Samsung SSD RMA process makes me never want to buy Samsung again...

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
M

Magical Penguin

Member
Jul 20, 2015
37
2
8
44
I have an 840 Samsung EVO failing.

To RMA the drive, I had to call in. Now I have to fill out a 14 question email (not even a form, but an email) to get an RMA approved. They need the bill of sale, and a photo of the drive label.

And they absolutely don't do advanced replacement.

Which means I pretty much have to buy a brand new drive tonight. I can't really think of any other options. And then I guess ebay the RMA drive once they send a new one.

I've done RMAs with Western Digital and Seagate that were so much easier. WD in particular stands out in my mind for being easy.

Which SSD companies have an easy RMA process?
 
Patriot

Patriot

Moderator
Apr 18, 2011
1,451
792
113
Define failing?

Running out of blocks to reallocate while under wpd?

I have never had an SSD fail on me to date...save maybe a used corsair force gt I got used and long out of warranty... and I think it is just a chipset incompatibility.
I have stacks of SSDs...heck even my OCZ vertex 1 is still kicking.

Hard drive companies are used to failure...
All the SSD manufacturers are memory companies...
 
M

Magical Penguin

Member
Jul 20, 2015
37
2
8
44
Between me and my partner this is the third out of seven or so Samsung SSDs. All of them bought at different times and used in completely different computers from different manufacturers with different OSs over the past three to four years.

It's showing the same signs that harddrives show when failing under Linux. Logs full of messages like this:

[4839597.453959] sd 4:0:0:0: [sda] FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_TIMEOUT
[4839597.453962] sd 4:0:0:0: [sda] CDB:
[4839597.453964] Write(10): 2a 00 59 6f 05 40 00 05 40 00
[4839597.453970] blk_update_request: I/O error, dev sda, sector 1500448064
[4839597.453996] EXT4-fs warning (device dm-1): ext4_end_bio:317: I/O error -5 writing to inode 6166928 (offset 301989888 size 8388608 starting block 187492520)
[4839597.453998] Buffer I/O error on device dm-1, logical block 187492520
[4839597.454000] Buffer I/O error on device dm-1, logical block 187492521
[4839597.454001] Buffer I/O error on device dm-1, logical block 187492522
[4839597.454002] Buffer I/O error on device dm-1, logical block 187492523
[4839597.454003] Buffer I/O error on device dm-1, logical block 187492524
[4839597.454005] Buffer I/O error on device dm-1, logical block 187492525
[4839597.454006] Buffer I/O error on device dm-1, logical block 187492526
[4839597.454007] Buffer I/O error on device dm-1, logical block 187492527
[4839597.454008] Buffer I/O error on device dm-1, logical block 187492528
[4839597.454009] Buffer I/O error on device dm-1, logical block 187492529
[4839597.454063] sd 4:0:0:0: [sda] FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_TIMEOUT
[4839597.454064] sd 4:0:0:0: [sda] CDB:
[4839597.454065] Write(10): 2a 00 59 6f 0a 80 00 05 40 00
[4839597.454069] blk_update_request: I/O error, dev sda, sector 1500449408
[4839597.454093] EXT4-fs warning (device dm-1): ext4_end_bio:317: I/O error -5 writing to inode 6166928 (offset 301989888 size 8388608 starting block 187492688)
[4839597.454150] sd 4:0:0:0: [sda] FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_TIMEOUT

and this

[4840061.919833] ata5.00: exception Emask 0x0 SAct 0x1fffe0 SErr 0x0 action 0x6 frozen
[4840061.919839] ata5.00: failed command: WRITE FPDMA QUEUED
[4840061.919843] ata5.00: cmd 61/00:28:40:2f:d1/16:00:59:00:00/40 tag 5 ncq 2883584 out
[4840061.919843] res 40/00:01:09:4f:c2/00:00:00:00:00/00 Emask 0x4 (timeout)
[4840061.919845] ata5.00: status: { DRDY }
[4840061.919847] ata5.00: failed command: WRITE FPDMA QUEUED
[4840061.919850] ata5.00: cmd 61/c0:30:40:45:d1/0f:00:59:00:00/40 tag 6 ncq 2064384 out
[4840061.919850] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
[4840061.919851] ata5.00: status: { DRDY }


That just go on and on.

Also despite the computer not having any load, operations that normally completely take hundred times longer or more.
 
T_Minus

T_Minus

Build. Break. Fix. Repeat
Feb 15, 2015
7,641
2,058
113
I'm sending my 1st RMA tomorrow for Intel SSD.

Intel had "No Record" of the drive in their warranty via website so I did a live chat, 10 minutes later I had RMA info in e-mail.

Basically print, mail, and wait...
 
M

Magical Penguin

Member
Jul 20, 2015
37
2
8
44
I don't think it is the trim bug, is that the messages the trim bug gives? I don't see anything about that message and the trim bug after a quick google search.

Drive has been in use for around a year and a half, current uptime on the laptop (which serves as a server and is never moved), is 56 days. And it just started having this problem, I would have noticed it if it happened before.

I haven't bothered with the firmware upgrade, it sounds like a total pain, and without great results.

Besides, looking at the patch Samsung provided for the trim bug, it seems to specifically have to do with Raid 0: [PATCH] raid0: data corruption when using trim — Linux RAID Storage

I am not running RAID.
 
M

Magical Penguin

Member
Jul 20, 2015
37
2
8
44
I went through all my archived syslogs, and the first sign of the problem was 6:50 pm tonight.

Smart shows this:

SMART Attributes Data Structure revision number: 1
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE
5 Reallocated_Sector_Ct 0x0033 100 100 010 Pre-fail Always - 0
9 Power_On_Hours 0x0032 095 095 000 Old_age Always - 24419
12 Power_Cycle_Count 0x0032 099 099 000 Old_age Always - 119
177 Wear_Leveling_Count 0x0013 095 095 000 Pre-fail Always - 51
179 Used_Rsvd_Blk_Cnt_Tot 0x0013 100 100 010 Pre-fail Always - 0
181 Program_Fail_Cnt_Total 0x0032 100 100 010 Old_age Always - 0
182 Erase_Fail_Count_Total 0x0032 100 100 010 Old_age Always - 0
183 Runtime_Bad_Block 0x0013 100 100 010 Pre-fail Always - 0
187 Reported_Uncorrect 0x0032 100 100 000 Old_age Always - 0
190 Airflow_Temperature_Cel 0x0032 053 037 000 Old_age Always - 47
195 Hardware_ECC_Recovered 0x001a 200 200 000 Old_age Always - 0
199 UDMA_CRC_Error_Count 0x003e 099 099 000 Old_age Always - 1
235 Unknown_Attribute 0x0012 099 099 000 Old_age Always - 42
241 Total_LBAs_Written 0x0032 099 099 000 Old_age Always - 25298802271
 
rubylaser

rubylaser

Active Member
Jan 4, 2013
846
236
43
Michigan, USA
Magical Penguin said:
I went through all my archived syslogs, and the first sign of the problem was 6:50 pm tonight.

Smart shows this:

SMART Attributes Data Structure revision number: 1
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE
5 Reallocated_Sector_Ct 0x0033 100 100 010 Pre-fail Always - 0
9 Power_On_Hours 0x0032 095 095 000 Old_age Always - 24419
12 Power_Cycle_Count 0x0032 099 099 000 Old_age Always - 119
177 Wear_Leveling_Count 0x0013 095 095 000 Pre-fail Always - 51
179 Used_Rsvd_Blk_Cnt_Tot 0x0013 100 100 010 Pre-fail Always - 0
181 Program_Fail_Cnt_Total 0x0032 100 100 010 Old_age Always - 0
182 Erase_Fail_Count_Total 0x0032 100 100 010 Old_age Always - 0
183 Runtime_Bad_Block 0x0013 100 100 010 Pre-fail Always - 0
187 Reported_Uncorrect 0x0032 100 100 000 Old_age Always - 0
190 Airflow_Temperature_Cel 0x0032 053 037 000 Old_age Always - 47
195 Hardware_ECC_Recovered 0x001a 200 200 000 Old_age Always - 0
199 UDMA_CRC_Error_Count 0x003e 099 099 000 Old_age Always - 1
235 Unknown_Attribute 0x0012 099 099 000 Old_age Always - 42
241 Total_LBAs_Written 0x0032 099 099 000 Old_age Always - 25298802271
Click to expand...
It's hard to see those values not in a code block, but all of these values look good to me. No reallocted sectors, no UDMA errors, no bad blocks, etc.
 
M

Magical Penguin

Member
Jul 20, 2015
37
2
8
44
rubylaser said:
It's hard to see those values not in a code block, but all of these values look good to me. No reallocted sectors, no UDMA errors, no bad blocks, etc.
Click to expand...
Yep, the smart values look good to me as well. It's the logs and performance that are the problem. I'm in the middle of a second rsync of everything on the drive to a spare drive, once that is done I'll experiment more.
 
M

Magical Penguin

Member
Jul 20, 2015
37
2
8
44
Just an update...everyone who said the drive is probably okay seems to be correct. The problem seems to have disappeared. Knock on wood.
 
P

pricklypunter

Well-Known Member
Nov 10, 2015
1,714
520
113
Canada
I have seen a few issues like this, with various disks, being "fixed" by simply writing zero's to the whole disk with dd. I think now and then something in the garbage collection gets screwed up and the disk starts doing all sorts of funky things until it's cleaned :)
 
You must log in or register to reply here.
Top Bottom