[SOLVED] Reallocated sector error....

[stevep94]

I've just gone into my unRAID server and have got a yellow message at the top right saying "Unraid Cache Disk SMART health [5]"

 

The message goes on to say "Warning [UNRAID] - reallocated sector ct is 4096" and then lists the Samsung SSD that I've got installed as my cache drive.

 

Do I need to be worrying about this? I did a lot of reading and research when I set my unRAID server up but since then it's just sat there hosting my Plex server etc with very little input from me other than the occasional plugin/docker updates.

 

Thanks for any suggestions.

 

*edit*

 

I've just run the "SMART extended self-test" and it says "Completed Without Error" and under the attributes tab it says the "Type" was "Pre-Fail"?

Edited June 8, 2020 by stevep94

[Michael_P]

How old is it, and what's the TBW count?

[JorgeB]

It's never a good sign but if it remains stable it should be OK.

[stevep94]

8 hours ago, Michael_P said:

How old is it, and what's the TBW count?

Sorry but what is this (Total Bytes Written?) and how would I find it out? (although I've just done a search in my e-mails and the only Samsung 830 SSD that pops up, I bought in 2012! 😲 - had no idea it was almost 8 years old!!)

 

I've downloaded the SMART report and get the following:

 

ID# ATTRIBUTE_NAME          FLAGS    VALUE WORST THRESH FAIL RAW_VALUE
  5 Reallocated_Sector_Ct   PO--CK   099   099   010    -    4096
  9 Power_On_Hours          -O--CK   094   094   000    -    27785
 12 Power_Cycle_Count       -O--CK   097   097   000    -    2148
177 Wear_Leveling_Count     PO--C-   079   079   000    -    734
179 Used_Rsvd_Blk_Cnt_Tot   PO--C-   099   099   010    -    2
181 Program_Fail_Cnt_Total  -O--CK   099   099   010    -    1
182 Erase_Fail_Count_Total  -O--CK   100   100   010    -    0
183 Runtime_Bad_Block       PO--C-   099   099   010    -    1
187 Uncorrectable_Error_Cnt -O--CK   100   100   000    -    0
190 Airflow_Temperature_Cel -O--CK   071   056   000    -    29
195 ECC_Error_Rate          -O-RC-   200   200   000    -    0
199 CRC_Error_Count         -OSRCK   253   253   000    -    0
235 POR_Recovery_Count      -O--C-   099   099   000    -    63
241 Total_LBAs_Written      -O--CK   099   099   000    -    109794768043

 

Not sure if this is any use at all??

 

As an aside I've noticed this morning that my CPU usage is showing as 100% as shown in the following pic - one minute it was CPU3 in the left column and then next it has changed to CPU1 in the right hand column?? (I've not noticed any of the CPU cores registering 100% before?)

Edited June 9, 2020 by stevep94

[Michael_P]

Total LBAs Written looks like around ~50 TBW

 

For a 250G drive you're nearing EOL - if it's a lower capacity you've exceeded it's expected life, so I'd consider a replacement.

[stevep94]

14 minutes ago, Michael_P said:

Total LBAs Written looks like around ~50 TBW

 

For a 250G drive you're nearing EOL - if it's a lower capacity you've exceeded it's expected life, so I'd consider a replacement.

Thanks! If you'd put a gun to my head I would never have said it was 8 years old! 😲

 

It's a 256GB drive so I'm going to look for a newer replacement I think!

 

Can anyone point me towards a guide to replacing the cache drive? Although, I actually don't use it as a cache drive to be honest, it's just set up as the home for my dockers (I think!) - as it goes I'm only using 10% of the drive capacity so assume I could get away with a far smaller drive?

 

I never really understood the theory behind having a cache drive, whenever I need to copy anything to my array I just do it straight to the HDD's as speed of copying isn't a massive issue for me - are there any other benefits of having a cache drive??

 

Thanks again for your continued help!

[JorgeB]

28 minutes ago, stevep94 said:

Can anyone point me towards a guide to replacing the cache drive?

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?tab=comments#comment-511923

 

[Michael_P]

1 hour ago, stevep94 said:

 as it goes I'm only using 10% of the drive capacity so assume I could get away with a far smaller drive?

Just keep in mind, larger capacity drives have a higher expected TBW lifetime - should you decide to tax it a bit more

 

Your 1 core 100% is probably WSSD going wild, you can verify with the 'top' command

[stevep94]

5 hours ago, Michael_P said:

Just keep in mind, larger capacity drives have a higher expected TBW lifetime - should you decide to tax it a bit more

 

Your 1 core 100% is probably WSSD going wild, you can verify with the 'top' command

Sorry (again) for being thick but what is "WSSD" and how to I verify with the "top" command? (and verify what exactly?) - it's still doing it by the way, but right now it's the 2nd column on CPU3 for some reason!)

 

Also, is there any real reason to have a cache drive if I'm not actually using it as a cache drive? The reason I ask is that typically speed of writing to the array isn't a huge issue for me - there are occasions whereby I've downloaded something and need to copy it to a different share to access it but my understanding of a cache drive is that copying from one share to another there wouldn't be any benefit anyway would there? (isn't data just held on the faster drive and then moved when invoking the mover?)

 

At the moment the cache drive is sat there with just my app data (basically dockers as I don't have any VM's running) on but would there be any detriment speed-wise if that information was on one of the HDD's?

Edited June 9, 2020 by stevep94

[Michael_P]

Log into the command line and run the command: top

it will list all the processes running, by order of which is consuming the most CPU - press Q to quit

 

You don't to use it as a cache drive, but it is recommended for dockers and VMs for the speed and responsiveness.

[stevep94]

1 hour ago, Michael_P said:

Log into the command line and run the command: top

it will list all the processes running, by order of which is consuming the most CPU - press Q to quit

 

You don't to use it as a cache drive, but it is recommended for dockers and VMs for the speed and responsiveness.

I'll source another SSD if it's going to help responsiveness!

 

Regards the CPU usage, you really have to understand how low my knowledge base is here - I'm very much a GUI person, I wouldn't know how to even log into the command line? Do you mean physically use a keyboard attached to the server or is this something I can do within the UI?

[Michael_P]

Click this in the top right to get to the terminal

 

 

From there, type: top

Note what's at the top of the list, then press Q to quit, then u can close the terminal with: exit

 

 

[stevep94]

21 minutes ago, Michael_P said:

Click this in the top right to get to the terminal

 

 

From there, type: top

Note what's at the top of the list, then press Q to quit, then u can close the terminal with: exit

 

 

Ta (again)

 

Ran the command and it does indeed look like WSDD - whatever that is? Is it something I need to be worrying about or stopping somehow?

[Michael_P]

You can disable it in SMB settings, discovery will stop working for Windows 10 clients tho (the server won't show up under network, you can still reach it by name/IP)

[stevep94]

6 minutes ago, Michael_P said:

You can disable it in SMB settings, discovery will stop working for Windows 10 clients tho (the server won't show up under network, you can still reach it by name/IP)

 

Thanks - I do have a drive letter mapped to the server and find it quite useful - after my post I saw someone recommend shutting down the server and rebooting - I've done this and it isn't showing as 100% use any more (although I appreciate this isn't a fix!)

 

Is there any detriment to just leaving it alone?

 

Also, while I've been dithering about which SSD to buy to replace my cache drive, I've realised that I actually have a 250GB Crucial MX500 that I bought in May 2019 that is sat in a laptop that I don't use anymore!

 

Easy enough to whip it out and slot it in place of the Samsung but do I need to prepare it in any way or will unRAID take care of that when I put it in? (it's got Windows 10 on it at the moment!)

Edited June 9, 2020 by stevep94

[Michael_P]

Access to the server isn't affected at all, drive maps and all that will still work - just when you go to to network it won't show up in the list. If you type the name in the address bar you can browse to it as normal. It's an inconvenience thing, really.

 

Restarting the server will restart the WSDD service, but eventually it will bug out again. Best bet is to pin the server under favorites in explorer in your Windows 10 client and just disable WSDD.

 

Here's the procedure for replacing the cache drive: https://wiki.unraid.net/Replace_A_Cache_Drive

 

 

[stevep94]

 

15 minutes ago, Michael_P said:

Here's the procedure for replacing the cache drive: https://wiki.unraid.net/Replace_A_Cache_Drive

 

How long should the mover be taking for the 3 folders I've got on the cache drive? I followed the instructions above but the mover has been going for a good 10 minutes now and there is no indication of how long it should be taking?

 

*edit*

 

I've just looked and my "appdata" and "domains" folders are on the array but the "system" folder isn't yet - although looking in the folder my "docker.img" is 21.5GB - presumably this is why it's taking a while? (also, is this normal size-wise?)

Edited June 9, 2020 by stevep94

[stevep94]

Never mind, it's done now!

[stevep94]

OK, things seem to be going as planned but here's a thought, what file system type do I want for the cache drive?

 

I've clicked on it after assigning it and it looks to have defaulted to "btrfs" but when I did a search there were various comments saying that it is better being "xfs" (although these posts were a few years old so I assume things may have changed?) and like a complete tool, I forgot to check what file system type my previous drive was!

[Michael_P]

btrfs is how I have mine

[stevep94]

Wow, looks like everything has actually worked! 😃

 

Thank you so much @Michael_P and @johnnie.black for your time in replying - this is one of the best things about Unraid - the community is amazing so even someone with as little knowledge as me can get things right!