Another SSD with possible problems

sorcerer

Active Member
Reaction score
76
Location
Preston, Lancs, UK
Looking for an SSD to put in my Zalman and I've recently had two Crucial MX100s that have been faulty and had to be returned. I bought them locally because I only live about 1.5 miles from CPC but after two faulty ones I decided on a refund in case they have a faulty batch.

Couple of days ago I put an order in with Ebuyer and noticed that they had a Kingston SSD (SV300) on offer and that it scored 9.3 over more than 350 reviews at Revoo, so I thought it's got to be worth a punt. It arrived this morning but, having been bitten twice, I thought I'd run a GSmartControl test first before chucking it in the Zalman. It passed the extended test without errors but has a couple of attributes flagged. Question is, I've no experience of SSDs so would you return it because of these or would it be ok?

ssd.jpg
 
Kingston SSD (SV300) on offer and that it scored 9.3 over more than 350 reviews at Revoo

End user scores are worth diddly squat :(

I'm running an Intel SSD in mine. No problems so far.

I believe E-Buyer has a good return policy. Tell them you're not happy with it, and would like to exchange it for an Intel:
240GB: http://www.ebuyer.com/523678-intel-530-series-240gb-2-5inch-sata-6gbs-ssd-ssdsc2bw240a4011
120GB: http://www.ebuyer.com/567933-intel-530-series-120gb-2-5inch-7mm-oem-ssd-ssdsc2bw120a401


Andy
 
I have one Intel and one Samsung 250 GB in my personal rig that runs 24/7 and about 10 other In client machines without issue. I only trust those 2 brands so far and I make sure that all of the machines have image backups as I only trust them to a point.
 
Think I'll take Andy's advice and swap it for an Intel. Gotta say though, I wouldn't put an SSD in a computer of mine given my (admittedly very limited) experience of them so far. I only thought of putting one in the Zalman because of power issues reported by some people when using a mechanical drive.
 
It arrived this morning but, having been bitten twice, I thought I'd run a GSmartControl test first before chucking it in the Zalman.
Is this the first power-on, before any use of the drive? There's no way to tell the state of the actual memory cells on a new drive, so I think that that smartctl report is meaningless.

Try running a long self-test (which will at least exercise every location), power-off, power-on and then see what smartctl reports.

Edit to add: we ran hard drives for years before SMART was invented and never worried ...

Edit further add: I select my SSDs on price, primarily, with a quick Google to weed out any obvious dogs. I currently have 2x OCZ Solid3 (60 GB), SanDisk Ultra Plus (256 GB) and Crucial MX100 (256 GB) in service. Treat them as you would (should) any hard drive and have a backup. Golden.
 
Last edited:
Is this the first power-on, before any use of the drive?

Yes it is (was). I created one big NTFS partition and then ran smartctl.

Try running a long self-test (which will at least exercise every location), power-off, power-on and then see what smartctl reports.

Did that and it still reports the same. Something is wrong somewhere I think because it took 35 minutes to run the extended test - surely it should be faster than that on a 120GB SSD? The Crucial MX100 only took about 5 minutes if I remember correctly.
 
I have various brands of SSD's in every laptop we have, a couple of Macs, and several Windows machines, including the old eePC I carry to job sites everyday.
I also have put them in at least 20 Macbook Pros for clients over the last year.

I have used Samsung, Kingston, Crucial, and Ocz. Never received a bad one and so far no failures. A small sample size, but I don't hesitate to use them.
Compare that to the 1 in 5 failure rate I have been getting from new WD spinning drives.

Then again, every computer I put them in, I set up some type of automated backup, so I've got that going for me.
 
Something is wrong somewhere I think because it took 35 minutes to run the extended test - surely it should be faster than that on a 120GB SSD? The Crucial MX100 only took about 5 minutes if I remember correctly.
(gSmartControl, under the Perform Tests tab) what is the estimated duration quoted for Extended self-test?

Certainly update the firmware as @AndyM suggested (don't you do that as a matter of course with anything having firmware?) and use the Kingston tools. It's entirely possible that Kingston is using SMART counters for non-standard purposes (though I have no evidence for this).

OCZ are now a subsidiary of Toshiba. Some of their practises were a bit .... ummm..."iffy" prior to the acquisition. Some people here will testify to that.
That was just an indication that I don't buy by brand, but those are older (60 GB Solid3) drives, when OCZ was OCZ. I bought them just after the floods in Thailand (2011?) sent HDD prices through the roof. One was the system drive on my daily workstation until late last year (and is now in a Zalman case); the other is in my drag-to-site laptop. Never had a problem with either, personally.
 
(gSmartControl, under the Perform Tests tab) what is the estimated duration quoted for Extended self-test?

Estimated duration is 48 minutes - actual time taken was 35 minutes.

Certainly update the firmware as @AndyM suggested (don't you do that as a matter of course with anything having firmware?) and use the Kingston tools. It's entirely possible that Kingston is using SMART counters for non-standard purposes (though I have no evidence for this).

It's already on the latest firmware. The one Andy linked to above is the 525 but the drive arrived with the 600 revision already installed and Kingston Tools reports that there is currently no update available.
 
@sorcerer What does the Kingston tool report for those SMART attributes? gSmartControl is interpreting the names, certainly, and may be presenting numbers that have no meaning. gSC does the same for my Sandisk SSD. (And your screenshot doesn't show the SMART attribute number column on the left.) Use the commandline smartctl -a /dev/sd... for raw data.

See also this link to Kingston's SMART Attributes documentation. Those big numbers are actually concatenated smaller numbers, in my interpretation.
 
I have a Kingston SSDNow here for recovery as we speak. Under the hood, the controller is a Toshiba proprietary controller that is on par with the uncrackable Sandforce controllers.

I'd vote for Intel, as they are the one device that we have consistently been able to recover data from. That said, my opinion will likely evolve over the years.
 
@sorcerer What does the Kingston tool report for those SMART attributes? gSmartControl is interpreting the names, certainly, and may be presenting numbers that have no meaning. gSC does the same for my Sandisk SSD. (And your screenshot doesn't show the SMART attribute number column on the left.) Use the commandline smartctl -a /dev/sd... for raw data.

See also this link to Kingston's SMART Attributes documentation. Those big numbers are actually concatenated smaller numbers, in my interpretation.

Sorry for the delayed reply - been very hectic here. This is the Smart data as interpreted by Kingston Tools:

SMART READ DATA
Revision: 10
Attributes List
1: (SSD Raw Read Error Rate) Normalized Rate: 120 Sectors Read: 0 Read Errors: 0
5: (SSD Retired Block Count) Spare blocks remaining 100% Retired Block 0
9: (SSD Power-On Hours) Value 100 Total 10 hrs 1 mins
12: (SSD Power Cycle Count) Power Cycle Life Remaining 100% Number of power cycles 10
171: (SSD Program Fail Count) Program Error Count 0
172: (SSD Erase Fail Count) Erase Error Count 0
174: (SSD Unexpected power loss count) Unexpected power loss Count 4
177: (Wear Range Delta) Wear Range Delta 0%
181: (Program Fail Count) Program Error Count 0
182: (Erase Fail Count) Erase Error Count 0
187: (SSD Reported Uncorrectable Errors) Normalized Value 100 lifetime URAISE Errors 0
189: (Unrecognized Attribute) Value: 17 Raw Data: 11 00 1c 00 0f 00 00
194: (SSD Temperature Monitoring) Normalized temp 17 Current 17 High 28 Low 15
195: (SSD ECC On-the-fly Count) Normalized Value 120 Sectors Read 0 UECC Count 0
196: (SSD Reallocation Event Count) Normalized Value 100 Reallocation Event Count 0
201: (SSD Uncorrectable Soft Read Error Rate)Normalized Value 120 Sectors Read 0 Uncorrectable Soft Error Count 0
204: (SSD Soft ECC Correction Rate (RAISE) Normalized Value 120 Sectors Read 0 Soft ECC Correction Count 0
230: (SSD Life Curve Status) Normalized Value 100
231: (SSD Life Left) Life Remaining 100%
233: (SSD Internal Reserved) 0
234: (SSD Internal Reserved) 0
241: (SSD Lifetime writes from host) lifetime writes 0
242: (SSD Lifetime reads from host) lifetime reads 0
244: (Unrecognized Attribute) Value: 100 Raw Data: 15 00 1f 00 00 00 00

Hope that helps, and thanks for the continuing help.
 
This is the Smart data as interpreted by Kingston Tools:
I am neither a doctor nor a lawyer ... :eek: but I'd declare that drive fit and serviceable. Maybe @lcoughey could offer a more practised view?

As for the 'Unrecognized Attribute' entries, I think they're just data that Kingston may like to be able to see, but are not for public consumption. I don't think they're significant and I would be happy to use that drive. Put it to work for a few weeks/months and check again with the Kingston tools, comparing with today's results.
 
Back
Top