3

Please, I received this report by S.M.A.R.T

Checking /dev/sdb....
Errors:
 Please note the following marginal Attributes:
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
190 Airflow_Temperature_Cel 0x0022   073   037   045    Old_age   Always   In_the_past 27 (3 44 30 26 0)

Checking /dev/sdc....
Errors:
 Please note the following marginal Attributes:
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
190 Airflow_Temperature_Cel 0x0022   073   038   045    Old_age   Always   In_the_past 27 (0 181 30 26 0)

Checking /dev/sdd....
Errors:
 Please note the following marginal Attributes:
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
190 Airflow_Temperature_Cel 0x0022   071   040   045    Old_age   Always   In_the_past 29 (1 6 32 27 0)

Please, Is this a serious mistake ? Possible cause problems in the future ?

MayeulC
  • 130
  • 9
Mohammed OK
  • 51
  • 1
  • 1
  • 5

1 Answers1

14

At some time in the past, your drives recorded air temperatures of 63, 62, and 60 degrees, respectively. The drives currently report temperatures of 27, 27 and 29 degrees. The attribute alarms at 55 degrees. (These values are calculated by subtracting the raw values in the VALUE, WORST and THRESH columns from 100.)

Unless you have some other temperature monitors in the server, or some other knowledge of the server's environment, it's not possible to say when this occurred or for how long.

There is probably no permanent damage to the drives, if the thermal events were of a short duration. But the drives are probably more likely to fail in future, especially if they become too hot again.

You can set up smartd to monitor the SMART attributes of your drives and send you email if there are any significant events. Or you can use another monitoring system such as nagios, zabbix, etc. If your server is an enterprise-grade server such as HP ProLiant or Dell PowerEdge, it will also include monitoring tools which can notify you of problems by email or SNMP trap.

Community
  • 1
Michael Hampton
  • 237,123
  • 42
  • 477
  • 940