Google Git
Sign in
apache / aurora-website / refs/heads/asf-site / . / content / documentation / latest / reference / configuration / index.html
blob: c0f4f281346f1275b0cc182135e1474a41d22a73 [file] [log] [blame]
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>Apache Aurora</title>
<link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.1/css/bootstrap.min.css">
<link href="/assets/css/main.css" rel="stylesheet">
<!-- Analytics -->
<script type="text/javascript">
var _gaq = _gaq || [];
_gaq.push(['_setAccount', 'UA-45879646-1']);
_gaq.push(['_setDomainName', 'apache.org']);
_gaq.push(['_trackPageview']);
(function() {
var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true;
ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js';
var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s);
})();
</script>
</head>
<body>
<div class="container-fluid section-header">
<div class="container">
<div class="nav nav-bar">
<a href="/"><img src="/assets/img/aurora_logo_dkbkg.svg" width="300" alt="Transparent Apache Aurora logo with dark background"/></a>
<ul class="nav navbar-nav navbar-right">
<li><a href="/documentation/latest/">Documentation</a></li>
<li><a href="/community/">Community</a></li>
<li><a href="/downloads/">Downloads</a></li>
<li><a href="/blog/">Blog</a></li>
</ul>
</div>
</div>
</div>
<div class="container-fluid">
<div class="container content">
<div class="col-md-12 documentation">
<h5 class="page-header text-uppercase">Documentation
<select onChange="window.location.href='/documentation/' + this.value + '/reference/configuration/'"
value="latest">
<option value="0.22.0"
>
0.22.0
(latest)
</option>
<option value="0.21.0"
>
0.21.0
</option>
<option value="0.20.0"
>
0.20.0
</option>
<option value="0.19.1"
>
0.19.1
</option>
<option value="0.19.0"
>
0.19.0
</option>
<option value="0.18.1"
>
0.18.1
</option>
<option value="0.18.0"
>
0.18.0
</option>
<option value="0.17.0"
>
0.17.0
</option>
<option value="0.16.0"
>
0.16.0
</option>
<option value="0.15.0"
>
0.15.0
</option>
<option value="0.14.0"
>
0.14.0
</option>
<option value="0.13.0"
>
0.13.0
</option>
<option value="0.12.0"
>
0.12.0
</option>
<option value="0.11.0"
>
0.11.0
</option>
<option value="0.10.0"
>
0.10.0
</option>
<option value="0.9.0"
>
0.9.0
</option>
<option value="0.8.0"
>
0.8.0
</option>
<option value="0.7.0-incubating"
>
0.7.0-incubating
</option>
<option value="0.6.0-incubating"
>
0.6.0-incubating
</option>
<option value="0.5.0-incubating"
>
0.5.0-incubating
</option>
</select>
</h5>
<h1 id="aurora-configuration-reference">Aurora Configuration Reference</h1>
<p>Don&rsquo;t know where to start? The Aurora configuration schema is very
powerful, and configurations can become quite complex for advanced use
cases.</p>
<p>For examples of simple configurations to get something up and running
quickly, check out the <a href="../../getting-started/tutorial/">Tutorial</a>. When you feel comfortable with the basics, move
on to the <a href="../configuration-tutorial/">Configuration Tutorial</a> for more in-depth coverage of
configuration design.</p>
<ul>
<li><a href="#process-schema">Process Schema</a>
<ul>
<li><a href="#process-objects">Process Objects</a></li>
</ul></li>
<li><a href="#task-schema">Task Schema</a>
<ul>
<li><a href="#task-object">Task Object</a></li>
<li><a href="#constraint-object">Constraint Object</a></li>
<li><a href="#resource-object">Resource Object</a></li>
</ul></li>
<li><a href="#job-schema">Job Schema</a>
<ul>
<li><a href="#job-objects">Job Objects</a></li>
<li><a href="#updateconfig-objects">UpdateConfig Objects</a></li>
<li><a href="#healthcheckconfig-objects">HealthCheckConfig Objects</a></li>
<li><a href="#announcer-objects">Announcer Objects</a></li>
<li><a href="#container">Container Objects</a></li>
<li><a href="#lifecycleconfig-objects">LifecycleConfig Objects</a></li>
<li><a href="#slapolicy-objects">SlaPolicy Objects</a></li>
</ul></li>
<li><a href="#specifying-scheduling-constraints">Specifying Scheduling Constraints</a></li>
<li><a href="#template-namespaces">Template Namespaces</a>
<ul>
<li><a href="#mesos-namespace">mesos Namespace</a></li>
<li><a href="#thermos-namespace">thermos Namespace</a></li>
</ul></li>
</ul>
<h1 id="process-schema">Process Schema</h1>
<p>Process objects consist of required <code>name</code> and <code>cmdline</code> attributes. You can customize Process
behavior with its optional attributes. Remember, Processes are handled by Thermos.</p>
<h3 id="process-objects">Process Objects</h3>
<table><thead>
<tr>
<th><strong>Attribute Name</strong></th>
<th style="text-align: center"><strong>Type</strong></th>
<th><strong>Description</strong></th>
</tr>
</thead><tbody>
<tr>
<td><strong>name</strong></td>
<td style="text-align: center">String</td>
<td>Process name (Required)</td>
</tr>
<tr>
<td><strong>cmdline</strong></td>
<td style="text-align: center">String</td>
<td>Command line (Required)</td>
</tr>
<tr>
<td><strong>max_failures</strong></td>
<td style="text-align: center">Integer</td>
<td>Maximum process failures (Default: 1)</td>
</tr>
<tr>
<td><strong>daemon</strong></td>
<td style="text-align: center">Boolean</td>
<td>When True, this is a daemon process. (Default: False)</td>
</tr>
<tr>
<td><strong>ephemeral</strong></td>
<td style="text-align: center">Boolean</td>
<td>When True, this is an ephemeral process. (Default: False)</td>
</tr>
<tr>
<td><strong>min_duration</strong></td>
<td style="text-align: center">Integer</td>
<td>Minimum duration between process restarts in seconds. (Default: 5)</td>
</tr>
<tr>
<td><strong>final</strong></td>
<td style="text-align: center">Boolean</td>
<td>When True, this process is a finalizing one that should run last. (Default: False)</td>
</tr>
<tr>
<td><strong>logger</strong></td>
<td style="text-align: center">Logger</td>
<td>Struct defining the log behavior for the process. (Default: Empty)</td>
</tr>
</tbody></table>
<h4 id="name">name</h4>
<p>The name is any valid UNIX filename string (specifically no
slashes, NULLs or leading periods). Within a Task object, each Process name
must be unique.</p>
<h4 id="cmdline">cmdline</h4>
<p>The command line run by the process. The command line is invoked in a bash
subshell, so can involve fully-blown bash scripts. However, nothing is
supplied for command-line arguments so <code>$*</code> is unspecified.</p>
<h4 id="max_failures">max_failures</h4>
<p>The maximum number of failures (non-zero exit statuses) this process can
have before being marked permanently failed and not retried. If a
process permanently fails, Thermos looks at the failure limit of the task
containing the process (usually 1) to determine if the task has
failed as well.</p>
<p>Setting <code>max_failures</code> to 0 makes the process retry
indefinitely until it achieves a successful (zero) exit status.
It retries at most once every <code>min_duration</code> seconds to prevent
an effective denial of service attack on the coordinating Thermos scheduler.</p>
<h4 id="daemon">daemon</h4>
<p>By default, Thermos processes are non-daemon. If <code>daemon</code> is set to True, a
successful (zero) exit status does not prevent future process runs.
Instead, the process reinvokes after <code>min_duration</code> seconds.
However, the maximum failure limit still applies. A combination of
<code>daemon=True</code> and <code>max_failures=0</code> causes a process to retry
indefinitely regardless of exit status. This should be avoided
for very short-lived processes because of the accumulation of
checkpointed state for each process run. When running in Mesos
specifically, <code>max_failures</code> is capped at 100.</p>
<h4 id="ephemeral">ephemeral</h4>
<p>By default, Thermos processes are non-ephemeral. If <code>ephemeral</code> is set to
True, the process&rsquo; status is not used to determine if its containing task
has completed. For example, consider a task with a non-ephemeral
webserver process and an ephemeral logsaver process
that periodically checkpoints its log files to a centralized data store.
The task is considered finished once the webserver process has
completed, regardless of the logsaver&rsquo;s current status.</p>
<h4 id="min_duration">min_duration</h4>
<p>Processes may succeed or fail multiple times during a single task&rsquo;s
duration. Each of these is called a <em>process run</em>. <code>min_duration</code> is
the minimum number of seconds the scheduler waits before running the
same process.</p>
<h4 id="final">final</h4>
<p>Processes can be grouped into two classes: ordinary processes and
finalizing processes. By default, Thermos processes are ordinary. They
run as long as the task is considered healthy (i.e., no failure
limits have been reached.) But once all regular Thermos processes
finish or the task reaches a certain failure threshold, it
moves into a &ldquo;finalization&rdquo; stage and runs all finalizing
processes. These are typically processes necessary for cleaning up the
task, such as log checkpointers, or perhaps e-mail notifications that
the task completed.</p>
<p>Finalizing processes may not depend upon ordinary processes or
vice-versa, however finalizing processes may depend upon other
finalizing processes and otherwise run as a typical process
schedule.</p>
<h4 id="logger">logger</h4>
<p>The default behavior of Thermos is to store stderr/stdout logs in files which grow unbounded.
In the event that you have large log volume, you may want to configure Thermos to automatically
rotate logs after they grow to a certain size, which can prevent your job from using more than its
allocated disk space.</p>
<p>Logger objects specify a <code>destination</code> for Process logs which is, by default, <code>file</code> - a pair of
<code>stdout</code> and <code>stderr</code> files. Its also possible to specify <code>console</code> to get logs output to
the Process stdout and stderr streams, <code>none</code> to suppress any logs output or <code>both</code> to send logs to
files and console streams.</p>
<p>The default Logger <code>mode</code> is <code>standard</code> which lets the stdout and stderr streams grow without bound.</p>
<table><thead>
<tr>
<th><strong>Attribute Name</strong></th>
<th style="text-align: center"><strong>Type</strong></th>
<th><strong>Description</strong></th>
</tr>
</thead><tbody>
<tr>
<td><strong>destination</strong></td>
<td style="text-align: center">LoggerDestination</td>
<td>Destination of logs. (Default: <code>file</code>)</td>
</tr>
<tr>
<td><strong>mode</strong></td>
<td style="text-align: center">LoggerMode</td>
<td>Mode of the logger. (Default: <code>standard</code>)</td>
</tr>
<tr>
<td><strong>rotate</strong></td>
<td style="text-align: center">RotatePolicy</td>
<td>An optional rotation policy. (Default: <code>Empty</code>)</td>
</tr>
</tbody></table>
<p>A RotatePolicy describes log rotation behavior for when <code>mode</code> is set to <code>rotate</code> and it is ignored
otherwise. If <code>rotate</code> is <code>Empty</code> or <code>RotatePolicy()</code> when the <code>mode</code> is set to <code>rotate</code> the
defaults below are used.</p>
<table><thead>
<tr>
<th><strong>Attribute Name</strong></th>
<th style="text-align: center"><strong>Type</strong></th>
<th><strong>Description</strong></th>
</tr>
</thead><tbody>
<tr>
<td><strong>log_size</strong></td>
<td style="text-align: center">Integer</td>
<td>Maximum size (in bytes) of an individual log file. (Default: 100 MiB)</td>
</tr>
<tr>
<td><strong>backups</strong></td>
<td style="text-align: center">Integer</td>
<td>The maximum number of backups to retain. (Default: 5)</td>
</tr>
</tbody></table>
<p>An example process configuration is as follows:</p>
<pre class="highlight plaintext"><code> process = Process(
name='process',
logger=Logger(
destination=LoggerDestination('both'),
mode=LoggerMode('rotate'),
rotate=RotatePolicy(log_size=5*MB, backups=5)
)
)
</code></pre>
<h1 id="task-schema">Task Schema</h1>
<p>Tasks fundamentally consist of a <code>name</code> and a list of Process objects stored as the
value of the <code>processes</code> attribute. Processes can be further constrained with
<code>constraints</code>. By default, <code>name</code>&rsquo;s value inherits from the first Process in the
<code>processes</code> list, so for simple <code>Task</code> objects with one Process, <code>name</code>
can be omitted. In Mesos, <code>resources</code> is also required.</p>
<h3 id="task-object">Task Object</h3>
<table><thead>
<tr>
<th><strong>param</strong></th>
<th style="text-align: center"><strong>type</strong></th>
<th><strong>description</strong></th>
</tr>
</thead><tbody>
<tr>
<td><code>name</code></td>
<td style="text-align: center">String</td>
<td>Process name (Required) (Default: <code>processes0.name</code>)</td>
</tr>
<tr>
<td><code>processes</code></td>
<td style="text-align: center">List of <code>Process</code> objects</td>
<td>List of <code>Process</code> objects bound to this task. (Required)</td>
</tr>
<tr>
<td><code>constraints</code></td>
<td style="text-align: center">List of <code>Constraint</code> objects</td>
<td>List of <code>Constraint</code> objects constraining processes.</td>
</tr>
<tr>
<td><code>resources</code></td>
<td style="text-align: center"><code>Resource</code> object</td>
<td>Resource footprint. (Required)</td>
</tr>
<tr>
<td><code>max_failures</code></td>
<td style="text-align: center">Integer</td>
<td>Maximum process failures before being considered failed (Default: 1)</td>
</tr>
<tr>
<td><code>max_concurrency</code></td>
<td style="text-align: center">Integer</td>
<td>Maximum number of concurrent processes (Default: 0, unlimited concurrency.)</td>
</tr>
<tr>
<td><code>finalization_wait</code></td>
<td style="text-align: center">Integer</td>
<td>Amount of time allocated for finalizing processes, in seconds. (Default: 30)</td>
</tr>
</tbody></table>
<h4 id="name">name</h4>
<p><code>name</code> is a string denoting the name of this task. It defaults to the name of the first Process in
the list of Processes associated with the <code>processes</code> attribute.</p>
<h4 id="processes">processes</h4>
<p><code>processes</code> is an unordered list of <code>Process</code> objects. To constrain the order
in which they run, use <code>constraints</code>.</p>
<h5 id="constraints">constraints</h5>
<p>A list of <code>Constraint</code> objects. Currently it supports only one type,
the <code>order</code> constraint. <code>order</code> is a list of process names
that should run in the order given. For example,</p>
<pre class="highlight plaintext"><code> process = Process(cmdline = "echo hello {{name}}")
task = Task(name = "echoes",
processes = [process(name = "jim"), process(name = "bob")],
constraints = [Constraint(order = ["jim", "bob"]))
</code></pre>
<p>Constraints can be supplied ad-hoc and in duplicate. Not all
Processes need be constrained, however Tasks with cycles are
rejected by the Thermos scheduler.</p>
<p>Use the <code>order</code> function as shorthand to generate <code>Constraint</code> lists.
The following:</p>
<pre class="highlight plaintext"><code> order(process1, process2)
</code></pre>
<p>is shorthand for</p>
<pre class="highlight plaintext"><code> [Constraint(order = [process1.name(), process2.name()])]
</code></pre>
<p>The <code>order</code> function accepts Process name strings <code>(&#39;foo&#39;, &#39;bar&#39;)</code> or the processes
themselves, e.g. <code>foo=Process(name=&#39;foo&#39;, ...)</code>, <code>bar=Process(name=&#39;bar&#39;, ...)</code>,
<code>constraints=order(foo, bar)</code>.</p>
<h4 id="resources">resources</h4>
<p>Takes a <code>Resource</code> object, which specifies the amounts of CPU, memory, and disk space resources
to allocate to the Task.</p>
<h4 id="max_failures">max_failures</h4>
<p><code>max_failures</code> is the number of failed processes needed for the <code>Task</code> to be
marked as failed.</p>
<p>For example, assume a Task has two Processes and a <code>max_failures</code> value of <code>2</code>:</p>
<pre class="highlight plaintext"><code> template = Process(max_failures=10)
task = Task(
name = "fail",
processes = [
template(name = "failing", cmdline = "exit 1"),
template(name = "succeeding", cmdline = "exit 0")
],
max_failures=2)
</code></pre>
<p>The <code>failing</code> Process could fail 10 times before being marked as permanently
failed, and the <code>succeeding</code> Process could succeed on the first run. However,
the task would succeed despite only allowing for two failed processes. To be more
specific, there would be 10 failed process runs yet 1 failed process. Both processes
would have to fail for the Task to fail.</p>
<h4 id="max_concurrency">max_concurrency</h4>
<p>For Tasks with a number of expensive but otherwise independent
processes, you may want to limit the amount of concurrency
the Thermos scheduler provides rather than artificially constraining
it via <code>order</code> constraints. For example, a test framework may
generate a task with 100 test run processes, but wants to run it on
a machine with only 4 cores. You can limit the amount of parallelism to
4 by setting <code>max_concurrency=4</code> in your task configuration.</p>
<p>For example, the following task spawns 180 Processes (&ldquo;mappers&rdquo;)
to compute individual elements of a 180 degree sine table, all dependent
upon one final Process (&ldquo;reducer&rdquo;) to tabulate the results:</p>
<pre class="highlight plaintext"><code>def make_mapper(id):
return Process(
name = "mapper%03d" % id,
cmdline = "echo 'scale=50;s(%d\*4\*a(1)/180)' | bc -l &gt;
temp.sine_table.%03d" % (id, id))
def make_reducer():
return Process(name = "reducer", cmdline = "cat temp.\* | nl \&gt; sine\_table.txt
&amp;&amp; rm -f temp.\*")
processes = map(make_mapper, range(180))
task = Task(
name = "mapreduce",
processes = processes + [make\_reducer()],
constraints = [Constraint(order = [mapper.name(), 'reducer']) for mapper
in processes],
max_concurrency = 8)
</code></pre>
<h4 id="finalization_wait">finalization_wait</h4>
<p>Process execution is organizued into three active stages: <code>ACTIVE</code>,
<code>CLEANING</code>, and <code>FINALIZING</code>. The <code>ACTIVE</code> stage is when ordinary processes run.
This stage lasts as long as Processes are running and the Task is healthy.
The moment either all Processes have finished successfully or the Task has reached a
maximum Process failure limit, it goes into <code>CLEANING</code> stage and send
SIGTERMs to all currently running Processes and their process trees.
Once all Processes have terminated, the Task goes into <code>FINALIZING</code> stage
and invokes the schedule of all Processes with the &ldquo;final&rdquo; attribute set to True.</p>
<p>This whole process from the end of <code>ACTIVE</code> stage to the end of <code>FINALIZING</code>
must happen within <code>finalization_wait</code> seconds. If it does not
finish during that time, all remaining Processes are sent SIGKILLs
(or if they depend upon uncompleted Processes, are
never invoked.)</p>
<p>When running on Aurora, the <code>finalization_wait</code> is capped at 60 seconds.</p>
<h3 id="constraint-object">Constraint Object</h3>
<p>Current constraint objects only support a single ordering constraint, <code>order</code>,
which specifies its processes run sequentially in the order given. By
default, all processes run in parallel when bound to a <code>Task</code> without
ordering constraints.</p>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td>order</td>
<td style="text-align: center">List of String</td>
<td>List of processes by name (String) that should be run serially.</td>
</tr>
</tbody></table>
<h3 id="resource-object">Resource Object</h3>
<p>Specifies the amount of CPU, Ram, and disk resources the task needs. See the
<a href="../../features/resource-isolation/">Resource Isolation document</a> for suggested values and to understand how
resources are allocated.</p>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>cpu</code></td>
<td style="text-align: center">Float</td>
<td>Fractional number of cores required by the task.</td>
</tr>
<tr>
<td><code>ram</code></td>
<td style="text-align: center">Integer</td>
<td>Bytes of RAM required by the task.</td>
</tr>
<tr>
<td><code>disk</code></td>
<td style="text-align: center">Integer</td>
<td>Bytes of disk required by the task.</td>
</tr>
<tr>
<td><code>gpu</code></td>
<td style="text-align: center">Integer</td>
<td>Number of GPU cores required by the task</td>
</tr>
</tbody></table>
<h1 id="job-schema">Job Schema</h1>
<h3 id="job-objects">Job Objects</h3>
<p><em>Note: Specifying a <code>Container</code> object as the value of the <code>container</code> property is
deprecated in favor of setting its value directly to the appropriate <code>Docker</code> or <code>Mesos</code>
container type</em></p>
<p><em>Note: Specifying preemption behavior of tasks through <code>production</code> flag is deprecated in favor of
electing appropriate task tier via <code>tier</code> attribute.</em></p>
<table><thead>
<tr>
<th>name</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>task</code></td>
<td style="text-align: center">Task</td>
<td>The Task object to bind to this job. Required.</td>
</tr>
<tr>
<td><code>name</code></td>
<td style="text-align: center">String</td>
<td>Job name. (Default: inherited from the task attribute&rsquo;s name)</td>
</tr>
<tr>
<td><code>role</code></td>
<td style="text-align: center">String</td>
<td>Job role account. Required.</td>
</tr>
<tr>
<td><code>cluster</code></td>
<td style="text-align: center">String</td>
<td>Cluster in which this job is scheduled. Required.</td>
</tr>
<tr>
<td><code>environment</code></td>
<td style="text-align: center">String</td>
<td>Job environment, default <code>devel</code>. By default must be one of <code>prod</code>, <code>devel</code>, <code>test</code> or <code>staging&lt;number&gt;</code> but it can be changed by the Cluster operator using the scheduler option <code>allowed_job_environments</code>.</td>
</tr>
<tr>
<td><code>contact</code></td>
<td style="text-align: center">String</td>
<td>Best email address to reach the owner of the job. For production jobs, this is usually a team mailing list.</td>
</tr>
<tr>
<td><code>instances</code></td>
<td style="text-align: center">Integer</td>
<td>Number of instances (sometimes referred to as replicas or shards) of the task to create. (Default: 1)</td>
</tr>
<tr>
<td><code>cron_schedule</code></td>
<td style="text-align: center">String</td>
<td>Cron schedule in cron format. May only be used with non-service jobs. See <a href="../../features/cron-jobs/">Cron Jobs</a> for more information. Default: None (not a cron job.)</td>
</tr>
<tr>
<td><code>cron_collision_policy</code></td>
<td style="text-align: center">String</td>
<td>Policy to use when a cron job is triggered while a previous run is still active. KILL_EXISTING Kill the previous run, and schedule the new run CANCEL_NEW Let the previous run continue, and cancel the new run. (Default: KILL_EXISTING)</td>
</tr>
<tr>
<td><code>update_config</code></td>
<td style="text-align: center"><code>UpdateConfig</code> object</td>
<td>Parameters for controlling the rate and policy of rolling updates.</td>
</tr>
<tr>
<td><code>constraints</code></td>
<td style="text-align: center">dict</td>
<td>Scheduling constraints for the tasks. See the section on the <a href="#specifying-scheduling-constraints">constraint specification language</a></td>
</tr>
<tr>
<td><code>service</code></td>
<td style="text-align: center">Boolean</td>
<td>If True, restart tasks regardless of success or failure. (Default: False)</td>
</tr>
<tr>
<td><code>max_task_failures</code></td>
<td style="text-align: center">Integer</td>
<td>Maximum number of failures after which the task is considered to have failed (Default: 1) Set to -1 to allow for infinite failures</td>
</tr>
<tr>
<td><code>priority</code></td>
<td style="text-align: center">Integer</td>
<td>Preemption priority to give the task (Default 0). Tasks with higher priorities may preempt tasks at lower priorities.</td>
</tr>
<tr>
<td><code>production</code></td>
<td style="text-align: center">Boolean</td>
<td>(Deprecated) Whether or not this is a production task that may <a href="../../features/multitenancy/#preemption">preempt</a> other tasks (Default: False). Production job role must have the appropriate <a href="../../features/multitenancy/#preemption">quota</a>.</td>
</tr>
<tr>
<td><code>health_check_config</code></td>
<td style="text-align: center"><code>HealthCheckConfig</code> object</td>
<td>Parameters for controlling a task&rsquo;s health checks. HTTP health check is only used if a health port was assigned with a command line wildcard.</td>
</tr>
<tr>
<td><code>container</code></td>
<td style="text-align: center">Choice of <code>Container</code>, <code>Docker</code> or <code>Mesos</code> object</td>
<td>An optional container to run all processes inside of.</td>
</tr>
<tr>
<td><code>lifecycle</code></td>
<td style="text-align: center"><code>LifecycleConfig</code> object</td>
<td>An optional task lifecycle configuration that dictates commands to be executed on startup/teardown. HTTP lifecycle is enabled by default if the &ldquo;health&rdquo; port is requested. See <a href="#lifecycleconfig-objects">LifecycleConfig Objects</a> for more information.</td>
</tr>
<tr>
<td><code>tier</code></td>
<td style="text-align: center">String</td>
<td>Task tier type. The default scheduler tier configuration allows for 3 tiers: <code>revocable</code>, <code>preemptible</code>, and <code>preferred</code>. If a tier is not elected, Aurora assigns the task to a tier based on its choice of <code>production</code> (that is <code>preferred</code> for production and <code>preemptible</code> for non-production jobs). See the section on <a href="../../features/multitenancy/#configuration-tiers">Configuration Tiers</a> for more information.</td>
</tr>
<tr>
<td><code>announce</code></td>
<td style="text-align: center"><code>Announcer</code> object</td>
<td>Optionally enable Zookeeper ServerSet announcements. See [Announcer Objects] for more information.</td>
</tr>
<tr>
<td><code>enable_hooks</code></td>
<td style="text-align: center">Boolean</td>
<td>Whether to enable <a href="../client-hooks/">Client Hooks</a> for this job. (Default: False)</td>
</tr>
<tr>
<td><code>partition_policy</code></td>
<td style="text-align: center"><code>PartitionPolicy</code> object</td>
<td>An optional partition policy that allows job owners to define how to handle partitions for running tasks (in partition-aware Aurora clusters)</td>
</tr>
<tr>
<td><code>metadata</code></td>
<td style="text-align: center">list of <code>Metadata</code> objects</td>
<td>list of <code>Metadata</code> objects for user&rsquo;s customized metadata information.</td>
</tr>
<tr>
<td><code>executor_config</code></td>
<td style="text-align: center"><code>ExecutorConfig</code> object</td>
<td>Allows choosing an alternative executor defined in <code>custom_executor_config</code> to be used instead of Thermos. Tasks will be launched with Thermos as the executor by default. See <a href="../../features/custom-executors/">Custom Executors</a> for more info.</td>
</tr>
<tr>
<td><code>sla_policy</code></td>
<td style="text-align: center">Choice of <code>CountSlaPolicy</code>, <code>PercentageSlaPolicy</code> or <code>CoordinatorSlaPolicy</code> object</td>
<td>An optional SLA policy that allows job owners to describe the SLA requirements for the job. See <a href="#slapolicy-objects">SlaPolicy Objects</a> for more information.</td>
</tr>
</tbody></table>
<h3 id="updateconfig-objects">UpdateConfig Objects</h3>
<p>Parameters for controlling the rate and policy of rolling updates.</p>
<table><thead>
<tr>
<th>object</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>batch_size</code></td>
<td style="text-align: center">Integer</td>
<td>Maximum number of shards to be updated in one iteration (Default: 1)</td>
</tr>
<tr>
<td><code>watch_secs</code></td>
<td style="text-align: center">Integer</td>
<td>Minimum number of seconds a shard must remain in <code>RUNNING</code> state before considered a success (Default: 45)</td>
</tr>
<tr>
<td><code>max_per_shard_failures</code></td>
<td style="text-align: center">Integer</td>
<td>Maximum number of restarts per shard during update. Increments total failure count when this limit is exceeded. (Default: 0)</td>
</tr>
<tr>
<td><code>max_total_failures</code></td>
<td style="text-align: center">Integer</td>
<td>Maximum number of shard failures to be tolerated in total during an update. Cannot be greater than or equal to the total number of tasks in a job. (Default: 0)</td>
</tr>
<tr>
<td><code>rollback_on_failure</code></td>
<td style="text-align: center">boolean</td>
<td>When False, prevents auto rollback of a failed update (Default: True)</td>
</tr>
<tr>
<td><code>wait_for_batch_completion</code></td>
<td style="text-align: center">boolean</td>
<td>When True, all threads from a given batch will be blocked from picking up new instances until the entire batch is updated. This essentially simulates the legacy sequential updater algorithm. (Default: False)</td>
</tr>
<tr>
<td><code>pulse_interval_secs</code></td>
<td style="text-align: center">Integer</td>
<td>Indicates a <a href="../../features/job-updates/#coordinated-job-updates">coordinated update</a>. If no pulses are received within the provided interval the update will be blocked. Beta-updater only. Will fail on submission when used with client updater. (Default: None)</td>
</tr>
<tr>
<td><code>update_strategy</code></td>
<td style="text-align: center">Choice of <code>QueueUpdateStrategy</code>, <code>BatchUpdateStrategy</code>, or <code>VariableBatchUpdateStrategy</code> object</td>
<td>Indicate which update strategy to use for this update.</td>
</tr>
<tr>
<td><code>sla_aware</code></td>
<td style="text-align: center">boolean</td>
<td>When True, updates will only update an instance if it does not break the task&rsquo;s specified <a href="../../features/sla-requirements/">SLA Requirements</a>. (Default: None)</td>
</tr>
</tbody></table>
<h3 id="queueupdatestrategy-objects">QueueUpdateStrategy Objects</h3>
<p>Update strategy which will keep the active updating instances at size <code>batch_size</code> throughout the update until there are no more instances left to update.</p>
<table><thead>
<tr>
<th>object</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>batch_size</code></td>
<td style="text-align: center">Integer</td>
<td>Maximum number of shards to be updated in one iteration (Default: 1)</td>
</tr>
</tbody></table>
<h3 id="batchupdatestrategy-objects">BatchUpdateStrategy Objects</h3>
<p>Update strategy which will wait until a maximum of <code>batch_size</code> number of instances are updated before continuing on to the next group until all instances are updated.</p>
<table><thead>
<tr>
<th>object</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>batch_size</code></td>
<td style="text-align: center">Integer</td>
<td>Maximum number of shards to be updated in one iteration (Default: 1)</td>
</tr>
<tr>
<td><code>autopause_after_batch</code></td>
<td style="text-align: center">Boolean</td>
<td>Automatically pauses update after completing a batch. (Default: False)</td>
</tr>
</tbody></table>
<h3 id="variablebatchupdatestrategy-objects">VariableBatchUpdateStrategy Objects</h3>
<p>Similar to Batch Update strategy, this strategy will wait until all instances in a current group are
updated before updating more instances. However, instead of maintaining a static group size, the
size of each group may change as the update progresses. For example, an update which modifies a
total of 10 instances may be done in batch sizes of 2, 3, and 5. If the number of instances to
be updated are greater than the sum of the groups, the last group size will be used in
perpetuity until all instances are updated. Following the previous example, if instead of 10
instances 20 instances are modified, the update groups would become: 2, 3, 5, 5, 5.</p>
<table><thead>
<tr>
<th>object</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>batch_sizes</code></td>
<td style="text-align: center">List(Integer)</td>
<td>Maximum number of shards to be updated per iteration. As each iteration completes, the next iteration&rsquo;s group size may change. If there are still instances that need to be updated after all sizes are used, the last size will be reused for the remainder of the update.</td>
</tr>
<tr>
<td><code>autopause_after_batch</code></td>
<td style="text-align: center">Boolean</td>
<td>Automatically pauses update before starting a new batch. (Default: False)</td>
</tr>
</tbody></table>
<h4 id="using-the-sla_aware-option">Using the <code>sla_aware</code> option</h4>
<p>There are some nuances around the <code>sla_aware</code> option that users should be aware of:</p>
<ul>
<li>SLA-aware updates work in tandem with maintenance. Draining a host that has an instance of the
job being updated affects the SLA and thus will be taken into account when the update determines
whether or not it is safe to update another instance.</li>
<li>SLA-aware updates will use the <a href="../../features/sla-requirements/#custom-sla">SLAPolicy</a> of the
<em>newest</em> configuration when determining whether or not it is safe to update an instance. For
example, if the current configuration specifies a
<a href="../../features/sla-requirements/#percentageslapolicy-objects">PercentageSlaPolicy</a> that allows for
5% of instances to be down and the updated configuration increaes this value to 10%, the SLA
calculation will be done using the 10% policy. Be mindful of this when doing an update that
modifies the <code>SLAPolicy</code> since it may be possible to put the old configuration in a bad state
that the new configuration would not be affected by. Additionally, if the update is rolled back,
then the rollback will use the old <code>SLAPolicy</code> (or none if there was not one previously).</li>
<li>If using the <a href="../../features/sla-requirements/#coordinatorslapolicy-objects">CoordinatorSlaPolicy</a>,
it is important to pay attention to the <code>batch_size</code> of the update. If you have a complex SLA
requirement, then you may be limiting the throughput of your updates with an insufficient
<code>batch_size</code>. For example, imagine you have a job with 9 instance that represents three
replicated caches, and you can only update one instance per replica set: <code>[0 1 2]
[3 4 5] [6 7 8]</code> (the number indicates the instance ID and the brackets represent replica
sets). If your <code>batch_size</code> is 3, then you will slowly update one replica set at a time. If your
<code>batch_size</code> is 9, then you can update all replica sets in parallel and thus speeding up the update.</li>
<li>If an instance fails an SLA check for an update, then it will be rechecked starting at a delay
from <code>sla_aware_kill_retry_min_delay</code> and exponentially increasing up to
<code>sla_aware_kill_retry_max_delay</code>. These are cluster-operator set values.</li>
</ul>
<h3 id="healthcheckconfig-objects">HealthCheckConfig Objects</h3>
<p>Parameters for controlling a task&rsquo;s health checks via HTTP or a shell command.</p>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>health_checker</code></td>
<td style="text-align: center">HealthCheckerConfig</td>
<td>Configure what kind of health check to use.</td>
</tr>
<tr>
<td><code>initial_interval_secs</code></td>
<td style="text-align: center">Integer</td>
<td>Initial grace period (during which health-check failures are ignored) while performing health checks. (Default: 15)</td>
</tr>
<tr>
<td><code>interval_secs</code></td>
<td style="text-align: center">Integer</td>
<td>Interval on which to check the task&rsquo;s health. (Default: 10)</td>
</tr>
<tr>
<td><code>max_consecutive_failures</code></td>
<td style="text-align: center">Integer</td>
<td>Maximum number of consecutive failures that will be tolerated before considering a task unhealthy (Default: 0)</td>
</tr>
<tr>
<td><code>min_consecutive_successes</code></td>
<td style="text-align: center">Integer</td>
<td>Minimum number of consecutive successful health checks required before considering a task healthy (Default: 1)</td>
</tr>
<tr>
<td><code>timeout_secs</code></td>
<td style="text-align: center">Integer</td>
<td>Health check timeout. (Default: 1)</td>
</tr>
</tbody></table>
<h3 id="healthcheckerconfig-objects">HealthCheckerConfig Objects</h3>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>http</code></td>
<td style="text-align: center">HttpHealthChecker</td>
<td>Configure health check to use HTTP. (Default)</td>
</tr>
<tr>
<td><code>shell</code></td>
<td style="text-align: center">ShellHealthChecker</td>
<td>Configure health check via a shell command.</td>
</tr>
</tbody></table>
<h3 id="httphealthchecker-objects">HttpHealthChecker Objects</h3>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>endpoint</code></td>
<td style="text-align: center">String</td>
<td>HTTP endpoint to check (Default: /health)</td>
</tr>
<tr>
<td><code>expected_response</code></td>
<td style="text-align: center">String</td>
<td>If not empty, fail the HTTP health check if the response differs. Case insensitive. (Default: ok)</td>
</tr>
<tr>
<td><code>expected_response_code</code></td>
<td style="text-align: center">Integer</td>
<td>If not zero, fail the HTTP health check if the response code differs. (Default: 0)</td>
</tr>
</tbody></table>
<h3 id="shellhealthchecker-objects">ShellHealthChecker Objects</h3>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>shell_command</code></td>
<td style="text-align: center">String</td>
<td>An alternative to HTTP health checking. Specifies a shell command that will be executed. Any non-zero exit status will be interpreted as a health check failure.</td>
</tr>
</tbody></table>
<h3 id="partitionpolicy-objects">PartitionPolicy Objects</h3>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>reschedule</code></td>
<td style="text-align: center">Boolean</td>
<td>Whether or not to reschedule when running tasks become partitioned (Default: True)</td>
</tr>
<tr>
<td><code>delay_secs</code></td>
<td style="text-align: center">Integer</td>
<td>How long to delay transitioning to LOST when running tasks are partitioned. (Default: 0)</td>
</tr>
</tbody></table>
<h3 id="metadata-objects">Metadata Objects</h3>
<p>Describes a piece of user metadata in a key value pair</p>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>key</code></td>
<td style="text-align: center">String</td>
<td>Indicate which metadata the user provides</td>
</tr>
<tr>
<td><code>value</code></td>
<td style="text-align: center">String</td>
<td>Provide the metadata content for corresponding key</td>
</tr>
</tbody></table>
<h3 id="executorconfig-objects">ExecutorConfig Objects</h3>
<p>Describes an Executor name and data to pass to the Mesos Task</p>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>name</code></td>
<td style="text-align: center">String</td>
<td>Name of the executor to use for this task. Must match the name of an executor in <code>custom_executor_config</code> or Thermos (<code>AuroraExecutor</code>). (Default: AuroraExecutor)</td>
</tr>
<tr>
<td><code>data</code></td>
<td style="text-align: center">String</td>
<td>Data blob to pass on to the executor. (Default: &ldquo;&rdquo;)</td>
</tr>
</tbody></table>
<h3 id="announcer-objects">Announcer Objects</h3>
<p>If the <code>announce</code> field in the Job configuration is set, each task will be
registered in the ServerSet <code>/aurora/role/environment/jobname</code> in the
zookeeper ensemble configured by the executor (which can be optionally overriden by specifying
<code>zk_path</code> parameter). If no Announcer object is specified,
no announcement will take place. For more information about ServerSets, see the <a href="../../features/service-discovery/">Service Discover</a>
documentation.</p>
<p>By default, the hostname in the registered endpoints will be the <code>--hostname</code> parameter
that is passed to the mesos agent. To override the hostname value, the executor can be started
with <code>--announcer-hostname=&lt;overriden_value&gt;</code>. If you decide to use <code>--announcer-hostname</code> and if
the overriden value needs to change for every executor, then the executor has to be started inside a wrapper, see <a href="../../operations/configuration/#thermos-executor-wrapper">Executor Wrapper</a>.</p>
<p>For example, if you want the hostname in the endpoint to be an IP address instead of the hostname,
the <code>--hostname</code> parameter to the mesos agent can be set to the machine IP or the executor can
be started with <code>--announcer-hostname=&lt;host_ip&gt;</code> while wrapping the executor inside a script.</p>
<table><thead>
<tr>
<th>object</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>primary_port</code></td>
<td style="text-align: center">String</td>
<td>Which named port to register as the primary endpoint in the ServerSet (Default: <code>http</code>)</td>
</tr>
<tr>
<td><code>portmap</code></td>
<td style="text-align: center">dict</td>
<td>A mapping of additional endpoints to be announced in the ServerSet (Default: <code>{ &#39;aurora&#39;: &#39;{{primary_port}}&#39; }</code>)</td>
</tr>
<tr>
<td><code>zk_path</code></td>
<td style="text-align: center">String</td>
<td>Zookeeper serverset path override (executor must be started with the <code>--announcer-allow-custom-serverset-path</code> parameter)</td>
</tr>
</tbody></table>
<h4 id="port-aliasing-with-the-announcer-portmap">Port aliasing with the Announcer <code>portmap</code></h4>
<p>The primary endpoint registered in the ServerSet is the one allocated to the port
specified by the <code>primary_port</code> in the <code>Announcer</code> object, by default
the <code>http</code> port. This port can be referenced from anywhere within a configuration
as <code>{{thermos.ports[http]}}</code>.</p>
<p>Without the port map, each named port would be allocated a unique port number.
The <code>portmap</code> allows two different named ports to be aliased together. The default
<code>portmap</code> aliases the <code>aurora</code> port (i.e. <code>{{thermos.ports[aurora]}}</code>) to
the <code>http</code> port. Even though the two ports can be referenced independently,
only one port is allocated by Mesos. Any port referenced in a <code>Process</code> object
but which is not in the portmap will be allocated dynamically by Mesos and announced as well.</p>
<p>It is possible to use the portmap to alias names to static port numbers, e.g.
<code>{&#39;http&#39;: 80, &#39;https&#39;: 443, &#39;aurora&#39;: &#39;http&#39;}</code>. In this case, referencing
<code>{{thermos.ports[aurora]}}</code> would look up <code>{{thermos.ports[http]}}</code> then
find a static port 80. No port would be requested of or allocated by Mesos.</p>
<p>Static ports should be used cautiously as Aurora does nothing to prevent two
tasks with the same static port allocations from being co-scheduled.
External constraints such as agent attributes should be used to enforce such
guarantees should they be needed.</p>
<h3 id="container-objects">Container Objects</h3>
<p>Describes the container the job&rsquo;s processes will run inside. If not using Docker or the Mesos
unified-container, the container can be omitted from your job config.</p>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>mesos</code></td>
<td style="text-align: center">Mesos</td>
<td>A native Mesos container to use.</td>
</tr>
<tr>
<td><code>docker</code></td>
<td style="text-align: center">Docker</td>
<td>A Docker container to use (via Docker engine)</td>
</tr>
</tbody></table>
<h3 id="mesos-object">Mesos Object</h3>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>image</code></td>
<td style="text-align: center">Choice(AppcImage, DockerImage)</td>
<td>An optional filesystem image to use within this container.</td>
</tr>
<tr>
<td><code>volumes</code></td>
<td style="text-align: center">List(Volume)</td>
<td>An optional list of volume mounts for this container.</td>
</tr>
</tbody></table>
<h3 id="volume-object">Volume Object</h3>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>container_path</code></td>
<td style="text-align: center">String</td>
<td>Path on the host to mount.</td>
</tr>
<tr>
<td><code>host_path</code></td>
<td style="text-align: center">String</td>
<td>Mount point in the container.</td>
</tr>
<tr>
<td><code>mode</code></td>
<td style="text-align: center">Enum</td>
<td>Mode of the mount, can be &lsquo;RW&rsquo; or &#39;RO&rsquo;.</td>
</tr>
</tbody></table>
<h3 id="appcimage">AppcImage</h3>
<p>Describes an AppC filesystem image.</p>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>name</code></td>
<td style="text-align: center">String</td>
<td>The name of the appc image.</td>
</tr>
<tr>
<td><code>image_id</code></td>
<td style="text-align: center">String</td>
<td>The <a href="https://github.com/appc/spec/blob/master/spec/aci.md#image-id">image id</a> of the appc image.</td>
</tr>
</tbody></table>
<h3 id="dockerimage">DockerImage</h3>
<p>Describes a Docker filesystem image.</p>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>name</code></td>
<td style="text-align: center">String</td>
<td>The name of the docker image.</td>
</tr>
<tr>
<td><code>tag</code></td>
<td style="text-align: center">String</td>
<td>The tag that identifies the docker image.</td>
</tr>
</tbody></table>
<h3 id="docker-object">Docker Object</h3>
<p><em>Note: In order to correctly execute processes inside a job, the Docker container must have Python 2.7 installed.</em>
<em>Note: For private docker registry, mesos mandates the docker credential file to be named as <code>.dockercfg</code>, even though docker may create a credential file with a different name on various platforms. Also, the <code>.dockercfg</code> file needs to be copied into the sandbox using the <code>-thermos_executor_resources</code> flag, specified while starting Aurora.</em></p>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>image</code></td>
<td style="text-align: center">String</td>
<td>The name of the docker image to execute. If the image does not exist locally it will be pulled with <code>docker pull</code>.</td>
</tr>
<tr>
<td><code>parameters</code></td>
<td style="text-align: center">List(Parameter)</td>
<td>Additional parameters to pass to the Docker engine.</td>
</tr>
</tbody></table>
<h3 id="docker-parameter-object">Docker Parameter Object</h3>
<p>Docker CLI parameters. This needs to be enabled by the scheduler <code>-allow_docker_parameters</code> option.
See <a href="https://docs.docker.com/reference/commandline/run/">Docker Command Line Reference</a> for valid parameters.</p>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>name</code></td>
<td style="text-align: center">String</td>
<td>The name of the docker parameter. E.g. volume</td>
</tr>
<tr>
<td><code>value</code></td>
<td style="text-align: center">String</td>
<td>The value of the parameter. E.g. /usr/local/bin:/usr/bin:rw</td>
</tr>
</tbody></table>
<h3 id="lifecycleconfig-objects">LifecycleConfig Objects</h3>
<p><em>Note: The only lifecycle configuration supported is the HTTP lifecycle via the HttpLifecycleConfig.</em></p>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>http</code></td>
<td style="text-align: center">HttpLifecycleConfig</td>
<td>Configure the lifecycle manager to send lifecycle commands to the task via HTTP.</td>
</tr>
</tbody></table>
<h3 id="httplifecycleconfig-objects">HttpLifecycleConfig Objects</h3>
<p><em>Note: The combined <code>graceful_shutdown_wait_secs</code> and <code>shutdown_wait_secs</code> is implicitly upper bounded by the <code>--stop_timeout_in_secs</code> flag exposed by the executor (see options <a href="https://github.com/apache/aurora/blob/master/src/main/python/apache/aurora/executor/bin/thermos_executor_main.py">here</a>, default is 2 minutes). Therefore, if the user specifies values that add up to more than <code>--stop_timeout_in_secs</code>, the task will be killed earlier than the user anticipates (see the termination lifecycle <a href="https://aurora.apache.org/documentation/latest/reference/task-lifecycle/#forceful-termination-killing-restarting">here</a>). Furthermore, <code>stop_timeout_in_secs</code> itself is implicitly upper bounded by two scheduler options: <code>transient_task_state_timeout</code> and <code>preemption_slot_hold_time</code> (see reference <a href="http://aurora.apache.org/documentation/latest/reference/scheduler-configuration/">here</a>. If the <code>stop_timeout_in_secs</code> exceeds either of these scheduler options, tasks could be designated as LOST or tasks utilizing preemption could lose their desired slot respectively. Cluster operators should be aware of these timings should they change the defaults.</em></p>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>port</code></td>
<td style="text-align: center">String</td>
<td>The named port to send POST commands. (Default: health)</td>
</tr>
<tr>
<td><code>graceful_shutdown_endpoint</code></td>
<td style="text-align: center">String</td>
<td>Endpoint to hit to indicate that a task should gracefully shutdown. (Default: /quitquitquit)</td>
</tr>
<tr>
<td><code>shutdown_endpoint</code></td>
<td style="text-align: center">String</td>
<td>Endpoint to hit to give a task its final warning before being killed. (Default: /abortabortabort)</td>
</tr>
<tr>
<td><code>graceful_shutdown_wait_secs</code></td>
<td style="text-align: center">Integer</td>
<td>The amount of time (in seconds) to wait after hitting the <code>graceful_shutdown_endpoint</code> before proceeding with the <a href="https://aurora.apache.org/documentation/latest/reference/task-lifecycle/#forceful-termination-killing-restarting">task termination lifecycle</a>. (Default: 5)</td>
</tr>
<tr>
<td><code>shutdown_wait_secs</code></td>
<td style="text-align: center">Integer</td>
<td>The amount of time (in seconds) to wait after hitting the <code>shutdown_endpoint</code> before proceeding with the <a href="https://aurora.apache.org/documentation/latest/reference/task-lifecycle/#forceful-termination-killing-restarting">task termination lifecycle</a>. (Default: 5)</td>
</tr>
</tbody></table>
<h4 id="graceful_shutdown_endpoint">graceful_shutdown_endpoint</h4>
<p>If the Job is listening on the port as specified by the HttpLifecycleConfig
(default: <code>health</code>), a HTTP POST request will be sent over localhost to this
endpoint to request that the task gracefully shut itself down. This is a
courtesy call before the <code>shutdown_endpoint</code> is invoked
<code>graceful_shutdown_wait_secs</code> seconds later.</p>
<h4 id="shutdown_endpoint">shutdown_endpoint</h4>
<p>If the Job is listening on the port as specified by the HttpLifecycleConfig
(default: <code>health</code>), a HTTP POST request will be sent over localhost to this
endpoint to request as a final warning before being shut down. If the task
does not shut down on its own after <code>shutdown_wait_secs</code> seconds, it will be
forcefully killed.</p>
<h3 id="slapolicy-objects">SlaPolicy Objects</h3>
<p>Configuration for specifying custom <a href="../../features/sla-requirements/">SLA requirements</a> for a job. There are 3 supported SLA policies
namely, <a href="#countslapolicy-objects"><code>CountSlaPolicy</code></a>, <a href="#percentageslapolicy-objects"><code>PercentageSlaPolicy</code></a> and <a href="#coordinatorslapolicy-objects"><code>CoordinatorSlaPolicy</code></a>.</p>
<h3 id="countslapolicy-objects">CountSlaPolicy Objects</h3>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>count</code></td>
<td style="text-align: center">Integer</td>
<td>The number of active instances required every <code>durationSecs</code>.</td>
</tr>
<tr>
<td><code>duration_secs</code></td>
<td style="text-align: center">Integer</td>
<td>Minimum time duration a task needs to be <code>RUNNING</code> to be treated as active.</td>
</tr>
</tbody></table>
<h3 id="percentageslapolicy-objects">PercentageSlaPolicy Objects</h3>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>percentage</code></td>
<td style="text-align: center">Float</td>
<td>The percentage of active instances required every <code>durationSecs</code>.</td>
</tr>
<tr>
<td><code>duration_secs</code></td>
<td style="text-align: center">Integer</td>
<td>Minimum time duration a task needs to be <code>RUNNING</code> to be treated as active.</td>
</tr>
</tbody></table>
<h3 id="coordinatorslapolicy-objects">CoordinatorSlaPolicy Objects</h3>
<table><thead>
<tr>
<th>param</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>coordinator_url</code></td>
<td style="text-align: center">String</td>
<td>The URL to the <a href="../../features/sla-requirements/#coordinator">Coordinator</a> service to be contacted before performing SLA affecting actions (job updates, host drains etc).</td>
</tr>
<tr>
<td><code>status_key</code></td>
<td style="text-align: center">String</td>
<td>The field in the Coordinator response that indicates the SLA status for working on the task. (Default: <code>drain</code>)</td>
</tr>
</tbody></table>
<h1 id="specifying-scheduling-constraints">Specifying Scheduling Constraints</h1>
<p>In the <code>Job</code> object there is a map <code>constraints</code> from String to String
allowing the user to tailor the schedulability of tasks within the job.</p>
<p>The constraint map&rsquo;s key value is the attribute name in which we
constrain Tasks within our Job. The value is how we constrain them.
There are two types of constraints: <em>limit constraints</em> and <em>value
constraints</em>.</p>
<table><thead>
<tr>
<th>constraint</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td>Limit</td>
<td>A string that specifies a limit for a constraint. Starts with <code>&#39;limit:</code> followed by an Integer and closing single quote, such as <code>&#39;limit:1&#39;</code>.</td>
</tr>
<tr>
<td>Value</td>
<td>A string that specifies a value for a constraint. To include a list of values, separate the values using commas. To negate the values of a constraint, start with a <code>!</code> <code>.</code></td>
</tr>
</tbody></table>
<p>Further details can be found in the <a href="../../features/constraints/">Scheduling Constraints</a> feature
description.</p>
<h1 id="template-namespaces">Template Namespaces</h1>
<p>Currently, a few Pystachio namespaces have special semantics. Using them
in your configuration allow you to tailor application behavior
through environment introspection or interact in special ways with the
Aurora client or Aurora-provided services.</p>
<h3 id="mesos-namespace">mesos Namespace</h3>
<p>The <code>mesos</code> namespace contains variables which relate to the <code>mesos</code> agent
which launched the task. The <code>instance</code> variable can be used
to distinguish between Task replicas.</p>
<table><thead>
<tr>
<th>variable name</th>
<th style="text-align: center">type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td><code>instance</code></td>
<td style="text-align: center">Integer</td>
<td>The instance number of the created task. A job with 5 replicas has instance numbers 0, 1, 2, 3, and 4.</td>
</tr>
<tr>
<td><code>hostname</code></td>
<td style="text-align: center">String</td>
<td>The instance hostname that the task was launched on.</td>
</tr>
</tbody></table>
<p>Please note, there is no uniqueness guarantee for <code>instance</code> in the presence of
network partitions. If that is required, it should be baked in at the application
level using a distributed coordination service such as Zookeeper.</p>
<h3 id="thermos-namespace">thermos Namespace</h3>
<p>The <code>thermos</code> namespace contains variables that work directly on the
Thermos platform in addition to Aurora. This namespace is fully
compatible with Tasks invoked via the <code>thermos</code> CLI.</p>
<table><thead>
<tr>
<th style="text-align: center">variable</th>
<th>type</th>
<th>description</th>
</tr>
</thead><tbody>
<tr>
<td style="text-align: center"><code>ports</code></td>
<td>map of string to Integer</td>
<td>A map of names to port numbers</td>
</tr>
<tr>
<td style="text-align: center"><code>task_id</code></td>
<td>string</td>
<td>The task ID assigned to this task.</td>
</tr>
</tbody></table>
<p>The <code>thermos.ports</code> namespace is automatically populated by Aurora when
invoking tasks on Mesos. When running the <code>thermos</code> command directly,
these ports must be explicitly mapped with the <code>-P</code> option.</p>
<p>For example, if &rsquo;{{<code>thermos.ports[http]</code>}}&rsquo; is specified in a <code>Process</code>
configuration, it is automatically extracted and auto-populated by
Aurora, but must be specified with, for example, <code>thermos -P http:12345</code>
to map <code>http</code> to port 12345 when running via the CLI.</p>
</div>
</div>
</div>
<div class="container-fluid section-footer buffer">
<div class="container">
<div class="row">
<div class="col-md-2 col-md-offset-1"><h3>Quick Links</h3>
<ul>
<li><a href="/downloads/">Downloads</a></li>
<li><a href="/community/">Mailing Lists</a></li>
<li><a href="http://issues.apache.org/jira/browse/AURORA">Issue Tracking</a></li>
<li><a href="/documentation/latest/contributing/">How To Contribute</a></li>
</ul>
</div>
<div class="col-md-2"><h3>The ASF</h3>
<ul>
<li><a href="http://www.apache.org/licenses/">License</a></li>
<li><a href="http://www.apache.org/foundation/sponsorship.html">Sponsorship</a></li>
<li><a href="http://www.apache.org/foundation/thanks.html">Thanks</a></li>
<li><a href="http://www.apache.org/security/">Security</a></li>
</ul>
</div>
<div class="col-md-6">
<p class="disclaimer">&copy; 2014-2017 <a href="http://www.apache.org/">Apache Software Foundation</a>. Licensed under the <a href="http://www.apache.org/licenses/">Apache License v2.0</a>. The <a href="https://www.flickr.com/photos/trondk/12706051375/">Aurora Borealis IX photo</a> displayed on the homepage is available under a <a href="https://creativecommons.org/licenses/by-nc-nd/2.0/">Creative Commons BY-NC-ND 2.0 license</a>. Apache, Apache Aurora, and the Apache feather logo are trademarks of The Apache Software Foundation.</p>
</div>
</div>
</div>
</body>
</html>