Re: preclear_disk.sh - a new utility to burn-in and pre-clear disks for quick add

[slowrolling]

I'm a windows guy that knows just enough to get by in linux.  Is there a command like ps or something else that I can use to check and see if preclear_disk.sh is still running?

 

I think a 2TB green drive should take ~30hours or so right?  ... I'm already 16 hours in if it is in fact still running, would be a shame to restart now.

[BRiT]

ps -ef | grep preclear

 

[slowrolling]

here is the output

 

root    6192  5657  0 02:19 pts/1    00:00:00 grep preclear

 

 

 

 

looks like I should restart then?

[Joe L.]

it is not running

 

Sorry.

[squirrellydw]

Hey Joe, do you still plan to combine this with unmenu?  I saw this posted on the first page

[Joe L.]

Hey Joe, do you still plan to combine this with unmenu?  I saw this posted on the first page

At some point, but have not as of yet.

[squirrellydw]

Anyway to stop the preclear once it's started?  I would rather do something else with the server right now

[Joe L.]

Anyway to stop the preclear once it's started?  I would rather do something else with the server right now

Control-C

 

 

[cj0r]

You should install the screen pkg that way you can have a preclear going on in the background while you do other things (assuming you're going through a telnet sessions and excluding rebooting).

[chanders]

Do brand new drives need to be precleared ? (I did them anyways). I ask this because a coworker was wondering if to do the same (he is now setting up his first array). His drives have been used before and was thinking to just let unRAID format them.

 

Looking forward to your replies

[jazzysmooth]

The advantage to preclearing is it allows you to perform the formatting outside of Unraid so your array remains online for that potentially extended period of time.  In addition you get SMART reports on the drives which will show you whether your drive has potential problems before ever trusting your data to it.

[Rajahal]

Do brand new drives need to be precleared ? (I did them anyways). I ask this because a coworker was wondering if to do the same (he is now setting up his first array). His drives have been used before and was thinking to just let unRAID format them.

 

Looking forward to your replies

 

It is highly recommended, yes.

[JustinAiken]

Okay, now I have all my movies/music/TV copied to my Unraid, so now I just started to build the parity drive, and I just started preclear scripts for 4 drives at once (now that the movies are done with them).

 

Will unraid be able to do all this at the same time? I cleared 3 drives at once before, but not 4, and not while parity was building... I have 2GB of RAM in now..

[jazzysmooth]

Personally, I don't run other tasks while important things like parity creation are occurring.  Can Unraid do it all at the same time?  Sure, but if you do hit a memory limit, or max out your PSU you chance interfering with the parity calculation.  And since parity protection is a primary reason why we use Unraid - I'd let the complete before doing the preclears.

[cal87]

I accidentally disconnected my network half way through the post-read.  Is there a way to restart the program at the post-read stage to see the results?  Don't want to have to start from the beginning.

[Joe L.]

I accidentally disconnected my network half way through the post-read.  Is there a way to restart the program at the post-read stage to see the results?  Don't want to have to start from the beginning.

Sorry. No.

[CrashnBrn]

I understand most of what my results are but am confused by this:

 

  3 Spin_Up_Time            0x0027  067  067  021    Pre-fail  Always      -      13683

 

This is on a new 2tb WD Black, it only took 33 hours to run.

 

Also most of my threshold values are 000, other than that I think every looks good.

 

TIA!

[BRiT]

The raw value is only meaningful to the mfg. Your worst value is your current value, which is 67. The threshold is 21. You are above that, so no need to worry. Only the mfg knows how quickly this value will decrease or what the value is brand new.

 

 

Now this following bit, I'm not completely certain about, but it's what I think, so someone correct me if I'm wrong. Even if you exceed the threshold value (drop below 021), odds are it'll still work as the status type is 'pre-fail'. The pre-fail type seems to be an indication that something may go wrong, but not a sign of a genuine failure.

[CrashnBrn]

The raw value is only meaningful to the mfg. Your worst value is your current value, which is 67. The threshold is 21. You are above that, so no need to worry. Only the mfg knows how quickly this value will decrease or what the value is brand new.

 

 

Now this following bit, I'm not completely certain about, but it's what I think, so someone correct me if I'm wrong. Even if you exceed the threshold value (drop below 021), odds are it'll still work as the status type is 'pre-fail'. The pre-fail type seems to be an indication that something may go wrong, but not a sign of a genuine failure.

 

Wow totally makes sense! I learn a little everyday! What about threshold values of 000? I'm typing (thinking) out loud That means that as the number hits zero that's the point of failure correct?

 

Edit: I missed the other thread that I was supposed to post about the results

[disco]

I'm running preclear on a new WD20EARS drive that I just received. I have the jumper placed on the 7-8 pins so the drive should be in compatibility mode.

 

The speed of the "Copying zeros to remainder of disk to clear it" is jumping around a ton--I've never seen this with a drive before. Usually my clears (on 1/2TB EADS drives) start off at around  90MB/sec and gradually slow down as the clearing progresses.

 

With the WD20EARS drive (the first of this kind that I've ever used)...the speed is jumping from around 250MB/sec (which seems TOO fast?) to around 3MB/sec.. Is this normal? Does this indicate something is wrong with the drive?

[disco]

This is what it looks like currently:

 

275114483712 bytes (275 GB) copied, 832.136 s, 331 MB/s

[Joe L.]

It probably indicates that sometimes the writes are being written to the buffer memory in the disk drive, and other times it is not. (or having to wait while the buffer is being written to the disk platters )

 

I would not worry...  It works that way with all drives, just the buffering might be more smooth.

[shawn]

[pre]

ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE

  1 Raw_Read_Error_Rate     0x000b   100   100   016    Pre-fail  Always       -       0

...

[/pre]

 

I think right now the preclear script does not have the attribute headers, so I did not understand how to read the values until I saw this post. Is it possible for this to be added to the script to help those who are unfamiliar with the format?

[Joe L.]

[pre]

ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE

  1 Raw_Read_Error_Rate     0x000b   100   100   016    Pre-fail  Always       -       0

...

[/pre]

 

I think right now the preclear script does not have the attribute headers, so I did not understand how to read the values until I saw this post. Is it possible for this to be added to the script to help those who are unfamiliar with the format?

A good suggestion.  I plan on re-writing the final output based on a lot of what we've learned since I originally wrote the pre-clear script.

 

Joe L.

[dgaschk]

Is there anyway to run preclear_disk.sh in the background? And have the results mailed

 

 

Would this work?

 

preclear_disk.sh -m [email protected] /dev/hdk > /dev/null &

 

Then I can disconnect my terminal and wait for an email.

 

What if output where redirected to a file. Could I then login and check the tail of the file for a status update?