title: Monitor HDD health statistics using S.M.A.R.T
agents: linux
author: Lars Michelsen <lm@mathias-kettner.de>
license: GPL
distribution: check_mk
description:
 This check monitors health statistics like error counters of HDDs using the
 information reported by S.M.A.R.T. This check works only for HDDs which
 report the value {Temperature_Celsius} via {smartctl -A <device>}.

 The check uses information provided by the Check_MK linux agent which are
 available when the agent has been extended with the {smart} agent plugin.

perfdata:
 One variable for each of the known state counters.

inventory:
 If there is at least one supported value available for a HDD one service is
 inventorized.

examples:
 # These are the default thresholds of the check.
 smart_stats_default_levels = {
   'realloc_events':  (1,  1),
   'realloc_sectors': (1,  1),
   'spin_retries':    (1,  1),
   'pending_retries': (1,  1),
   'cmd_timeouts':    (5, 10),
   'end_to_end':      (1,  1),
   'uncorr_errs':     (1,  1),
   'udma_crcs':       (1,  1),
 }


[parameters]
params(dict from string to (int, int)): A python dictionary which contains
 tuples of two integers which represent the WARNING and the CRITICAL
 threshold threshold for each supported value.

[configuration]
smart_stats_default_levels (dict from string to (int, int)): default levels
 used for all inventorized checks.