Custom Query (1110 matches)

Filters
 
Or
 
  
 
Columns

Show under each result:


Results (37 - 39 of 1110)

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Ticket Resolution Summary Owner Reporter
#249 fixed smartd: health check failed if SCSI drive is performing test somebody Alex Samorukov
Description

Instead of SMART HEALTH "OK" smartd shows Device: /dev/sdb, non-SMART asc,ascq: 4,9 in the logs if test is running. smartcl -H /dev/sdb shows that smart status if fine.

#1080 wontfix smartd.service: Use Type=forking to wait for successful initialisation Christian Franke
Description

From Debian Bug 865912:

Forking to the background (Type=forking) is the recommended way to signal to systemd that a daemon has initialized successfully and is ready for operation; at least for daemons without explicit systemd support (Type=notify).

In this particular case, "systemd restart smartd" will give immediate feedback of a failure if the user makes a mistake in /etc/smartd.conf.

Proposed patch:

  • smartd.service.in

     
    66[Service]
     7Type=forking
    78EnvironmentFile=-/usr/local/etc/sysconfig/smartmontools
    8 ExecStart=/usr/local/sbin/smartd -n $smartd_opts
     9ExecStart=/usr/local/sbin/smartd $smartd_opts
    910ExecReload=/bin/kill -HUP $MAINPID
#1081 fixed smartd.service: Add support for Type=notify Christian Franke Christian Franke
Description

Add support to notify systemd about success/failure of smartd startup via sd_notify(3).

For example:

  // On success:
  sd_notifyf(0, "READY=1\n"
    "STATUS=Monitoring %d ATA/SATA, %d SCSI/SAS and %d NVMe devices\n"
    "MAINPID=%lu", ... );

  // On failure:
  sd_notifyf(0, "STATUS=Error in configuration file");

See also ticket #1080.

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Note: See TracQuery for help on using queries.