Hard Drive Burn-In Testing - Discussion Thread

[ggoldfingerd]

badblocks -ns /dev/adaX

It might be nice for you to add that a statement mentioning that badblocks is limited to 32-bit addresses. Badblocks will return "Value too large for defined data type invalid end block ..." on 6TB HDD's. The solution is to increase the blocksize to 4096, which is native to the drive I have. I see that @radian23 mentioned this too.

Code:

badblocks -b 4096 -ns /dev/adaX

 

[qwertymodo]

but that wipes out all the data on the SSD, it's not selective.

That's why I specified secure *file* deletion, rather than secure full-drive deletion. There is no way to tell a SSD to securely delete an individual file. Even then, I've seen varying reports as to the actual security of even the full-drive deletion.

 

[joeschmuck]

Even then, I've seen varying reports as to the actual security of even the full-drive deletion

Shreading the drive is the only secure deletion. That is what the NSA does when we ship them our drives to destroy.

 

[qwertymodo]

Shreading the drive is the only secure deletion. That is what the NSA does when we ship them our drives to destroy.

With the NSA's track record lately, I have to wonder if they really do that or if they just hold on to them "for safe keeping". But tinfoil hats aside, you're right... that should do it: https://www.youtube.com/watch?v=sQYPCPB1g3o

 

[cyberjock]

I think that was meant as a joke quertymodo. ;)

 

[joeschmuck]

Nope, this is what I'm told the NSA does with the hard drives we send them for destruction, it is the only acceptable method for the level of classified data we process. Actually I'm under the impression that the overall shred operation would leave every chip completely destroyed so the material in the video would need to go through another finer shredder. But for all I know, they melt it down next.

 

[qwertymodo]

I absolutely agree that shredding is the only truly secure method for classified information. My earlier discussion of secure deletion on SSD's was in reference to "secure" in terms of non-classified data on consumer devices, and was more concerned with discussing the behavior of the wear-leveling controller, rather than going in-depth wrt secure deletion. And my jab at the NSA was half joking (though not entirely... seriously, you might want to consider finding somebody else to do the shredding for you or invest in your own shredder).

 

[Arwen]

During one of my fill-in jobs as a Technical Support Engineer at Sun Microsystems, (and later Oracle), we
occasionally got bad disk replacement tickets from US government secure sites. As long as their support contract
had a "do not need hard disk drive return" clause, I'd either ship them a replacement. Or send a pre-screened field
engineer with the replacement.

I remember my first case that occured. Having working with highly secure sites before, (as a software and
hardware developer), it was easy to understand the need. It was just paper work to make sure they paid for the
priviledge to keep the bad, old drive.

That old job of mine meant that any media, (E)EPROM, floppy, CD/DVD, hard disk drive brought into the site,
never left in usable condition.

 

[Ruff.Hi]

Now, before we can perform raw disk I/O, we need to enable the kernel geometry debug flags. This carries some inherent risk, and should probably not be done on a production system. This does not survive through a reboot, so when you're done, just reboot the machine to disable it:

Code:

sysctl kern.geom.debugflags=0x10

Now, we can run badblocks. THIS TEST WILL DESTROY ANY DATA ON THE DISK SO ONLY RUN THIS ON A NEW DISK WITHOUT DATA ON IT OR BACK UP ANY DATA FIRST:​

Code:

badblocks -ws /dev/adaX

I'm running the badblocks tests on my NAS at home at the moment ... 5 HDDs, 4TB each.

FreeNAS is going nuts with issues such as ...

Device: /dev/ada4, Read SMART Self-Test Log Failed
Device: /dev/ada1, Read SMART Self-Test Log Failed
Device: /dev/ada2, Read SMART Self-Test Log Failed
Device: /dev/ada2, Read SMART Error Log Failed
Device: /dev/ada1, Read SMART Error Log Failed
Device: /dev/ada3, failed to read SMART Attribute Data
Device: /dev/ada2, failed to read SMART Attribute Data
Device: /dev/ada1, failed to read SMART Attribute Data
Device: /dev/ada3, Read SMART Error Log Failed

I'm putting this down to the badblock tests going on at the moment. Am I ok with ignoring these reports or is this something to really worry about?

Edit: And now I get 'The volume DuffleBag (ZFS) state is DEGRADED: One or more devices has been removed by the administrator. Sufficient replicas exist for the pool to continue functioning in a degraded state.' which does sound bad. I'll look at it when I get home.

Seems like my pool is crashing. Lucky the data on it is just a copy of my production data at the moment ... nothing to really lose.

 

[wisebear]

Hello guys.
I have bought 3 new WD Red 3 TB drives for my NAS. I have already returned one drive because it was not recognized in Bios. I have followed this guide and made tests with the other two drives. I have now some doubts whether I should replace another disk under warranty. With this disk I got errors during badblocks (11/0/0) test and some alerts in FreeNAS (3 Currently unreadable (pending) sectors and some other smart errors). Read and write speeds were also noticeably lower on this disk. Here are my smart results obtained after badblocks test:

=== START OF READ SMART DATA SECTION ===
SMART Attributes Data Structure revision number: 16
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE
1 Raw_Read_Error_Rate 0x002f 200 200 051 Pre-fail Always - 76
3 Spin_Up_Time 0x0027 181 179 021 Pre-fail Always - 5916
4 Start_Stop_Count 0x0032 100 100 000 Old_age Always - 33
5 Reallocated_Sector_Ct 0x0033 200 200 140 Pre-fail Always - 0
7 Seek_Error_Rate 0x002e 200 200 000 Old_age Always - 0
9 Power_On_Hours 0x0032 100 100 000 Old_age Always - 102
10 Spin_Retry_Count 0x0032 100 253 000 Old_age Always - 0
11 Calibration_Retry_Count 0x0032 100 253 000 Old_age Always - 0
12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 33
192 Power-Off_Retract_Count 0x0032 200 200 000 Old_age Always - 19
193 Load_Cycle_Count 0x0032 200 200 000 Old_age Always - 49
194 Temperature_Celsius 0x0022 120 114 000 Old_age Always - 30
196 Reallocated_Event_Count 0x0032 200 200 000 Old_age Always - 0
197 Current_Pending_Sector 0x0032 200 200 000 Old_age Always - 0
198 Offline_Uncorrectable 0x0030 100 253 000 Old_age Offline - 0
199 UDMA_CRC_Error_Count 0x0032 200 200 000 Old_age Always - 0
200 Multi_Zone_Error_Rate 0x0008 200 200 000 Old_age Offline - 1

I am a complete newbie regarding s.m.a.r.t test results. Should I be worried with raw value 1 of Multi_Zone_Error_Rate and with 76 Raw_Read_Error_Rate?

Sorry for my English and formatting.

 

[qwertymodo]

Unreadable/pending sectors on a brand new drive is an absolute red flag, go ahead and RMA it.

 

[Robert Trevellyan]

Unreadable/pending sectors on a brand new drive is an absolute red flag, go ahead and RMA it.

I don't see any reallocated, unreadable or pending reallocations in these SMART data.

@wisebear are you 100% sure this is the drive that was reported as having errors?

 

[qwertymodo]

I don't see any reallocated, unreadable or pending reallocations in these SMART data.

You're right. I thought there were. SMART tests are notoriously impossible to read on this board thanks to the code tag being broken.

 

[Bidule0hm]

Actually it's not broken to this point but just a bit. It's just that the OP doesn't used the code tag.

 

[qwertymodo]

Huh, they must've fixed it. Last time I tried the code tag it didn't use a monospace font and it seriously screwed up whitespace.

 

[Ericloewe]

Ideally, you'd want to RMA with a multi-zone error, but I doubt they'll play along nicely.

 

[wisebear]

@Robert Trevellyan: I am 100 % sure this drive was reported to have the issues.

I don't understand what do you mean with code tags...this means that s.m.a.r.t does not detects issues on my drive correctly or what? Errors were definitely present during badblocks test. Is there any other way to ''make'' my smart test result bad? I want to RMA it because the disk is not in perfect condition...also read and write speeds are not comparable to other wd red 3 tb drives I have.

 

[depasseg]

Code tags relate to the reply in the thread. It makes it easier for us to read the results you paste.

 

[Robert Trevellyan]

alerts in FreeNAS (3 Currently unreadable (pending) sectors and some other smart errors)

this means that s.m.a.r.t does not detects issues on my drive correctly or what?

It is possible for attribute #197 Current_Pending_Sector to return to 0 if the sectors in question were successfully written later. According to Wikipedia, this is a shortcoming of the SMART technology, since it may indicated 'unstable' sectors that are hard to read but appear to write successfully.

You might have to put the drive into service and wait for it to have a persistent SMART failure. Either that or keep running badblocks and SMART tests on it. It should only take one failed SMART test to qualify for RMA.

 

[wisebear]

What about the log below? Is it usual that failed self-test is followed by a successful one?

Code:

SMART Self-test log structure revision number 1
Num  Test_Description    Status                  Remaining  LifeTime(hours)  LBA_of_first_error
# 1  Extended offline    Completed without error       00%        99         -
# 2  Extended offline    Completed: read failure       70%        11         1817685136
# 3  Conveyance offline  Completed without error       00%         9         -
# 4  Short offline       Completed without error       00%         9         -
1 of 1 failed self-tests are outdated by newer successful extended offline self-test # 1