netdata/health/health.d/memory.conf

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

    alarm: 1hour_memory_hw_corrupted
       on: mem.hwcorrupt
    class: Errors
     type: System
component: Memory
       os: linux
    hosts: *
     calc: $HardwareCorrupted
    units: MB
    every: 10s
     warn: $this > 0
    delay: down 1h multiplier 1.5 max 1h
     info: amount of memory corrupted due to a hardware failure
       to: sysadmin

## ECC Controller

 template: ecc_memory_mc_correctable
       on: mem.edac_mc
    class: Errors
     type: System
component: Memory
       os: linux
    hosts: *
   lookup: sum -10m unaligned of correctable, correctable_noinfo
    units: errors
    every: 1m
     warn: $this > 0
    delay: down 1h multiplier 1.5 max 1h
     info: memory controller ${label:controller} ECC correctable errors in the last 10 minutes
       to: sysadmin

 template: ecc_memory_mc_uncorrectable
       on: mem.edac_mc
    class: Errors
     type: System
component: Memory
       os: linux
    hosts: *
   lookup: sum -10m unaligned of uncorrectable,uncorrectable_noinfo
    units: errors
    every: 1m
     crit: $this > 0
    delay: down 1h multiplier 1.5 max 1h
     info: memory controller ${label:controller} ECC uncorrectable errors in the last 10 minutes
       to: sysadmin

## ECC DIMM

 template: ecc_memory_dimm_correctable
       on: mem.edac_mc_dimm
    class: Errors
     type: System
component: Memory
       os: linux
    hosts: *
   lookup: sum -10m unaligned of correctable
    units: errors
    every: 1m
     warn: $this > 0
    delay: down 1h multiplier 1.5 max 1h
     info: DIMM ${label:dimm} controller ${label:controller} (location ${label:dimm_location}) ECC correctable errors in the last 10 minutes
       to: sysadmin

 template: ecc_memory_dimm_uncorrectable
       on: mem.edac_mc_dimm
    class: Errors
     type: System
component: Memory
       os: linux
    hosts: *
   lookup: sum -10m unaligned of uncorrectable
    units: errors
    every: 1m
     crit: $this > 0
    delay: down 1h multiplier 1.5 max 1h
     info: DIMM ${label:dimm} controller ${label:controller} (location ${label:dimm_location}) ECC uncorrectable errors in the last 10 minutes
       to: sysadmin