netdata/health/health.d/disks.conf

# you can disable an alarm notification by setting the 'to' line to: silent


# -----------------------------------------------------------------------------
# low disk space

# checking the latest collected values
# raise an alarm if the disk is low on
# available disk space

 template: disk_space_usage
       on: disk.space
    class: Utilization
     type: System
component: Disk
       os: linux freebsd
    hosts: *
chart labels: mount_point=!/dev !/dev/* !/run !/run/* *
     calc: $used * 100 / ($avail + $used)
    units: %
    every: 1m
     warn: $this > (($status >= $WARNING ) ? (80) : (90))
     crit: ($this > (($status == $CRITICAL) ? (90) : (98))) && $avail < 5
    delay: up 1m down 15m multiplier 1.5 max 1h
     info: disk ${label:mount_point} space utilization
       to: sysadmin

 template: disk_inode_usage
       on: disk.inodes
    class: Utilization
     type: System
component: Disk
       os: linux freebsd
    hosts: *
chart labels: mount_point=!/dev !/dev/* !/run !/run/* *
     calc: $used * 100 / ($avail + $used)
    units: %
    every: 1m
     warn: $this > (($status >= $WARNING)  ? (80) : (90))
     crit: $this > (($status == $CRITICAL) ? (90) : (98))
    delay: up 1m down 15m multiplier 1.5 max 1h
     info: disk ${label:mount_point} inode utilization
       to: sysadmin


# -----------------------------------------------------------------------------
# disk fill rate

# calculate the rate the disk fills
# use as base, the available space change
# during the last hour

# this is just a calculation - it has no alarm
# we will use it in the next template to find
# the hours remaining

template: disk_fill_rate
      on: disk.space
      os: linux freebsd
   hosts: *
  lookup: min -10m at -50m unaligned of avail
    calc: ($this - $avail) / (($now - $after) / 3600)
   every: 1m
   units: GB/hour
    info: average rate the disk fills up (positive), or frees up (negative) space, for the last hour

# calculate the hours remaining
# if the disk continues to fill
# in this rate

template: out_of_disk_space_time
      on: disk.space
      os: linux freebsd
   hosts: *
    calc: ($disk_fill_rate > 0) ? ($avail / $disk_fill_rate) : (inf)
   units: hours
   every: 10s
    warn: $this > 0 and $this < (($status >= $WARNING)  ? (48) : (8))
    crit: $this > 0 and $this < (($status == $CRITICAL) ? (24) : (2))
   delay: down 15m multiplier 1.2 max 1h
    info: estimated time the disk will run out of space, if the system continues to add data with the rate of the last hour
      to: silent


# -----------------------------------------------------------------------------
# disk inode fill rate

# calculate the rate the disk inodes are allocated
# use as base, the available inodes change
# during the last hour

# this is just a calculation - it has no alarm
# we will use it in the next template to find
# the hours remaining

template: disk_inode_rate
      on: disk.inodes
      os: linux freebsd
   hosts: *
  lookup: min -10m at -50m unaligned of avail
    calc: ($this - $avail) / (($now - $after) / 3600)
   every: 1m
   units: inodes/hour
    info: average rate at which disk inodes are allocated (positive), or freed (negative), for the last hour

# calculate the hours remaining
# if the disk inodes are allocated
# in this rate

template: out_of_disk_inodes_time
      on: disk.inodes
      os: linux freebsd
   hosts: *
    calc: ($disk_inode_rate > 0) ? ($avail / $disk_inode_rate) : (inf)
   units: hours
   every: 10s
    warn: $this > 0 and $this < (($status >= $WARNING)  ? (48) : (8))
    crit: $this > 0 and $this < (($status == $CRITICAL) ? (24) : (2))
   delay: down 15m multiplier 1.2 max 1h
    info: estimated time the disk will run out of inodes, if the system continues to allocate inodes with the rate of the last hour
      to: silent


# -----------------------------------------------------------------------------
# disk congestion

# raise an alarm if the disk is congested
# by calculating the average disk utilization
# for the last 10 minutes

 template: 10min_disk_utilization
       on: disk.util
    class: Utilization
     type: System
component: Disk
       os: linux freebsd
    hosts: *
   lookup: average -10m unaligned
    units: %
    every: 1m
     warn: $this > 98 * (($status >= $WARNING)  ? (0.7) : (1))
    delay: down 15m multiplier 1.2 max 1h
     info: average percentage of time ${label:device} disk was busy over the last 10 minutes
       to: silent


# raise an alarm if the disk backlog
# is above 1000ms (1s) per second
# for 10 minutes
# (i.e. the disk cannot catch up)

 template: 10min_disk_backlog
       on: disk.backlog
    class: Latency
     type: System
component: Disk
       os: linux
    hosts: *
   lookup: average -10m unaligned
    units: ms
    every: 1m
     warn: $this > 5000 * (($status >= $WARNING)  ? (0.7) : (1))
    delay: down 15m multiplier 1.2 max 1h
     info: average backlog size of the ${label:device} disk over the last 10 minutes
       to: silent