Issue
I have created an RPM package for my daemon which on installation create service init script under /etc/init.d/ Now I wanted to port this RPM package for CentOS7.1 and use systemd framework for service startup through boot as well as by admin through start/stop command.
I couldn't find any tutorial. Please help.
EDIT: I tried the suggestion given by msuchy.
I tried and here is my observation
On sysV based framework (CentOS6.5)
[root@adil work]# /etc/init.d/daemon_script status
Service daemon: Stopped
[root@adil work]# /etc/init.d/daemon_script start
Starting daemon: Initializing daemon... [ OK ]
[root@adil work]#
[root@adil work]# /etc/init.d/daemon_script status
Service daemon: Running
[root@adil work]#
[root@adil work]# /etc/init.d/daemon_script stop
Shutting down parent daemon: [ OK ]
[root@adil work]# /etc/init.d/daemon_script status
Service daemon: Stopped
[root@adil work]#
===========
On systemd based framework, I install the same RPM on CentOS7.1
[root@localhost x86_64]# /etc/init.d/daemon_script
Usage: /etc/init.d/daemon_script {start|stop|restart|status}
[root@localhost x86_64]# /etc/init.d/daemon_script start
Starting daemon_script (via systemctl): Warning: Unit file of daemon_script.service changed on disk, 'systemctl daemon-reload' recommended.
Job for daemon_script.service failed. See 'systemctl status daemon_script.service' and 'journalctl -xn' for details.
[FAILED]
[root@localhost x86_64]# systemctl daemon-reload
[root@localhost x86_64]# systemctl status daemon_script.service
daemon_script.service - SYSV: start and stop Test daemon service.
Loaded: loaded (/etc/rc.d/init.d/daemon_script)
Active: failed (Result: exit-code) since Fri 2015-09-11 15:30:44 IST; 32s ago
Sep 11 15:30:44 localhost.localdomain systemd[1]: Starting SYSV: start and st...
Sep 11 15:30:44 localhost.localdomain systemd[1]: daemon_script.service: cont...
Sep 11 15:30:44 localhost.localdomain systemd[1]: Failed to start SYSV: start...
Sep 11 15:30:44 localhost.localdomain systemd[1]: Unit daemon_script.service ...
Hint: Some lines were ellipsized, use -l to show in full.
[root@localhost x86_64]# systemctl status daemon_script.service -l
daemon_script.service - SYSV: start and stop Test daemon service.
Loaded: loaded (/etc/rc.d/init.d/daemon_script)
Active: failed (Result: exit-code) since Fri 2015-09-11 15:30:44 IST; 46s ago
Sep 11 15:30:44 localhost.localdomain systemd[1]: Starting SYSV: start and stop Test daemon service....
Sep 11 15:30:44 localhost.localdomain systemd[1]: daemon_script.service: control process exited, code=exited status=203
Sep 11 15:30:44 localhost.localdomain systemd[1]: Failed to start SYSV: start and stop Test daemon service..
Sep 11 15:30:44 localhost.localdomain systemd[1]: Unit daemon_script.service entered failed state.
[root@localhost x86_64]#
output of journalctl -xn
-- Logs begin at Fri 2015-09-11 14:50:35 IST, end at Fri 2015-09-11 15:40:01 IST. --
Sep 11 15:31:03 localhost.localdomain systemd[1]: [/usr/lib/systemd/system/dm-event.socket:10] Unknown lvalue 'RemoveOnStop' in section 'Socket'
Sep 11 15:39:33 localhost.localdomain systemd[1]: Starting SYSV: start and stop Test daemon service....
-- Subject: Unit daemon_script.service has begun with start-up
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
--
-- Unit daemon_script.service has begun starting up.
Sep 11 15:39:33 localhost.localdomain systemd[8509]: Failed at step EXEC spawning /etc/rc.d/init.d/daemon_script: Exec format error
-- Subject: Process /etc/rc.d/init.d/daemon_script could not be executed
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
--
-- The process /etc/rc.d/init.d/daemon_script could not be executed and failed.
--
-- The error number returned while executing this process is 8.
Sep 11 15:39:33 localhost.localdomain systemd[1]: daemon_script.service: control process exited, code=exited status=203
Sep 11 15:39:33 localhost.localdomain systemd[1]: Failed to start SYSV: start and stop Test daemon service..
-- Subject: Unit daemon_script.service has failed
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
--
-- Unit daemon_script.service has failed.
--
-- The result is failed.
Sep 11 15:39:33 localhost.localdomain systemd[1]: Unit daemon_script.service entered failed state.
Sep 11 15:40:01 localhost.localdomain systemd[1]: Created slice user-0.slice.
-- Subject: Unit user-0.slice has finished start-up
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
--
-- Unit user-0.slice has finished starting up.
--
-- The start-up result is done.
Sep 11 15:40:01 localhost.localdomain systemd[1]: Starting Session 7 of user root.
-- Subject: Unit session-7.scope has begun with start-up
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
--
-- Unit session-7.scope has begun starting up.
Sep 11 15:40:01 localhost.localdomain systemd[1]: Started Session 7 of user root.
-- Subject: Unit session-7.scope has finished start-up
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
--
-- Unit session-7.scope has finished starting up.
--
-- The start-up result is done.
Sep 11 15:40:01 localhost.localdomain CROND[8528]: (root) CMD (/usr/lib64/sa/sa1 1 1)
[root@localhost x86_64]#
Solution
You do not need to migrate your scripts.
It is probably better and you can utilize some interesting functions of Systemd, your files will be smaller. But you do not need to migrate. Systemd works correctly with SysV init files too.
You can not locate any SysV init files on CentOS 7 installation because Red Hat packagers (who created CentOS) put an effort into packaging and migrated all SysV files to unit files. But you do not need to.
The is only one task you need to do. Once you place new SysV file, you must reload the systemd manager configuration using
# systemctl daemon-reload
That is all. I give you small example
# cat /etc/init.d/foo
#!/usr/bin/sh
echo ahoy
# chmod a+x /etc/init.d/foo
# systemctl start foo
Failed to start foo.service: Unit foo.service failed to load: No such file or directory.
# systemctl daemon-reload
# systemctl start foo
# journalctl -xn
-- Subject: Unit foo.service has finished start-up
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
--
-- Unit foo.service has finished starting up.
--
-- The start-up result is done.
# service foo start
ahoy
So you can use all command (chkconfig, service) as you are used to from CentOS 6. And when you have time, you can study man systemd.unit(5) and bunch of other man pages (see "SEE ALSO" of that man page).
Answered By - msuchy Answer Checked By - David Goodson (WPSolving Volunteer)