Excessive SMART Error Rates Logged for Read and Verify ECC Errors on Certain Enterprise Hard Drives

-

-

Table of Contents

1. Introduction
2. Description
3. Solution
4. Further Information

 

---

Introduction

S.M.A.R.T. (Self-Monitoring, Analysis and Reporting Technology; often written as SMART) is a health monitoring system for hard-drives and solid-state drives which has been internationally standardised.

The primary function of SMART is to detect and report various indicators of drive reliability, with the intent of anticipating imminent hardware failures, allowing the hardware to be replaced before failure and the integrity of the data to be maintained.

Dell EMC have collaborated with our hard-drive vendors in the interpretation of these values.

 

---

Description

It has been found, whilst reviewing the SMART status of some enterprise hard-drives, high read and verify ECC (Error Correction Code) correction rates may be observed on some models. If these values are compared against other hard-drives, it may appear to indicate that some particular models of hard-drives have a much higher error rate than others which may report zero ECC corrections. In some cases this error rate will appear as hundreds of millions of ECC corrections and may increment rapidly as more I/O transactions occur.

 

An example of this situation is provided below. This example was gathered by running the command "smartctl -a /dev/sdX" under Linux OS.

Figure 1: Error Counter Log

 

---

Solution

The SMART specification allows vendors to provide these counters, such as the ones shown in the above example, for informational purposes. The counters are not necessarily a count of soft or hard faults within the ECC logic. This allows each drive vendor flexibility as to what will be displayed in the available SMART fields. For some vendors, no error data is presented in the ECC read or verify categories. In the example above, the vendor has chosen to use the counters for monitoring the ECC functionality. The values which are presented do not represent an error-rate. Similarily, a higher rate of events on some disks in comparison to others does not indicate that a performance problem exists.
 

If you have a specific query in relation to the health counters on a certain model of Enterprise HDD, please contact our support technicians who can work to obtain an answer from Dell EMC's engineering teams.

 

---

Further Information

• For more information on the international standardisation of SMART values and other SCSI Storage Interfaces please see the homepage of the T10 technical committee located here.
• Smartmontools is a set of utility programs to control and monitor computer storage systems using the Self-Monitoring, Analysis and Reporting Technology system built into most modern ATA, Serial ATA, SCSI/SAS and NVMe hard drives. It is not a DellEMC tool. More information on Smartmontools can be found here.

 

---