack</code> handles replicas, except during <a href="#rolling-deployment">rolling deployments</a>.</p>
+<details class="info">
+<summary>HTTP request configuration</summary>
+<p>You can configure the HTTP request method, headers, and other properties. To include secret values in probe requests, use environment variable interpolation, which is enabled for the <code>url</code>, <code>headers[i].value</code>, and <code>body</code> properties.</p>
+<div editor-title="service.dstack.yml">
+<div class="highlight"><pre><span></span><code><span class="nt">type</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">service</span>
+<span class="nt">name</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">my-service</span>
+<span class="nt">port</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">80</span>
+<span class="nt">image</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">my-app:latest</span>
+<span class="nt">env</span><span class="p">:</span>
+<span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">PROBES_API_KEY</span>
+<span class="nt">probes</span><span class="p">:</span>
+<span class="p p-Indicator">-</span><span class="w"> </span><span class="nt">type</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">http</span>
+<span class="w">  </span><span class="nt">method</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">post</span>
+<span class="w">  </span><span class="nt">url</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">/check-health</span>
+<span class="w">  </span><span class="nt">headers</span><span class="p">:</span>
+<span class="w">  </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="nt">name</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">X-API-Key</span>
+<span class="w">    </span><span class="nt">value</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">${{ env.PROBES_API_KEY }}</span>
+<span class="w">  </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="nt">name</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">Content-Type</span>
+<span class="w">    </span><span class="nt">value</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">application/json</span>
+<span class="w">  </span><span class="nt">body</span><span class="p">:</span><span class="w"> </span><span class="s">&#39;{&quot;level&quot;:</span><span class="nv"> </span><span class="s">2}&#39;</span>
+<span class="w">  </span><span class="nt">timeout</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">20s</span>
+</code></pre></div>
+</div>
+</details>
 <p>See the <a href="../../reference/dstack.yml/service/#probes">reference</a> for more probe configuration options.</p>
 <h3 id="path-prefix">Path prefix<a class="headerlink" href="#path-prefix" title="Permanent link">&para;</a></h3>
 <p>If your <code>dstack</code> project doesn't have a <a href="../gateways/">gateway</a>, services are hosted with the

docs/reference/api/python/index.html

@@ -6539,6 +6539,17 @@ <h3 id="dstack.api.BackendType"><code>dstack.api.BackendType</code><a class="hea
               </div>
             </td>
           </tr>
+          <tr class="doc-section-item">
+            <td><code><span title="dstack.api.BackendType.HOTAISLE">HOTAISLE</span></code></td>
+            <td>
+                  <code><span title="dstack._internal.core.models.backends.base.BackendType">BackendType</span></code>
+            </td>
+            <td>
+              <div class="doc-md-description">
+                <p>Hot Aisle</p>
+              </div>
+            </td>
+          </tr>
           <tr class="doc-section-item">
             <td><code><span title="dstack.api.BackendType.KUBERNETES">KUBERNETES</span></code></td>
             <td>

docs/reference/api/rest/openapi.json

@@ -4147,9 +4147,16 @@ <h6 class="reference-item" id="header"><code>header</code> -  Name of the header
 <h3 id="probes"><code>probes</code><a class="headerlink" href="#probes" title="Permanent link">&para;</a></h3>
 <h4 id="probesn"><code>probes[n]</code><a class="headerlink" href="#probesn" title="Permanent link">&para;</a></h4>
 <h6 class="reference-item" id="url"><code>url</code> - (Optional) The URL to request. Defaults to <code>/</code>.<a class="headerlink" href="#url" title="Permanent link">&para;</a></h6>
+<h6 class="reference-item" id="method"><code>method</code> - (Optional) The HTTP method to use for the probe (e.g., <code>get</code>, <code>post</code>, etc.). Defaults to <code>get</code>.<a class="headerlink" href="#method" title="Permanent link">&para;</a></h6>
+<h6 class="reference-item" id="headers"><code>headers</code> - (Optional) A list of HTTP headers to include in the request.<a class="headerlink" href="#headers" title="Permanent link">&para;</a></h6>
+<h6 class="reference-item" id="body"><code>body</code> - (Optional) The HTTP request body to send with the probe.<a class="headerlink" href="#body" title="Permanent link">&para;</a></h6>
 <h6 class="reference-item" id="timeout"><code>timeout</code> - (Optional) Maximum amount of time the HTTP request is allowed to take. Defaults to <code>10s</code>.<a class="headerlink" href="#timeout" title="Permanent link">&para;</a></h6>
 <h6 class="reference-item" id="interval"><code>interval</code> - (Optional) Minimum amount of time between the end of one probe execution and the start of the next. Defaults to <code>15s</code>.<a class="headerlink" href="#interval" title="Permanent link">&para;</a></h6>
 <h6 class="reference-item" id="ready_after"><code>ready_after</code> - (Optional) The number of consecutive successful probe executions required for the replica to be considered ready. Used during rolling deployments. Defaults to <code>1</code>.<a class="headerlink" href="#ready_after" title="Permanent link">&para;</a></h6>
+<h5 id="probesnheaders"><code>probes[n].headers</code><a class="headerlink" href="#probesnheaders" title="Permanent link">&para;</a></h5>
+<h6 id="probesnheadersm"><code>probes[n].headers[m]</code><a class="headerlink" href="#probesnheadersm" title="Permanent link">&para;</a></h6>
+<h6 class="reference-item" id="name"><code>name</code> -  The name of the HTTP header.<a class="headerlink" href="#name" title="Permanent link">&para;</a></h6>
+<h6 class="reference-item" id="value"><code>value</code> -  The value of the HTTP header.<a class="headerlink" href="#value" title="Permanent link">&para;</a></h6>
 <h3 id="retry"><code>retry</code><a class="headerlink" href="#retry" title="Permanent link">&para;</a></h3>
 <h6 class="reference-item" id="on_events"><code>on_events</code> - (Optional) The list of events that should be handled with retry. Supported events are <code>no-capacity</code>, <code>interruption</code>, <code>error</code>. Omit to retry on all events.<a class="headerlink" href="#on_events" title="Permanent link">&para;</a></h6>
 <h6 class="reference-item" id="duration"><code>duration</code> - (Optional) The maximum period of retrying the run, e.g., <code>4h</code> or <code>1d</code>.<a class="headerlink" href="#duration" title="Permanent link">&para;</a></h6>

docs/reference/dstack.yml/service/index.html

@@ -3916,6 +3916,10 @@ <h2 id="server">Server<a class="headerlink" href="#server" title="Permanent link
 <li><code id="DSTACK_SERVER_BACKGROUND_PROCESSING_DISABLED">DSTACK_SERVER_BACKGROUND_PROCESSING_DISABLED</code> - Disables background processing if set to any value. Useful to run only web frontend and API server.</li>
 <li><code id="DSTACK_SERVER_MAX_PROBES_PER_JOB">DSTACK_SERVER_MAX_PROBES_PER_JOB</code> - Maximum number of probes allowed in a run configuration. Validated at apply time.</li>
 <li><code id="DSTACK_SERVER_MAX_PROBE_TIMEOUT">DSTACK_SERVER_MAX_PROBE_TIMEOUT</code> - Maximum allowed timeout for a probe. Validated at apply time.</li>
+<li><code id="DSTACK_SERVER_METRICS_RUNNING_TTL_SECONDS">DSTACK_SERVER_METRICS_RUNNING_TTL_SECONDS</code> – Maximum age of metrics samples for running jobs.</li>
+<li><code id="DSTACK_SERVER_METRICS_FINISHED_TTL_SECONDS">DSTACK_SERVER_METRICS_FINISHED_TTL_SECONDS</code> – Maximum age of metrics samples for finished jobs.</li>
+<li><code id="DSTACK_SERVER_INSTANCE_HEALTH_TTL_SECONDS">DSTACK_SERVER_INSTANCE_HEALTH_TTL_SECONDS</code> – Maximum age of instance health checks.</li>
+<li><code id="DSTACK_SERVER_INSTANCE_HEALTH_MIN_COLLECT_INTERVAL_SECONDS">DSTACK_SERVER_INSTANCE_HEALTH_MIN_COLLECT_INTERVAL_SECONDS</code> – Minimum time interval between consecutive health checks of the same instance.</li>
 </ul>
 <details class="info">
 <summary>Internal environment variables</summary>

docs/reference/environment-variables/index.html

@@ -3921,6 +3921,14 @@ <h6 class="reference-item" id="data"><code>data</code> - (Optional) The contents
 <div class="tabbed-block"></div>
 </div>
 </div>
+<h5 id="hotaisle"><code>projects[n].backends[type=hotaisle]</code><a class="headerlink" href="#hotaisle" title="Permanent link">&para;</a></h5>
+<h6 class="reference-item" id="type"><code>type</code> -  The type of backend. Must be <code>hotaisle</code>.<a class="headerlink" href="#type" title="Permanent link">&para;</a></h6>
+<h6 class="reference-item" id="team_handle"><code>team_handle</code> -  The Hot Aisle team handle.<a class="headerlink" href="#team_handle" title="Permanent link">&para;</a></h6>
+<h6 class="reference-item" id="regions"><code>regions</code> - (Optional) The list of Hot Aisle regions. Omit to use all regions.<a class="headerlink" href="#regions" title="Permanent link">&para;</a></h6>
+<h6 class="reference-item" id="_creds"><a href="#hotaisle-creds"><code>creds</code></a> -  The credentials.<a class="headerlink" href="#_creds" title="Permanent link">&para;</a></h6>
+<h6 id="hotaisle-creds"><code>projects[n].backends[type=hotaisle].creds</code><a class="headerlink" href="#hotaisle-creds" title="Permanent link">&para;</a></h6>
+<h6 class="reference-item" id="type"><code>type</code> -  The type of credentials. Must be <code>api_key</code>.<a class="headerlink" href="#type" title="Permanent link">&para;</a></h6>
+<h6 class="reference-item" id="api_key"><code>api_key</code> -  The Hot Aisle API key.<a class="headerlink" href="#api_key" title="Permanent link">&para;</a></h6>
 <h5 id="lambda"><code>projects[n].backends[type=lambda]</code><a class="headerlink" href="#lambda" title="Permanent link">&para;</a></h5>
 <h6 class="reference-item" id="type"><code>type</code> -  The type of backend. Must be <code>lambda</code>.<a class="headerlink" href="#type" title="Permanent link">&para;</a></h6>
 <h6 class="reference-item" id="regions"><code>regions</code> - (Optional) The list of Lambda regions. Omit to use all regions.<a class="headerlink" href="#regions" title="Permanent link">&para;</a></h6>