Replacing oldest/most used drive


Recommended Posts

I am trying to upgrade some drives.  I have some aging 2-3 tb drives that I am targeting as I am out of drive slots.

 

I want to replace the best candidate.  At 2 tb an 8tb is the smartest, but only if one of the 3tb is not for some reason a better target.

 

Not sure if there is a simple report for this or not.

Link to comment

ran a diagnostic to collect a SMART report etc.

 

Looking at the reports and the power-on hours I have a few drives over 50 and 60k hours.  One that even hits 70k.

 

However, the thing that stuck out the most to me during review is the seagate drives.  Every one of my ST3000 drives show crazy values across the board.  My assumption is that it either has to be something with the drives and their firmware or the drives are really going bad, which would not actually surprise me.  I am actually looking at an ST3000 that never got used and never made it into the server before going bad.  I pre-cleared it and shelved it and now it is just bad.

tower-diagnostics-20190312-1921.zip

Link to comment

Seagate has chosen to report some of the SMART attributes in a different way compared to the other HD manufacturers. In addition they report some attributes (with id# above 200) that no other manufacturer will report. So it is not direct apples to apples comparison and the RAW values that look crazy may and in your case are just normal.

 

But if I was in your shoes I will also look at the max temperature a hard drive has run - it is usually way down in the SMART report. And in your case you have 5 HDs that were run at very high temperature at same time in the past - perhaps they shared 5in3 enclosure with a bad fan or a loose cable connection to the fan....

So here they are:

sdq - 66 degC - ~69kh

sdn - 65 - ~70.5kh

sdc - 64 - ~59kh

sdg - 59 - ~55.5kh

sdk - 58 - ~50.5kh

 

And when you replace a 2 or 3TB hard drive with 8TB one you will have spare 5-6 TB left - you can simply copy the content of 2 or 3 of these small aged drives, then zero them out and remove them from the array and you will end with more empty slots in your case (and perhaps save a tiny bit of electricity)

 

 

Link to comment
7 hours ago, bcbgboy13 said:

Seagate has chosen to report some of the SMART attributes in a different way compared to the other HD manufacturers. In addition they report some attributes (with id# above 200) that no other manufacturer will report. So it is not direct apples to apples comparison and the RAW values that look crazy may and in your case are just normal.

 

But if I was in your shoes I will also look at the max temperature a hard drive has run - it is usually way down in the SMART report. And in your case you have 5 HDs that were run at very high temperature at same time in the past - perhaps they shared 5in3 enclosure with a bad fan or a loose cable connection to the fan....

So here they are:

sdq - 66 degC - ~69kh

sdn - 65 - ~70.5kh

sdc - 64 - ~59kh

sdg - 59 - ~55.5kh

sdk - 58 - ~50.5kh

 

And when you replace a 2 or 3TB hard drive with 8TB one you will have spare 5-6 TB left - you can simply copy the content of 2 or 3 of these small aged drives, then zero them out and remove them from the array and you will end with more empty slots in your case (and perhaps save a tiny bit of electricity)

 

 

Thank you, yes, I believe that's what I called crazy, the raw values.

 

In that case, if my seagates are ok for now (knock on wood), I was going to Target sdq and sdn as they have the most hours, are 2tb and have experienced high temps.

 

As far as zeroing drives when replacing them, is that any different than the documented exclude from array/safe mode, copy off data, remove, reboot, start array and re-build parity?

 

I was thinking of removing a few drives, so I was reviewing the process a little.  First I must confirm the the supermicro backplane supports my shucked drive.

Link to comment
1 hour ago, smakovits said:

As far as zeroing drives when replacing them, is that any different than the documented exclude from array/safe mode, copy off data, remove, reboot, start array and re-build parity?

 

Not really sure what you mean by 'zeroing' a drive.  Moving everything off of the drive, so it looks 'empty', does NOT actually remove any data.  It just modifies an entry in the file system tables to show that the drive sectors, where the data and the file system entry are still stored, are now available for reuse.  That data will remain there until a new file is stored (overwriting the old data) on the drive.  If you actually remove any of these older drives from the array, you will have to rebuild Parity! 

Edited by Frank1940
Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.