[SOLVED] Recommendations for a new desktop computer for Linux Mint Mate version 20

michaelk · 01-26-2022, 01:57 PM

I would replace the drive. You have three critical parameters that are not good.

https://kb.acronis.com/content/9101

astrogeek · 01-26-2022, 02:02 PM

Its OK! The main thing that would help yourself and those offering help would be to try to read and consider suggestions as precisely as they are given. It often takes several posts to arrive at a point which seems simple to many of us, but is entirely new to you, making it frustrating for everyone.

Let's get the CODE tag thing under control so we don't have to revisit it again. Mrmazda provided two helpful links in post #57, and I have amended my own post #70 above to be very explicit about how to use them... you are trying but have not hit the target yet! Please review those and try to edit your post #74 as a simple example for yourself. It will be obvious when you get it right because you will see this...

Code:

tomgirl@tomgirl-OptiPlex-790 ~ $ sudo smartctl -x /dev/sda | pastebinit
[sudo] password for tomgirl:
Unable to read or parse the result page, it could be a server timeout or a change server-side, try with another Pastebin.
tomgirl@tomgirl-OptiPlex-790 ~ $

...instead of this!

code
tomgirl@tomgirl-OptiPlex-790 ~ $ sudo smartctl -x /dev/sda | pastebinit
[sudo] password for tomgirl:
Unable to read or parse the result page, it could be a server timeout or a change server-side, try with another Pastebin.
tomgirl@tomgirl-OptiPlex-790 ~ $
code

Once you get #74 to look right, go back and try the same thing on the longer data posts and see the difference it makes!

OrphanHome · 01-26-2022, 02:15 PM

MrMazda

Code:

tomgirl@tomgirl-OptiPlex-790 ~ $ sudo smartctl -x /dev/sda | pastebinit
[sudo] password for tomgirl: 
Unable to read or parse the result page, it could be a server timeout or a change server-side, try with another Pastebin.

tomgirl@tomgirl-OptiPlex-790 ~ $

OrphanHome · 01-26-2022, 02:32 PM

I had not seen the editing icons at the top line before. It's better, yes?

michaelk · 01-26-2022, 02:58 PM

We have necessary information.

Code:

SMART Attributes Data Structure revision number: 10
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME          FLAGS    VALUE WORST THRESH FAIL RAW_VALUE
  1 Raw_Read_Error_Rate     POSR--   060   053   006    -    123097403
  3 Spin_Up_Time            PO----   096   095   000    -    0
  4 Start_Stop_Count        -O--CK   096   096   020    -    4790
  5 Reallocated_Sector_Ct   PO--CK   100   100   036    -    8
  7 Seek_Error_Rate         POSR--   086   060   030    -    425666960
  9 Power_On_Hours          -O--CK   056   056   000    -    39249
 10 Spin_Retry_Count        PO--C-   100   100   097    -    0
 12 Power_Cycle_Count       -O--CK   096   096   020    -    4798
194 Temperature_Celsius     -O---K   039   057   000    -    39
195 Hardware_ECC_Recovered  -O-RC-   060   053   000    -    123097403
197 Current_Pending_Sector  -O--C-   100   100   000    -    0
198 Offline_Uncorrectable   ----C-   100   100   000    -    0
199 UDMA_CRC_Error_Count    -OSRCK   200   200   000    -    0
200 Multi_Zone_Error_Rate   ------   100   253   000    -    0
202 Data_Address_Mark_Errs  -O--CK   100   253   000    -    0

After reexamination the only critical parameter is the Reallocated_Sector_Ct. The other numbers while large probably are only an indication the drive is just old. However, IMHO it makes sense since your going to install a new operating system that now is the time to replace the drive.

astrogeek · 01-26-2022, 02:59 PM

Quote:

Originally Posted by OrphanHome

I had not seen the editing icons at the top line before. It's better, yes?

Ah yes! What a difference in readability that makes, Thanks!

Arnulf · 01-26-2022, 03:03 PM

Quote:

Originally Posted by OrphanHome

Code:

5 Reallocated_Sector_Ct   PO--CK   100   100   036    -    8

Any value > 0 for "Reallocated Sector Count" (Reallocated_Sector_Ct) is bad for a HDD in a system that's daily used or stores critical data. Affected HDD should be replaced. It may be used in a museal system without any critical data stored on it. Before doing this, run badblocks -wsv over this HDD. If Reallocated_Sector_Ct hasn't increased after badblocks run HDD can be used in a museal system. Otherwise it should be disposed.

OrphanHome · 01-26-2022, 03:24 PM

Thank you everyone for your assistance and patience. You're the best.

Tommi Jordon

mrmazda · 01-26-2022, 05:06 PM

Quote:

Originally Posted by Arnulf

Any value > 0 for "Reallocated Sector Count" (Reallocated_Sector_Ct) is bad for a HDD in a system that's daily used or stores critical data. Affected HDD should be replaced.

IME this is excessive paranoia. 8 bad sectors could have been there on a brand new HDD. Currently there are zero pending sectors, which is good. I would recheck periodically with smartctl and if you see them increasing, then yes, replace ASAP. If you go a month or 3 with no new ones, back off the testing and keep using until you have some other reason to replace. Need for more space or speed are reason enough.

You only have 4.5 years of power up time on what you have. There's likely at least that much more time left on it. Given it's 10 or so years old now, it might still be good 20 years from now.

obobskivich · 01-27-2022, 11:41 AM

Quote:

Originally Posted by mrmazda

IME this is excessive paranoia. 8 bad sectors could have been there on a brand new HDD. Currently there are zero pending sectors, which is good. I would recheck periodically with smartctl and if you see them increasing, then yes, replace ASAP. If you go a month or 3 with no new ones, back off the testing and keep using until you have some other reason to replace. Need for more space or speed are reason enough.

You only have 4.5 years of power up time on what you have. There's likely at least that much more time left on it. Given it's 10 or so years old now, it might still be good 20 years from now.

Fully agree. Also keep in mind these are 'RAW' values, which are meaningless without context (except POH) - nothing here looks 'scary' or 'dangerous' and it passed all tests. I don't see any critical failures or 'it will die instantly and all your data will be gone forever' concerns. Of course standard guidance to have backups always applies - no single device should be the arbiter of 'will my data survive' - the failure of a device is more about how much inconvenience can you tolerate, not gambling with data loss.

Also remember: this can all be done in a few clicks in a GUI as well.