SSD failure or Debian 10 Incompatible ?

[XDm9wf8aS2aZsKh]

Don't add parameter to bottom file, but edit default cmd line :
GRUB_CMDLINE_LINUX_DEFAULT="quiet libata.force=1.5G"

[fred44nl]

ok thanks
my modifications for to-day are:
- purged apparmor
- added libata.force=1.5G

will keep a close eye

looked at Mint 19.1 Cinnamon
I understand why it must have been slow
so many services running, all sorts of background jobs and Cinnamon is really heavy.

[tamir]

Looking carefully at the behavior presented as Multiple ata errors in dmesg output are seen and SSD completely disappears from BIOS as well as from gparted live, this might be an unfortunate coincidence with a bad SSD. Installing new OS or any other writing demanding operation with a faulty SSD would cause issues; reducing writing speed perhaps could help in such cases this is why symptoms are reduce when you do.

As for running smartctl tools which causing crashes - SSD controller does not support some SMART queries, hence it is not a reliable method to test an SSD

We suggest to perform the following:
# 1. Get SSD info supposing it was detected as /dev/sda:
$ sudo smartctl -i /dev/sda
Model Family: SanDisk based SSDs
Device Model: SanDisk SSD U110 16GB
Serial Number: 144658401078
...

# 2. Supposing tested SSD is a boot device and the only disk in the system write 1000MB file to it:
$ dd if=/dev/urandom of=~/1000MB bs=1MB count=1000
1000+0 records in
1000+0 records out
1000000000 bytes (1.0 GB, 954 MiB) copied, 34.1745 s, 29.3 MB/s

# 3. Then look at dmesg output, search ata errors. It should not be any errors.
$ dmesg | tail

Should you receive an error, the SSD is faulty and the machine should be RMA'd for repairs.
The RMA procedure is documented at - https://www.fit-pc.com/web/support/rma/

[fred44nl]

I have a Kingspec m2.SSD, which gives me no errors.
done the dd-stress-test before, without problems.
but what I find strange, is that dd if=/dev/zero of=test bs=1048576 count=2048 gives me a speed of about 75 MB/s
but the specs from Kingspec mentions much, much higher performance.
as a comparison, I ran the same dd-command on my external usb3-hdd and this gave me about 110 MB/s
so, usb3 is faster than m2.
is that correct ??

[XDm9wf8aS2aZsKh]

Hi, I did your commands.

Code: Select all

smartctl -i /dev/sda
smartctl 6.6 2017-11-05 r4594 [x86_64-linux-4.19.0-5-amd64] (local build)
Copyright (C) 2002-17, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Device Model:     NT-128
Serial Number:    087110270294
LU WWN Device Id: 0 000000 000000000
Firmware Version: Q0407A
User Capacity:    126 701 535 232 bytes [126 GB]
Sector Size:      512 bytes logical/physical
Rotation Rate:    Solid State Device
Form Factor:      2.5 inches
Device is:        Not in smartctl database [for details use: -P showall]
ATA Version is:   ACS-2 (minor revision not indicated)
SATA Version is:  SATA 3.1, 6.0 Gb/s (current: 1.5 Gb/s)
Local Time is:    Fri Jul 19 10:28:28 2019 CEST
SMART support is: Available - device has SMART capability.
SMART support is: Enabled

With libata.force=1.5G

Code: Select all

dd if=/dev/urandom of=/home/user/test/10000MB bs=1MB count=10000
10000+0 enregistrements lus
10000+0 enregistrements écrits
10000000000 octets (10 GB, 9,3 GiB) copiés, 336,188 s, 29,7 MB/s

Without libata.force=1.5G

Code: Select all

dd if=/dev/urandom of=/home/user/test/10000MB bs=1MB count=10000
10000+0 enregistrements lus
10000+0 enregistrements écrits
10000000000 octets (10 GB, 9,3 GiB) copiés, 343,25 s, 29,1 MB/s

No error on dmesg.
But without libata.force=1.5G, 5 minutes after test, computer freeze, (hard drive read only, but can't show error !)
My SSD speed seems very low !

[fred44nl]

My SSD speed seems very low !

I am not so sure

fred44nl@Debian:~$ dd if=/dev/zero of=test bs=1048576 count=10
10+0 records gelezen
10+0 records geschreven
10485760 bytes (10 MB, 10 MiB) copied, 0,0234415 s, 447 MB/s
fred44nl@Debian:~$
fred44nl@Debian:~$ dd if=/dev/zero of=test bs=1048576 count=100
100+0 records gelezen
100+0 records geschreven
104857600 bytes (105 MB, 100 MiB) copied, 1,18676 s, 88,4 MB/s
fred44nl@Debian:~$
fred44nl@Debian:~$ dd if=/dev/zero of=test bs=1048576 count=1000
1000+0 records gelezen
1000+0 records geschreven
1048576000 bytes (1,0 GB, 1000 MiB) copied, 33,1974 s, 31,6 MB/s
fred44nl@Debian:~$
fred44nl@Debian:~$ dd if=/dev/zero of=test bs=1048576 count=2000
2000+0 records gelezen
2000+0 records geschreven
2097152000 bytes (2,1 GB, 2,0 GiB) copied, 80,9427 s, 25,9 MB/s

now the same for my usbhdd:

fred44nl@Debian:~$ dd if=/dev/zero of=/mnt/usbhdd/shares/test bs=1048576 count=20
20+0 records gelezen
20+0 records geschreven
20971520 bytes (21 MB, 20 MiB) copied, 0,0466646 s, 449 MB/s
fred44nl@Debian:~$
fred44nl@Debian:~$ dd if=/dev/zero of=/mnt/usbhdd/shares/test bs=1048576 count=20
20+0 records gelezen
20+0 records geschreven
20971520 bytes (21 MB, 20 MiB) copied, 0,242074 s, 86,6 MB/s
fred44nl@Debian:~$
fred44nl@Debian:~$ dd if=/dev/zero of=/mnt/usbhdd/shares/test bs=1048576 count=20
20+0 records gelezen
20+0 records geschreven
20971520 bytes (21 MB, 20 MiB) copied, 0,222982 s, 94,1 MB/s
fred44nl@Debian:~$
fred44nl@Debian:~$ dd if=/dev/zero of=/mnt/usbhdd/shares/test bs=1048576 count=20
20+0 records gelezen
20+0 records geschreven
20971520 bytes (21 MB, 20 MiB) copied, 0,234141 s, 89,6 MB/s
fred44nl@Debian:~$
fred44nl@Debian:~$ dd if=/dev/zero of=/mnt/usbhdd/shares/test bs=1048576 count=200
200+0 records gelezen
200+0 records geschreven
209715200 bytes (210 MB, 200 MiB) copied, 2,31704 s, 90,5 MB/s
fred44nl@Debian:~$
fred44nl@Debian:~$ dd if=/dev/zero of=/mnt/usbhdd/shares/test bs=1048576 count=2000
2000+0 records gelezen
2000+0 records geschreven
2097152000 bytes (2,1 GB, 2,0 GiB) copied, 21,4429 s, 97,8 MB/s

it could have something to do with the contents of buffers or caches
this dd-command is nice for stressing the SSD, but does not say much about the actual performance

[XDm9wf8aS2aZsKh]

Here is a review of this SSD : https://androidpctv.com/review-kingspec-nt128-ssd-m2/

They have better results for the same disk

[fred44nl]

did we have a look at

Code: Select all

sudo hdparm -i /dev/sda

/dev/sda:

Model=NT-64, FwRev=Q0407A, SerialNo=977112170455
Config={ HardSect NotMFM Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=0, SectSize=576, ECCbytes=4
BuffType=unknown, BuffSize=unknown, MaxMultSect=1, MultSect=1
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=123731968
IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 udma2 udma3 udma4 udma5 *udma6
AdvancedPM=yes: unknown setting WriteCache=disabled
Drive conforms to: Unspecified: ATA/ATAPI-2,3,4,5,6,7

* signifies the current active mode

[XDm9wf8aS2aZsKh]

Code: Select all

sudo hdparm -i /dev/sda

Code: Select all

/dev/sda:

 Model=NT-128, FwRev=Q0407A, SerialNo=087110270294
 Config={ HardSect NotMFM Fixed DTR>10Mbs }
 RawCHS=16383/16/63, TrkSize=0, SectSize=576, ECCbytes=4
 BuffType=unknown, BuffSize=unknown, MaxMultSect=1, MultSect=1
 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=247463936
 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
 PIO modes:  pio0 pio1 pio2 pio3 pio4 
 DMA modes:  mdma0 mdma1 mdma2 
 UDMA modes: udma0 udma1 udma2 udma3 udma4 udma5 *udma6 
 AdvancedPM=yes: disabled (255) WriteCache=enabled
 Drive conforms to: Unspecified:  ATA/ATAPI-2,3,4,5,6,7

 * signifies the current active mode

[fred44nl]

UDMA modes: udma0 udma1 udma2 udma3 udma4 udma5 *udma6
AdvancedPM=yes: disabled (255) WriteCache=enabled

yours has WriteCache=enabled.
mine is disabled

edit1:
what do you have in /etc/hdparm.conf

# -W Disable/enable the IDE drive's write-caching feature
#write_cache = off

edit2:
because I wanted to see what smartctl did for me, I installed smartmontools
this gave me an extra running service and in /var/log/syslog I found:

Jul 20 12:05:09 Debian systemd[1]: Reloading.
Jul 20 12:05:10 Debian systemd[1]: Started Self Monitoring and Reporting Technology (SMART) Daemon.
Jul 20 12:05:10 Debian smartd[7754]: smartd 6.6 2017-11-05 r4594 [x86_64-linux-4.19.0-5-amd64] (local build)
Jul 20 12:05:10 Debian smartd[7754]: Copyright (C) 2002-17, Bruce Allen, Christian Franke, http://www.smartmontools.org
Jul 20 12:05:10 Debian smartd[7754]: Opened configuration file /etc/smartd.conf
Jul 20 12:05:10 Debian smartd[7754]: Drive: DEVICESCAN, implied '-a' Directive on line 21 of file /etc/smartd.conf
Jul 20 12:05:10 Debian smartd[7754]: Configuration file /etc/smartd.conf was parsed, found DEVICESCAN, scanning devices
Jul 20 12:05:10 Debian systemd[1]: Reloading.
Jul 20 12:05:10 Debian smartd[7754]: Device: /dev/sda, type changed from 'scsi' to 'sat'
Jul 20 12:05:10 Debian smartd[7754]: Device: /dev/sda [SAT], opened
Jul 20 12:05:10 Debian smartd[7754]: Device: /dev/sda [SAT], NT-64, S/N:977112170455, WWN:0-000000-000000000, FW:Q0407A, 63.3 GB
Jul 20 12:05:10 Debian smartd[7754]: Device: /dev/sda [SAT], not found in smartd database.
Jul 20 12:05:10 Debian smartd[7754]: Device: /dev/sda [SAT], can't monitor Current_Pending_Sector count - no Attribute 197
Jul 20 12:05:10 Debian smartd[7754]: Device: /dev/sda [SAT], can't monitor Offline_Uncorrectable count - no Attribute 198
Jul 20 12:05:10 Debian smartd[7754]: Device: /dev/sda [SAT], is SMART capable. Adding to "monitor" list.
Jul 20 12:05:10 Debian smartd[7754]: Device: /dev/sdb [SAT], opened
Jul 20 12:05:10 Debian smartd[7754]: Device: /dev/sdb [SAT], TOSHIBA MQ01UBD100, S/N:376DTLJHT, WWN:0-000000-000000000, FW:AX001U, 1.00 TB
Jul 20 12:05:10 Debian smartd[7754]: Device: /dev/sdb [SAT], found in smartd database: Toshiba 2.5" HDD MQ01UBD... (USB 3.0)

so, I rapidly purged smartmontools and rebooted