Run Splunk Enterprise as a systemd service

Splunk Enterprise 7.2.2 and higher provides support for systemd on Linux with an enhanced enable boot-start command that lets you automatically configure systemd to manage splunkd as a service.

What is systemd?

systemd is a system startup and service manager that is widely deployed as the default init system on most major Linux distributions. You can configure systemd to manage processes, such as splunkd, as services, and allocate system resources to those processes under cgroups.

systemd advantages

systemd offers the following general advantages:

• Enhanced parallel processing.
• Simplified configuration with standardized unit text files. No scripts required.
• Improved mechanism for expressing dependencies. For example, you can specify in the unit file that the network must be up before startup of the splunkd service occurs.

systemd offers these additional specific advantages for Splunk deployments:

• Start splunkd at boot.
• Monitor and manage splunkd service during runtime.
• Provides tools to debug and troubleshoot boot-time and service activities.
• Allows more control over plug-in monitoring tools that track the status of Splunk instances.
• Simplifies the set up of cgroups required for workload management in Splunk Enterprise. See Set up Linux for workload management in the Workload Management manual.

Configure systemd to manage splunkd

You can use either of the following two methods to configure systemd to manage splunkd as a service:

• Configure systemd using enable boot-start.
• Configure systemd manually.

If you configure systemd using enable boot-start, a Splunk service unit file is created automatically. No additional manual configuration is required.

System requirements

• To run splunkd as a systemd service requires one of the following supported Linux distributions:
  • RHEL 7, 8, and 9
  • CentOS 7 and 8
  • Ubuntu 16.04 LTS and later
  • Suse 12
• To configure systemd using enable boot-start requires Splunk Enterprise version 7.2.2 or later.
• To enable workload management in Splunk Enterprise under systemd requires systemd version 219 or higher. For more information, see Linux operating system requirements in the Workload Management manual.
• Workload management supports Linux cgroups v1 by default.

Workload management support for Linux cgroups v2 is now available for Early Access customers in Splunk Enterprise version 9.3.0. In the Early Access release stage, Splunk products may have limitations on customer access, features, maturity and regional availability. For additional information on Early Access please contact your Splunk representative.

Permissions requirements

The enable boot-start command and systemd have the following permissions requirements:

• Non-root users must have super user permissions to configure systemd using enable boot-start.
• Non-root users must have super user permissions to run splunk start|stop|restart operations under systemd.

For instructions on how to create a new user with super user permissions, see your Linux documentation.

Unprivileged users must use sudo to run splunk start|stop|restart. If you do not use sudo and attempt to run splunk start|stop|restart when managed by systemd, a prompt appears requesting authentication. For example:

==== AUTHENTICATING FOR org.freedesktop.systemd1.manage-units ===
Authentication is required to manage system services or units.
Multiple identities can be used for authentication:
 1.  <username_1>
 2.  <username_2>
Choose identity to authenticate as (1-2): 2
Password: 
==== AUTHENTICATION COMPLETE ===

Alternately, you can install polkit rules with the enable boot-start command to allow unprivileged users to run start|stop|restart operations under systemd without using sudo. For instructions, see Install polkit rules to elevate user permissions.

Unit file naming considerations

The enable boot-start command creates a systemd unit file named Splunkd.service. The unit file name is based on the SPLUNK_SERVER_NAME in splunk-launch.conf, which is set by default to Splunkd.

If for any reason you remove the SPLUNK_SERVER_NAME value from splunk-launch.conf, enable boot-start creates a unit file named splunkd.service (lower case "splunkd") and sets SPLUNK_SERVER_NAME=splunkd in the splunk-launch.conf file.

You can specify a different name of your choice for the unit file when you run enable boot-start. See Specify a different unit file name.

Configure systemd using enable boot-start

You can configure systemd to manage splunkd as a service using the enable boot-start command, as follows:

1. Log into the machine on which you want to configure systemd to manage splunkd as a service.
2. Stop splunkd.

   $SPLUNK_HOME/bin/splunk stop
   

3. If you previously enabled Splunk Enterprise to start at boot using the enable boot-start command, run disable boot-start to remove the splunk init script located in /etc/init.d and its symbolic links.

   [sudo] $SPLUNK_HOME/bin/splunk disable boot-start
   

   For instructions on how to reinstall the splunk init script, see Install splunk init script.

4. Run the enable boot-start command, specifying the -systemd-managed, -user, and -group parameters, as follows:

   [sudo] $SPLUNK_HOME/bin/splunk enable boot-start -systemd-managed 1 -user <username> -group <groupname>
   

   Specifying -user and -group is optional but recommended. If you do not specify -user, the SPLUNK_OS_USER in splunk-launch.conf is used. If SPLUNK_OS_USER is not defined, the owner of the splunk binary is used.

   This installs the following systemd service unit file, named Splunkd.service by default, in /etc/systemd/system. To specify a different unit file name, use the -systemd-unit-file-name option. See Specify a different unit file name.

   #This unit file replaces the traditional start-up script for systemd
   #configurations, and is used when enabling boot-start for Splunk on
   #systemd-based Linux distributions.
   
   [Unit]
   Description=Systemd service file for Splunk, generated by 'splunk enable boot-start'
   After=network.target
   
   [Service]
   Type=simple
   Restart=always
   ExecStart=/opt/splunk/bin/splunk _internal_launch_under_systemd
   KillMode=mixed
   KillSignal=SIGINT
   TimeoutStopSec=360
   LimitNOFILE=65536
   SuccessExitStatus=51 52
   RestartPreventExitStatus=51
   RestartForceExitStatus=52
   User=splunk
   Group=splunk
   Delegate=true
   CPUShares=1024
   MemoryLimit=<value>
   PermissionsStartOnly=true
   ExecStartPost=/bin/bash -c "chown -R splunk:splunk /sys/fs/cgroup/cpu/system.slice/%n"
   ExecStartPost=/bin/bash -c "chown -R splunk:splunk /sys/fs/cgroup/memory/system.slice/%n"
   
   [Install]
   WantedBy=multi-user.target
   

   The MemoryLimit value is set to the total system memory available in bytes when the service unit file is created. The MemoryLimit value will not update if the total available system memory changes. To update the MemoryLimit value in the unit file, you can manually edit the value or use the boot-start command to disable and re-enable systemd.

   The following unit file properties are required. Do not change these values without appropriate guidance.
   Type=simple
Restart=always
ExecStart=$SPLUNK_HOME/bin/splunk _internal_launch_under_systemd
Delegate=true This property is required for workload management. See Configure workload management.

   Do not use the following properties. These properties can cause splunkd to fail on restart.
   RemainAfterExit=yes
ExecStop

   For more information, see Systemd unit file properties.

5. Start splunkd.

   [sudo] $SPLUNK_HOME/bin/splunk start
   

   This starts splunkd as a systemd service.

   Under systemd, splunk start|stop|restart commands are mapped to systemctl start|stop|restart commands.

6. Verify that splunkd is running as a systemd service. For example:

   $SPLUNK_HOME/bin/splunk status
   splunkd is running (PID: 24772).
   splunk helpers are running (PIDs: 24843 24857 24984 25032).
   

   Alternatively, you can use systemctl status to check if the splunkd process is running. However, when using this command, a brief time lag can occur during which systemctl status shows "active" and splunk status shows "splunkd is not running".

   Configuring systemd to manage splunkd as a service creates CPU and Memory cgroups in these locations:

   CPU: /sys/fs/cgroup/cpu/system.slice/Splunkd.service
   Memory: /sys/fs/cgroup/memory/system.slice/Splunkd.service
   

7. For distributed deployments, repeat steps 1-8 on all search heads and indexers.

Additional options for enable boot-start

The enable boot-start command supports these additional options:

Install splunk init script

In version 7.2.2 and higher, the enable boot-start command adds a -systemd-managed 0|1 option that controls whether to install the splunk init script in /etc/init.d or the Splunkd.service unit file in /etc/systemd/system.

To install the splunk init script, specify -systemd-managed 0:

$SPLUNK_HOME/bin/splunk enable boot-start -systemd-managed 0 -user <username>
Init script installed at /etc/init.d/splunk.
Init script is configured to run at boot.

See Configure Splunk Enterprise to start at boot time.

In version 7.2.2 through 7.2.x, if you do not specify the -systemd-managed option, the enable boot-start command defaults to -systemd-managed 1 and installs the Splunkd.service unit file. In version 7.3.0 and later, this default behavior is reversed, and the enable boot-start command defaults to -systemd-managed 0 and installs the splunkinit file.

The init.d boot-start script is not compatible with RHEL 8 and higher.

Specify a different unit file name

The default splunkd unit file name is Splunkd.service. You can specify a different name for the unit file and update the SPLUNK_SERVER_NAME value in splunk-launch.conf using the -systemd-unit-file-name option. For example, to create a unit file with the name "splunk.service":

$SPLUNK_HOME/bin/splunk enable boot-start -systemd-managed 1 -systemd-unit-file-name splunk  
Systemd unit file installed at /etc/systemd/system/splunk.service.
Configured as systemd managed service.

For more information, see Unit file naming considerations.

Install polkit rules to elevate user permissions

In version 8.1.1 and higher, the enable boot-start command adds an option to install polkit rules that allow non-root users to run start, stop, and restart operations under systemd without using sudo. Installing the polkit rules can reduce overhead for admins that must otherwise add unprivileged users to the sudoers file to run these operations under systemd.

To install polkit rules:

Run the enable boot-start command, specifying the -create-polkit-rules option, as follows:

./splunk enable boot-start -systemd-managed 1 -create-polkit-rules 1 -user <username>

If you previously ran enable boot-start and specified a different user, you must change the owner of $SPLUNK_HOME to the new user for whom you create the polkit rules. For example:

chown -R <username> $SPLUNK_HOME

Before you can install polkit rules using the create-polkit-rules option, you must install the Polkit library on your system if you have not already done so.

Configure systemd on a clean install

To configure systemd on a clean installation of Splunk Enterprise:

1. Expand the install package in an appropriate directory. For example:

   tar xvfz splunk_package_name.tgz -C /opt
   

2. Run enable boot-start to install the Splunkd.service unit file:

   sudo $SPLUNK_HOME/bin/splunk enable boot-start -systemd-managed 1 -user <username>
   

   When running enable boot-start for the first time after a clean install, Splunk Enterprise prompts you to accept the Splunk software license agreement. To automatically accept the license without prompt, specify the --accept-license flag with the command.

3. Start splunkd.

   sudo $SPLUNK_HOME/bin/splunk start 
   

4. Verify that splunkd is running as a systemd service.

   $SPLUNK_HOME/bin/splunk status
   

Manage clusters under systemd

When managing an indexer cluster under systemd:

• You must use the sudo command to start, stop, and restart the cluster manager node or individual peer nodes using splunk start|stop|restart commands.
• You do not need sudo to perform a rolling restart using the splunk rolling-restart cluster-peers command, or to take a peer offline using the splunk offline command.

When managing a search head cluster under systemd:

• You must use the sudo command to start, stop, and restart cluster members using splunk start|stop|restart commands.
• You do not need sudo to perform a rolling restart using the splunk rolling-restart shcluster-members command, or to remove a cluster member using the splunk remove shcluster-members command.

Upgrade considerations for systemd

Upgrade from 8.0.x to 8.1

If you configured Splunk Enterprise version 8.0.x to run as a systemd service, upon upgrade to version 8.1, Splunk Enterprise adds the following properties to the Splunkd.service unit file:

User
Group
ExecStartPost

When Splunk Enterprise adds these unit file properties, it creates a new unit file that replaces the existing unit file Splunkd.service. It also renames the old unit file Splunkd.service_<timestamp>, which it saves for backup purposes only.

When upgrading directly from 7.3.x or lower to 8.1, Splunk Enterprise adds the Group property to the unit file.

Upgrade from 7.3.x or lower to 8.0

If you configured Splunk Enterprise version 7.3.x or lower to run as a systemd service, upon upgrade to version 8.0.x, on initial start, Splunk Enterprise modifies the existing systemd configuration as follows:

• It removes the ExecStartPost and User properties from the Splunkd.service unit file.
• It checks the systemd environment, identifies the cgroup path, and automatically sets permissions for the correct cgroup directories.

You must use sudo splunk start to perform the initial start of Splunk Enterprise after installing the version 8.0.0 upgrade tarball.

Using systemctl start to perform the initial start of Splunk Enterprise on upgrade to version 8.0.0 will fail.

For detailed information on upgrading Splunk Enterprise, see How to upgrade Splunk Enterprise in the Installation Manual.