Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.user > #243387 > unrolled thread

Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

Back to article view | Back to linux.debian.user

Contents

  Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action  0x0 Heladu <helaheladu38@gmail.com> - 2021-12-23 23:10 +0100
    Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0  action 0x0 "Alexander V. Makartsev" <avbetev@gmail.com> - 2021-12-23 23:40 +0100
      Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0 rhkramer@gmail.com - 2021-12-24 16:40 +0100
        Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0  action 0x0 "Alexander V. Makartsev" <avbetev@gmail.com> - 2021-12-24 17:30 +0100
        Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0  action 0x0 David Christensen <dpchrist@holgerdanske.com> - 2021-12-24 21:10 +0100
        Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0 gene heskett <gheskett@shentel.net> - 2021-12-24 22:30 +0100
      Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0  action 0x0 Heladu <helaheladu38@gmail.com> - 2021-12-24 19:10 +0100
        Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0  action 0x0 "Alexander V. Makartsev" <avbetev@gmail.com> - 2021-12-24 20:20 +0100
    Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0  action 0x0 Dan Ritter <dsr@randomstring.org> - 2021-12-24 00:40 +0100
      Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0  action 0x0 Heladu <helaheladu38@gmail.com> - 2021-12-24 19:10 +0100
        Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0  action 0x0 David Christensen <dpchrist@holgerdanske.com> - 2021-12-24 21:30 +0100
          Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0 "Thomas Schmitt" <scdbackup@gmx.net> - 2021-12-24 22:00 +0100
            Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0  action 0x0 David Christensen <dpchrist@holgerdanske.com> - 2021-12-25 01:20 +0100
    Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0  action 0x0 David Christensen <dpchrist@holgerdanske.com> - 2021-12-24 00:50 +0100
      Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0 Glenn <ve9gj@napan.com> - 2021-12-24 16:10 +0100
    Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0  action 0x0 James Dutton <james.dutton@gmail.com> - 2021-12-24 16:30 +0100
      Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0 Stefan Monnier <monnier@iro.umontreal.ca> - 2021-12-24 18:10 +0100
      Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0  action 0x0 Heladu <helaheladu38@gmail.com> - 2021-12-24 19:10 +0100

#243387 — Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

Greetings,
I've been experiencing a lot of slowness in general when the system attempts
to read from the hard drive disk. I use Debian 10 Buster with the MATE desktop
environment and simple things like opening the calendar applet or right
clicking to open the context menu takes longer than usual. I noticed the LED
indicator than turns on when reading from the disk also took longer to turn
off, so I decided to inspect the logs and I ran into these entries:

Dec 23 22:33:24 sigma kernel: [ 1250.853537] ata6.00: exception Emask 0x0 SAct
0x40000000 SErr 0x0 action 0x0
Dec 23 22:33:24 sigma kernel: [ 1250.853544] ata6.00: irq_stat 0x40000008
Dec 23 22:33:24 sigma kernel: [ 1250.853550] ata6.00: failed command: READ
FPDMA QUEUED
Dec 23 22:33:24 sigma kernel: [ 1250.853559] ata6.00: cmd
60/08:f0:10:96:2b/00:00:01:00:00/40 tag 30 ncq dma 4096 in
Dec 23 22:33:24 sigma kernel: [ 1250.853559]          res
41/40:00:10:96:2b/00:00:01:00:00/00 Emask 0x409 (media error) <F>
Dec 23 22:33:24 sigma kernel: [ 1250.853563] ata6.00: status: { DRDY ERR }
Dec 23 22:33:24 sigma kernel: [ 1250.853566] ata6.00: error: { UNC }
Dec 23 22:33:24 sigma kernel: [ 1250.855102] ata6.00: configured for UDMA/133
Dec 23 22:33:24 sigma kernel: [ 1250.855121] sd 5:0:0:0: [sda] tag#30 FAILED
Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
Dec 23 22:33:24 sigma kernel: [ 1250.855126] sd 5:0:0:0: [sda] tag#30 Sense
Key : Medium Error [current] 
Dec 23 22:33:24 sigma kernel: [ 1250.855130] sd 5:0:0:0: [sda] tag#30 Add.
Sense: Unrecovered read error - auto reallocate failed
Dec 23 22:33:24 sigma kernel: [ 1250.855135] sd 5:0:0:0: [sda] tag#30 CDB:
Read(10) 28 00 01 2b 96 10 00 00 08 00
Dec 23 22:33:24 sigma kernel: [ 1250.855139] print_req_error: I/O error, dev
sda, sector 19633680
Dec 23 22:33:24 sigma kernel: [ 1250.855162] ata6: EH complete

They happen every time the system experiences slow reads. Now, I did some
research and I've read some possible causes like a bad SATA cable or a
malfunctioning HDD or PSU. I booted from a Debian installer on an USB stick
and I ran fsck.ext4 to check the disk and it printed the partition was clean.

Given that fsck didn't print anything unusual, I decided to replace the SATA
cable. However, it's still happening.

The HDD is a 1TB 3.5" WD Blue SATA drive which was bought a year ago.

I'm certain this is not a software problem because I've been running the
system a whole year without any problem. Has anyone ever experienced this? Is
there a way I can reliably find the faulty component (HDD, PSU...) without
buying a new one and hoping that solves it?

Thank you very much in advance.

[toc] | [next] | [standalone]

#243388 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

[Multipart message — attachments visible in raw view] — view raw

On 24.12.2021 02:51, Heladu wrote:
> Greetings,
> I've been experiencing a lot of slowness in general when the system attempts
> to read from the hard drive disk. I use Debian 10 Buster with the MATE desktop
> environment and simple things like opening the calendar applet or right
> clicking to open the context menu takes longer than usual. I noticed the LED
> indicator than turns on when reading from the disk also took longer to turn
> off, so I decided to inspect the logs and I ran into these entries:
> ...
> They happen every time the system experiences slow reads. Now, I did some
> research and I've read some possible causes like a bad SATA cable or a
> malfunctioning HDD or PSU. I booted from a Debian installer on an USB stick
> and I ran fsck.ext4 to check the disk and it printed the partition was clean.
>
> Given that fsck didn't print anything unusual, I decided to replace the SATA
> cable. However, it's still happening.
>
> The HDD is a 1TB 3.5" WD Blue SATA drive which was bought a year ago.
>
> I'm certain this is not a software problem because I've been running the
> system a whole year without any problem. Has anyone ever experienced this? Is
> there a way I can reliably find the faulty component (HDD, PSU...) without
> buying a new one and hoping that solves it?
>
> Thank you very much in advance.
>
You can review SMART attributes which keep track of device's health and 
metrics.
This utility is part of "smartmontools" package.
Run this one-liner to see values of relevant attributes:
     $ sudo smartctl -A /dev/sda | grep -E '5 Realloc|183 Runtime|197 
Current|199 UDMA'

Here is what they should like on perfectly fine hard drive:
     $ sudo smartctl -A /dev/sda | grep -E '5 Realloc|183 Runtime|197 
Current|199 UDMA'
       5 Reallocated_Sector_Ct   0x0033   100   100   010 Pre-fail  
Always       -       0
     183 Runtime_Bad_Block       0x0032   100   100   000 Old_age   
Always       -       0
     197 Current_Pending_Sector  0x0012   100   100   000 Old_age   
Always       -       0
     199 UDMA_CRC_Error_Count    0x003e   200   200   000 Old_age   
Always       -       0

Raw values are displayed on the right and they all zeroes. Post the 
output you got with next reply.
You should backup or "ddrescue" your data from this drive and 
RMA\replace it or better switch to SSD disk.

-- 
With kindest regards, Alexander.

⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org
⠈⠳⣄⠀⠀⠀⠀

[toc] | [prev] | [next] | [standalone]

#243410 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

On Thursday, December 23, 2021 05:30:32 PM Alexander V. Makartsev wrote:
> You can review SMART attributes which keep track of device's health and
> metrics.
> This utility is part of "smartmontools" package.
> Run this one-liner to see values of relevant attributes:
>      $ sudo smartctl -A /dev/sda | grep -E '5 Realloc|183 Runtime|197
> Current|199 UDMA'

I'm not the OP, and not  very familiar with smartmon / smartctl, but I ran the 
recommended command on the two disks in my oldest system, and the results are 
posted below.  (Aside: at some point, in the near future, I'll read the 
relevant manpage to better understand that output.)

I also see the advice from Dave Christensen on additional tests to run and 
will try those in the near future, ideally next week.

/dev/sda is an SSD (which hold my system and doesn't get much writing), 
/dev/sdb is an HDD (which holds my "user data").

Should I be worried?

root@s19:~# smartctl -A /dev/sda | grep -E '5 Realloc|183 Runtime|197 
> Current|199 UDMA'                                                                                    
  5 Reallocated_Sector_Ct   0x0033   100   100   036    Pre-fail  Always       
-       0               
183 Runtime_Bad_Block       0x0032   100   100   000    Old_age   Always       
-       0               
197 Current_Pending_Sector  0x0012   100   100   000    Old_age   Always       
-       0               
199 UDMA_CRC_Error_Count    0x003e   200   200   000    Old_age   Always       
-       0               
root@s19:~# smartctl -A /dev/sdb | grep -E '5 Realloc|183 Runtime|197                                  
Current|199 UDMA'                                                                                      
  5 Reallocated_Sector_Ct   0x0032   100   100   000    Old_age   Always       
-       12              
root@s19:~#  

[toc] | [prev] | [next] | [standalone]

#243412 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

[Multipart message — attachments visible in raw view] — view raw

On 24.12.2021 20:31, rhkramer@gmail.com wrote:
> ...
> /dev/sdb is an HDD (which holds my "user data").
>
> Should I be worried?
>
> root@s19:~# smartctl -A /dev/sdb | grep -E '5 Realloc|183 Runtime|197
> Current|199 UDMA'
>    5 Reallocated_Sector_Ct   0x0032   100   100   000    Old_age   Always
> -       12
> root@s19:~#
>
Attribute #5 shows if the drive in question ever encountered and 
successfully remapped a 'bad block'. In your case it happened at least 
12 times.
They could've happen a few years ago, drive's firmware recovered from 
them and it was working fine ever since¹.
But they also could've happen recently within a few days and in that 
case the drive is failing and should be replaced ASAP.

For future investigation you should see full output of:
     # smartctl -A /dev/sdb

And also SMART logs, to see when was last media error encountered:
     # smartctl -l error /dev/sda

For now one thing is certain, your HDD had media errors, is not 
reliable² and you should have a good backup of data from it.
It's good idea to configure 'smartd' on all hosts to monitor health 
state of your drives and notify you by mail if something happened.


¹ I have a 320GB HDD for a five years with, I think, contaminated 
platter. When it tries to read from some LBA range it fails with media 
errors, but if I request to read past that LBA range it works fine.
As a workaround, I've repartitioned it effectively cutting off that 
faulty LBA range (around 50GB) and use it as a portable drive to carry 
some not important data and to store additional backup copies.
I always expect it to die, but it still works fine to this day.

² Any hardware, despite being new or old, could fail at any time. Always 
expect failure and have a good backup.

-- 
With kindest regards, Alexander.

⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org
⠈⠳⣄⠀⠀⠀⠀

[toc] | [prev] | [next] | [standalone]

#243425 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

On 12/24/21 7:31 AM, rhkramer@gmail.com wrote:
> On Thursday, December 23, 2021 05:30:32 PM Alexander V. Makartsev wrote:
>> You can review SMART attributes which keep track of device's health and
>> metrics.
>> This utility is part of "smartmontools" package.
>> Run this one-liner to see values of relevant attributes:
>>       $ sudo smartctl -A /dev/sda | grep -E '5 Realloc|183 Runtime|197
>> Current|199 UDMA'
> 
> I'm not the OP, and not  very familiar with smartmon / smartctl, but I ran the
> recommended command on the two disks in my oldest system, and the results are
> posted below.  (Aside: at some point, in the near future, I'll read the
> relevant manpage to better understand that output.)
> 
> I also see the advice from Dave Christensen on additional tests to run and
> will try those in the near future, ideally next week.
> 
> /dev/sda is an SSD (which hold my system and doesn't get much writing),
> /dev/sdb is an HDD (which holds my "user data").
> 
> Should I be worried?
> 
> root@s19:~# smartctl -A /dev/sda | grep -E '5 Realloc|183 Runtime|197
>> Current|199 UDMA'
>    5 Reallocated_Sector_Ct   0x0033   100   100   036    Pre-fail  Always
> -       0
> 183 Runtime_Bad_Block       0x0032   100   100   000    Old_age   Always
> -       0
> 197 Current_Pending_Sector  0x0012   100   100   000    Old_age   Always
> -       0
> 199 UDMA_CRC_Error_Count    0x003e   200   200   000    Old_age   Always
> -       0
> root@s19:~# smartctl -A /dev/sdb | grep -E '5 Realloc|183 Runtime|197
> Current|199 UDMA'
>    5 Reallocated_Sector_Ct   0x0032   100   100   000    Old_age   Always
> -       12
> root@s19:~#


Examining specific SMART parameters out of context may be useful for 
someone who is familiar with specific drives, but I suggest including 
the entire report when posting to a mailing list (I typically redact the 
serial number):

# smartctl -x /dev/sda

# smartctl -x /dev/sdb


While SMART reports are mostly standardized, each manufacturer may 
provide specific parameters and/or specific interpretations.  The whole 
report should include model number, hardware version, firmware version, 
etc., of the drive, which can be used to find a manufacturer document 
that explains these details.


David

[toc] | [prev] | [next] | [standalone]

#243429 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

On Friday, December 24, 2021 10:31:43 AM EST rhkramer@gmail.com wrote:
> On Thursday, December 23, 2021 05:30:32 PM Alexander V. Makartsev wrote:
> > You can review SMART attributes which keep track of device's health and
> > metrics.
> > This utility is part of "smartmontools" package.
> > 
> > Run this one-liner to see values of relevant attributes:
> >      $ sudo smartctl -A /dev/sda | grep -E '5 Realloc|183 Runtime|197
> > 
> > Current|199 UDMA'
> 
> I'm not the OP, and not  very familiar with smartmon / smartctl, but I ran
> the recommended command on the two disks in my oldest system, and the
> results are posted below.  (Aside: at some point, in the near future, I'll
> read the relevant manpage to better understand that output.)
> 
> I also see the advice from Dave Christensen on additional tests to run and
> will try those in the near future, ideally next week.
> 
> /dev/sda is an SSD (which hold my system and doesn't get much writing),
> /dev/sdb is an HDD (which holds my "user data").
> 
> Should I be worried?
> 
> root@s19:~# smartctl -A /dev/sda | grep -E '5 Realloc|183 Runtime|197
> 
> > Current|199 UDMA'
> 
>   5 Reallocated_Sector_Ct   0x0033   100   100   036    Pre-fail  Always
> -       0

Yes, and that 36 should be watched, any increase means that drive is on its 
way out. Usually quicker than next week. I'd replace it just for S&G, but 
today I'd replace it with an SSD, not spinning rust. Among other things the 
SSD is around 4x faster. I just had a new 2t spinning rust go belly up in the 
night, so I now boot from a 500G SSD, and put in 4 1T SSD's in a raid10 for
/home.

[...]

Cheers, Gene Heskett.
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page <http://geneslinuxbox.net:6309/gene>

[toc] | [prev] | [next] | [standalone]

#243418 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

Hello,
First of all, thanks for the reply.

El vie, 24-12-2021 a las 03:30 +0500, Alexander V. Makartsev escribió:
>  You can review SMART attributes which keep track of device's health and
> metrics.
> This utility is part of "smartmontools" package.
> Run this one-liner to see values of relevant attributes:
>     $ sudo smartctl -A /dev/sda | grep -E '5 Realloc|183 Runtime|197
> Current|199 UDMA'
> 
> Here is what they should like on perfectly fine hard drive:
>     $ sudo smartctl -A /dev/sda | grep -E '5 Realloc|183 Runtime|197
> Current|199 UDMA'
>       5 Reallocated_Sector_Ct   0x0033   100   100   010    Pre-fail 
> Always       -       0
>     183 Runtime_Bad_Block       0x0032   100   100   000    Old_age  
> Always       -       0
>     197 Current_Pending_Sector  0x0012   100   100   000    Old_age  
> Always       -       0
>     199 UDMA_CRC_Error_Count    0x003e   200   200   000    Old_age  
> Always       -       0
> 
> Raw values are displayed on the right and they all zeroes. Post the output
> you got with next reply.
> You should backup or "ddrescue" your data from this drive and RMA\replace it
> or better switch to SSD disk.
> 

Well, I installed the package in question and ran the very same command you
said. However, the Runtime_Bad_Block attribute doesn't appear. Here's the
output:
$ sudo smartctl -A /dev/sda | grep -E '5 Realloc|183 Runtime|197 Current|199
UDMA'
  5 Reallocated_Sector_Ct   0x0033   200   200   140    Pre-
fail  Always       -       0
197
Current_Pending_Sector  0x0032   200   200   000    Old_age   Always       -  
     1
199
UDMA_CRC_Error_Count    0x0032   200   200   000    Old_age   Always       -  
     0

If I understood correctly what I read, this means there's one sector waiting
for remapping, right? And why doesn't the Runtime_Bad_Block attribute appear?

[toc] | [prev] | [next] | [standalone]

#243422 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

[Multipart message — attachments visible in raw view] — view raw

On 24.12.2021 22:44, Heladu wrote:
> Hello,
> First of all, thanks for the reply.
>
> El vie, 24-12-2021 a las 03:30 +0500, Alexander V. Makartsev escribió:
>>   You can review SMART attributes which keep track of device's health and
>> metrics.
>> This utility is part of "smartmontools" package.
>> Run this one-liner to see values of relevant attributes:
>>      $ sudo smartctl -A /dev/sda | grep -E '5 Realloc|183 Runtime|197
>> Current|199 UDMA'
>>
>> Here is what they should like on perfectly fine hard drive:
>>      $ sudo smartctl -A /dev/sda | grep -E '5 Realloc|183 Runtime|197
>> Current|199 UDMA'
>>        5 Reallocated_Sector_Ct   0x0033   100   100   010    Pre-fail
>> Always       -       0
>>      183 Runtime_Bad_Block       0x0032   100   100   000    Old_age
>> Always       -       0
>>      197 Current_Pending_Sector  0x0012   100   100   000    Old_age
>> Always       -       0
>>      199 UDMA_CRC_Error_Count    0x003e   200   200   000    Old_age
>> Always       -       0
>>
>> Raw values are displayed on the right and they all zeroes. Post the output
>> you got with next reply.
>> You should backup or "ddrescue" your data from this drive and RMA\replace it
>> or better switch to SSD disk.
>>
> Well, I installed the package in question and ran the very same command you
> said. However, the Runtime_Bad_Block attribute doesn't appear. Here's the
> output:
> $ sudo smartctl -A /dev/sda | grep -E '5 Realloc|183 Runtime|197 Current|199
> UDMA'
>    5 Reallocated_Sector_Ct   0x0033   200   200   140    Pre-
> fail  Always       -       0
> 197
> Current_Pending_Sector  0x0032   200   200   000    Old_age   Always       -
>       1
> 199
> UDMA_CRC_Error_Count    0x0032   200   200   000    Old_age   Always       -
>       0
>
> If I understood correctly what I read, this means there's one sector waiting
> for remapping, right?
That is correct. Drive's firmware should take care of it automatically 
or during short or extended self-tests.
I suggest to backup your data before performing any self-tests on the drive.
They are non-destructive by nature, but you never know what could happen 
and what was the initial cause for 'bad blocks' to appear.

> And why doesn't the Runtime_Bad_Block attribute appear?
>
SMART attributes and functionality could be different, depending on 
manufacturer or\and model of the device.
My example was from HDD made by Seagate. Maybe it was a mistake on my 
part, suggesting a one-liner with 'grep', causing a confusion.


-- 
With kindest regards, Alexander.

⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org
⠈⠳⣄⠀⠀⠀⠀

[toc] | [prev] | [next] | [standalone]

#243389 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

Heladu wrote: 
> Greetings,
> I've been experiencing a lot of slowness in general when the system attempts
> to read from the hard drive disk. I use Debian 10 Buster with the MATE desktop
> environment and simple things like opening the calendar applet or right
> clicking to open the context menu takes longer than usual. I noticed the LED
> indicator than turns on when reading from the disk also took longer to turn
> off, so I decided to inspect the logs and I ran into these entries:
> 
> Dec 23 22:33:24 sigma kernel: [ 1250.853537] ata6.00: exception Emask 0x0 SAct
> 0x40000000 SErr 0x0 action 0x0
> Dec 23 22:33:24 sigma kernel: [ 1250.853544] ata6.00: irq_stat 0x40000008
> Dec 23 22:33:24 sigma kernel: [ 1250.853550] ata6.00: failed command: READ
> FPDMA QUEUED
> Dec 23 22:33:24 sigma kernel: [ 1250.853559] ata6.00: cmd
> 60/08:f0:10:96:2b/00:00:01:00:00/40 tag 30 ncq dma 4096 in
> Dec 23 22:33:24 sigma kernel: [ 1250.853559]          res
> 41/40:00:10:96:2b/00:00:01:00:00/00 Emask 0x409 (media error) <F>
> Dec 23 22:33:24 sigma kernel: [ 1250.853563] ata6.00: status: { DRDY ERR }
> Dec 23 22:33:24 sigma kernel: [ 1250.853566] ata6.00: error: { UNC }
> Dec 23 22:33:24 sigma kernel: [ 1250.855102] ata6.00: configured for UDMA/133
> Dec 23 22:33:24 sigma kernel: [ 1250.855121] sd 5:0:0:0: [sda] tag#30 FAILED
> Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
> Dec 23 22:33:24 sigma kernel: [ 1250.855126] sd 5:0:0:0: [sda] tag#30 Sense
> Key : Medium Error [current] 
> Dec 23 22:33:24 sigma kernel: [ 1250.855130] sd 5:0:0:0: [sda] tag#30 Add.
> Sense: Unrecovered read error - auto reallocate failed
> Dec 23 22:33:24 sigma kernel: [ 1250.855135] sd 5:0:0:0: [sda] tag#30 CDB:
> Read(10) 28 00 01 2b 96 10 00 00 08 00
> Dec 23 22:33:24 sigma kernel: [ 1250.855139] print_req_error: I/O error, dev
> sda, sector 19633680
> Dec 23 22:33:24 sigma kernel: [ 1250.855162] ata6: EH complete
> 
> They happen every time the system experiences slow reads. Now, I did some
> research and I've read some possible causes like a bad SATA cable or a
> malfunctioning HDD or PSU. I booted from a Debian installer on an USB stick
> and I ran fsck.ext4 to check the disk and it printed the partition was clean.

It's true a bad cable can cause this, but you swapped cables,
so:

Back up the data immediately; this disk is failing.

If it is under warranty, you'll get a replacement. That can take
weeks, though, so you should go buy another disk today.

Sorry.

-dsr-

[toc] | [prev] | [next] | [standalone]

#243419 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

El jue, 23-12-2021 a las 18:12 -0500, Dan Ritter escribió:
> It's true a bad cable can cause this, but you swapped cables,
> so:
> 
> Back up the data immediately; this disk is failing.
> 
> If it is under warranty, you'll get a replacement. That can take
> weeks, though, so you should go buy another disk today.
> 
> Sorry.
> 
> -dsr-

Yeah, it's clearly a faulty disk. But I wonder why it has started to fail
"sooner". All HDDs I've had have lasted for at least 5 years... I guess I
wasn't lucky with this one.

Thanks for the reply.

[toc] | [prev] | [next] | [standalone]

#243426 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

On 12/24/21 9:48 AM, Heladu wrote:
> El jue, 23-12-2021 a las 18:12 -0500, Dan Ritter escribió:
>> It's true a bad cable can cause this, but you swapped cables,
>> so:
>>
>> Back up the data immediately; this disk is failing.
>>
>> If it is under warranty, you'll get a replacement. That can take
>> weeks, though, so you should go buy another disk today.
>>
>> Sorry.
>>
>> -dsr-
> 
> Yeah, it's clearly a faulty disk. 


Did you test the power supply, the memory, run a long SMART test, and 
correctly interpret the complete SMART report?  Did you double-check 
your conclusion somehow?


Computers are complicated beasts.  A fault in one portion can cause 
problems in another portion.  Bad power supplies are the worst -- they 
can cause anything and everything to malfunction, they can permanently 
damage other hardware, and they can cause electrical shock, 
electrocution, and/or fires.


> But I wonder why it has started to fail
> "sooner". All HDDs I've had have lasted for at least 5 years... I guess I
> wasn't lucky with this one.
> 
> Thanks for the reply.


David

[toc] | [prev] | [next] | [standalone]

#243427 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

Hi,

David Christensen wrote:
> Did you test the power supply, the memory, run a long SMART test, and
> correctly interpret the complete SMART report?

The following lines from the original post forward a message from the
drive's firmware.

Heladu wrote:
> > Dec 23 22:33:24 sigma kernel: [ 1250.855121] sd 5:0:0:0: [sda] tag#30 FAILED
> > Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
> > Dec 23 22:33:24 sigma kernel: [ 1250.855126] sd 5:0:0:0: [sda] tag#30 Sense
> > Key : Medium Error [current]
> > Dec 23 22:33:24 sigma kernel: [ 1250.855130] sd 5:0:0:0: [sda] tag#30 Add.
> > Sense: Unrecovered read error - auto reallocate failed
> > Dec 23 22:33:24 sigma kernel: [ 1250.855135] sd 5:0:0:0: [sda] tag#30 CDB:
> > Read(10) 28 00 01 2b 96 10 00 00 08 00

The drive answers to an SCSI READ command from the computer that its
storage medium caused an error. No problem outside the drive should
be able to cause this.


Have a nice day :)

Thomas

[toc] | [prev] | [next] | [standalone]

#243432 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

On 12/24/21 12:58 PM, Thomas Schmitt wrote:
> Hi,
> 
> David Christensen wrote:
>> Did you test the power supply, the memory, run a long SMART test, and
>> correctly interpret the complete SMART report? >
> The following lines from the original post forward a message from the
> drive's firmware.
> 
> Heladu wrote:
>>> Dec 23 22:33:24 sigma kernel: [ 1250.855121] sd 5:0:0:0: [sda] tag#30 FAILED
>>> Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
>>> Dec 23 22:33:24 sigma kernel: [ 1250.855126] sd 5:0:0:0: [sda] tag#30 Sense
>>> Key : Medium Error [current]
>>> Dec 23 22:33:24 sigma kernel: [ 1250.855130] sd 5:0:0:0: [sda] tag#30 Add.
>>> Sense: Unrecovered read error - auto reallocate failed
>>> Dec 23 22:33:24 sigma kernel: [ 1250.855135] sd 5:0:0:0: [sda] tag#30 CDB:
>>> Read(10) 28 00 01 2b 96 10 00 00 08 00
> 
> The drive answers to an SCSI READ command from the computer that its
> storage medium caused an error. No problem outside the drive should
> be able to cause this.


Assuming a conventional desktop environment with no extremes or events, 
agreed.  But, I would still test the power supply, test the memory, run 
a long SMART test, and post the complete SMART report.  If the SMART 
overall-health self-assessment test result says "FAIL", then recycle the 
drive.  But if the result is "PASSED", then I would consider keeping it. 
  In any case, I would get more drives and set up RAID.


David

[toc] | [prev] | [next] | [standalone]

#243392 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

On 12/23/21 1:51 PM, Heladu wrote:
> Greetings,
> I've been experiencing a lot of slowness in general when the system attempts
> to read from the hard drive disk. I use Debian 10 Buster with the MATE desktop
> environment and simple things like opening the calendar applet or right
> clicking to open the context menu takes longer than usual. I noticed the LED
> indicator than turns on when reading from the disk also took longer to turn
> off, so I decided to inspect the logs and I ran into these entries:
> 
> Dec 23 22:33:24 sigma kernel: [ 1250.853537] ata6.00: exception Emask 0x0 SAct
> 0x40000000 SErr 0x0 action 0x0
> Dec 23 22:33:24 sigma kernel: [ 1250.853544] ata6.00: irq_stat 0x40000008
> Dec 23 22:33:24 sigma kernel: [ 1250.853550] ata6.00: failed command: READ
> FPDMA QUEUED
> Dec 23 22:33:24 sigma kernel: [ 1250.853559] ata6.00: cmd
> 60/08:f0:10:96:2b/00:00:01:00:00/40 tag 30 ncq dma 4096 in
> Dec 23 22:33:24 sigma kernel: [ 1250.853559]          res
> 41/40:00:10:96:2b/00:00:01:00:00/00 Emask 0x409 (media error) <F>
> Dec 23 22:33:24 sigma kernel: [ 1250.853563] ata6.00: status: { DRDY ERR }
> Dec 23 22:33:24 sigma kernel: [ 1250.853566] ata6.00: error: { UNC }
> Dec 23 22:33:24 sigma kernel: [ 1250.855102] ata6.00: configured for UDMA/133
> Dec 23 22:33:24 sigma kernel: [ 1250.855121] sd 5:0:0:0: [sda] tag#30 FAILED
> Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
> Dec 23 22:33:24 sigma kernel: [ 1250.855126] sd 5:0:0:0: [sda] tag#30 Sense
> Key : Medium Error [current]
> Dec 23 22:33:24 sigma kernel: [ 1250.855130] sd 5:0:0:0: [sda] tag#30 Add.
> Sense: Unrecovered read error - auto reallocate failed
> Dec 23 22:33:24 sigma kernel: [ 1250.855135] sd 5:0:0:0: [sda] tag#30 CDB:
> Read(10) 28 00 01 2b 96 10 00 00 08 00
> Dec 23 22:33:24 sigma kernel: [ 1250.855139] print_req_error: I/O error, dev
> sda, sector 19633680
> Dec 23 22:33:24 sigma kernel: [ 1250.855162] ata6: EH complete
> 
> They happen every time the system experiences slow reads. Now, I did some
> research and I've read some possible causes like a bad SATA cable or a
> malfunctioning HDD or PSU. I booted from a Debian installer on an USB stick
> and I ran fsck.ext4 to check the disk and it printed the partition was clean.
> 
> Given that fsck didn't print anything unusual, I decided to replace the SATA
> cable. However, it's still happening.
> 
> The HDD is a 1TB 3.5" WD Blue SATA drive which was bought a year ago.
> 
> I'm certain this is not a software problem because I've been running the
> system a whole year without any problem. Has anyone ever experienced this? Is
> there a way I can reliably find the faulty component (HDD, PSU...) without
> buying a new one and hoping that solves it?
> 
> Thank you very much in advance.


I own a power supply tester, so I would start by testing the power supply.


I suggest booting a memory test USB stick and letting it run for at 
least one pass (better, 24 hours).  Either of these should work:

https://www.memtest86.com/

https://www.memtest.org/


I suggest booting Debian and running a SMART long test:

# smartctl -t long /dev/sda


Check the test every hour until complete:

# smartctl -x /dev/sda


Please post the final smartctl(8) report.


David

[toc] | [prev] | [next] | [standalone]

#243404 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

[Multipart message — attachments visible in raw view] — view raw

99%+ this is a failing hard drive. If you value your data use ddrescue to transfer your failing drive to a new one as step one.   Then you can do further troubleshooting.
Besides smartctl tests what I do is boot a rescue disk such as sysrescuecd on a USB stick and then run 
dd_rescue -v /dev/sdX /dev/null
Any disk errors will become apparent. Make sure you understand dd_rescue read and write agrguments. You dont want to write to your disk!!

Good Luck
Glenn

On December 23, 2021 7:40:42 PM AST, David Christensen <dpchrist@holgerdanske.com> wrote:
>On 12/23/21 1:51 PM, Heladu wrote:
>> Greetings,
>> I've been experiencing a lot of slowness in general when the system attempts
>> to read from the hard drive disk. I use Debian 10 Buster with the MATE desktop
>> environment and simple things like opening the calendar applet or right
>> clicking to open the context menu takes longer than usual. I noticed the LED
>> indicator than turns on when reading from the disk also took longer to turn
>> off, so I decided to inspect the logs and I ran into these entries:
>> 
>> Dec 23 22:33:24 sigma kernel: [ 1250.853537] ata6.00: exception Emask 0x0 SAct
>> 0x40000000 SErr 0x0 action 0x0
>> Dec 23 22:33:24 sigma kernel: [ 1250.853544] ata6.00: irq_stat 0x40000008
>> Dec 23 22:33:24 sigma kernel: [ 1250.853550] ata6.00: failed command: READ
>> FPDMA QUEUED
>> Dec 23 22:33:24 sigma kernel: [ 1250.853559] ata6.00: cmd
>> 60/08:f0:10:96:2b/00:00:01:00:00/40 tag 30 ncq dma 4096 in
>> Dec 23 22:33:24 sigma kernel: [ 1250.853559]          res
>> 41/40:00:10:96:2b/00:00:01:00:00/00 Emask 0x409 (media error) <F>
>> Dec 23 22:33:24 sigma kernel: [ 1250.853563] ata6.00: status: { DRDY ERR }
>> Dec 23 22:33:24 sigma kernel: [ 1250.853566] ata6.00: error: { UNC }
>> Dec 23 22:33:24 sigma kernel: [ 1250.855102] ata6.00: configured for UDMA/133
>> Dec 23 22:33:24 sigma kernel: [ 1250.855121] sd 5:0:0:0: [sda] tag#30 FAILED
>> Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
>> Dec 23 22:33:24 sigma kernel: [ 1250.855126] sd 5:0:0:0: [sda] tag#30 Sense
>> Key : Medium Error [current]
>> Dec 23 22:33:24 sigma kernel: [ 1250.855130] sd 5:0:0:0: [sda] tag#30 Add.
>> Sense: Unrecovered read error - auto reallocate failed
>> Dec 23 22:33:24 sigma kernel: [ 1250.855135] sd 5:0:0:0: [sda] tag#30 CDB:
>> Read(10) 28 00 01 2b 96 10 00 00 08 00
>> Dec 23 22:33:24 sigma kernel: [ 1250.855139] print_req_error: I/O error, dev
>> sda, sector 19633680
>> Dec 23 22:33:24 sigma kernel: [ 1250.855162] ata6: EH complete
>> 
>> They happen every time the system experiences slow reads. Now, I did some
>> research and I've read some possible causes like a bad SATA cable or a
>> malfunctioning HDD or PSU. I booted from a Debian installer on an USB stick
>> and I ran fsck.ext4 to check the disk and it printed the partition was clean.
>> 
>> Given that fsck didn't print anything unusual, I decided to replace the SATA
>> cable. However, it's still happening.
>> 
>> The HDD is a 1TB 3.5" WD Blue SATA drive which was bought a year ago.
>> 
>> I'm certain this is not a software problem because I've been running the
>> system a whole year without any problem. Has anyone ever experienced this? Is
>> there a way I can reliably find the faulty component (HDD, PSU...) without
>> buying a new one and hoping that solves it?
>> 
>> Thank you very much in advance.
>
>
>I own a power supply tester, so I would start by testing the power supply.
>
>
>I suggest booting a memory test USB stick and letting it run for at 
>least one pass (better, 24 hours).  Either of these should work:
>
>https://www.memtest86.com/
>
>https://www.memtest.org/
>
>
>I suggest booting Debian and running a SMART long test:
>
># smartctl -t long /dev/sda
>
>
>Check the test every hour until complete:
>
># smartctl -x /dev/sda
>
>
>Please post the final smartctl(8) report.
>
>
>David
>

[toc] | [prev] | [next] | [standalone]

#243407 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

On Thu, 23 Dec 2021 at 22:09, Heladu <helaheladu38@gmail.com> wrote:
> Dec 23 22:33:24 sigma kernel: [ 1250.855130] sd 5:0:0:0: [sda] tag#30 Add.
> Sense: Unrecovered read error - auto reallocate failed
That is a faulty disk. Replace it. It has already lost the data stored
on one sector, and when this happens, more fail.
The "slow disk" you are experiencing is the disk trying to re-read a
faulty sector multiple times in the hope it will recover the data.
The disk will automatically try to relocate this data to a different
sector, and mark the bad one as bad.
This is the "reallocate" feature it mentions in the error message.
Except in this case, even though it tried to re-read the bad sector
multiple times in the hopes of recovering it, it failed to do so, thus
that sector's data is lost for-ever.
It is 100% a faulty disk, and 0% a cable problem.

[toc] | [prev] | [next] | [standalone]

#243413 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

> It is 100% a faulty disk, and 0% a cable problem.

It's indeed not a problem with the SATA connection.  And it is almost
assuredly a disk problem.  But FWIW it can also be a problem with the
power supply (speaking from experience, where such problems were
recurring for multiple drives on a specific (Orange-Pi-mini) machine,
depending on the load on the machine, the ambient temperature, ...).


        Stefan

[toc] | [prev] | [next] | [standalone]

#243417 — Re: Slow disk reads - exception Emask 0x0 SAct 0x6b0000 SErr 0x0 action 0x0

Hello,
El vie, 24-12-2021 a las 15:23 +0000, James Dutton escribió:
> Except in this case, even though it tried to re-read the bad sector
> multiple times in the hopes of recovering it, it failed to do so, thus
> that sector's data is lost for-ever.

I see... Well, in that case I won't bother trying to fix it and get a new one
instead.

Thanks for the reply.

[toc] | [prev] | [standalone]

Back to top | Article view | linux.debian.user

csiph-web