Bug#757973: smartmontools: race condition when restarting service when using systemd

[Marc Dequènes (Duck)]

Package: smartmontools
Version: 6.2+svn3841-1.2
Severity: normal


Coin,

Restarting smartd stopped but did not start the service again:
# service smartmontools restart
# service smartmontools status
smartmontools.service - LSB: SMART monitoring daemon
   Loaded: loaded (/etc/init.d/smartmontools)
   Active: active (exited) since Tue 2014-08-12 21:43:47 CEST; 3s ago


Aug 12 21:43:47 Elwing smartd[4489]: smartd received signal 15: Terminated

Aug 12 21:43:47 Elwing systemd[1]: Started LSB: SMART monitoring daemon.

Aug 12 21:43:47 Elwing smartd[4489]: smartd is exiting (exit status 0)

No smartd was running and the PID file was not present.

I tried: service smartmontools start

But nothing else happened.


  /etc/init.d/smartmontools start
But it did not work either.

Using:
  /etc/init.d/smartmontools stop
  /etc/init.d/smartmontools start






Regards


-- Package-specific info:
Output of /usr/share/bug/smartmontools:

-- System Information:
Debian Release: jessie/sid
  APT prefers testing
  APT policy: (500, 'testing')
Architecture: amd64 (x86_64)

Kernel: Linux 3.14-2-amd64 (SMP w/2 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages smartmontools depends on:
ii  debianutils  4.4
ii  libc6        2.19-7
ii  libcap-ng0   0.7.3-1.1
ii  libgcc1      1:4.9.1-4
ii  libselinux1  2.3-1
ii  libstdc++6   4.9.1-4
ii  lsb-base     4.1+Debian13

Versions of packages smartmontools recommends:
ii  bsd-mailx [mailx]  8.1.2-0.20131005cvs-1

--
Marc Dequènes (Duck)

Process: 20780 ExecStop=/etc/init.d/smartmontools stop (code=exited, status=0/SUCCESS) Process: 20784 ExecStart=/etc/init.d/smartmontools start (code=exited, status=0/SUCCESS) Aug 12 21:43:47 Elwing smartmontools[20784]: Starting S.M.A.R.T. daemon: smartd already running. Aug 12 21:43:47 Elwing smartd[4489]: Device: /dev/sda [SAT], state written to /var/lib/smartmontools/smartd.Hitachi_HDS722020ALA330-JK1175YAJ0NJ5X.ata.state Aug 12 21:43:47 Elwing smartd[4489]: Device: /dev/sdb [SAT], state written to /var/lib/smartmontools/smartd.Hitachi_HDS722020ALA330-JK1175YAJ05W2X.ata.state Aug 12 21:43:47 Elwing smartd[4489]: Device: /dev/sdc [SAT], state written to /var/lib/smartmontools/smartd.ST2000VX000_1CU164-W1E7SQ1S.ata.state Aug 12 21:43:47 Elwing smartd[4489]: Device: /dev/sdd [SAT], state written to /var/lib/smartmontools/smartd.ST2000VX000_1CU164-W1E7Q584.ata.state Which resulted in such messages in the logs: systemd[1]: Started LSB: SMART monitoring daemon. I really needed to start this service as i fear a failure, so i tried to start the service using the old method: started the daemon properly, and now using the 'service' command works perfectly. This machine recently migrated to systemd but had rebooted since then, so the service was already started through systemd. I think the race condition when restarting should be fixed and i'm just unable to reproduce it because the race does not occur anymore. Nevertheless i don't understand why the start action did not work afterwards.

pgptAe7_bvaNW.pgp
Description: PGP Digital Signature