Moving disks to a new server build

[Frank1940]

55 minutes ago, bjp999 said:

I am not picking on any drive in particular, but certainly looking at the reliability of drives is something everyone should do in making purchasing decisions. So many factors affect longevity, including heat, temperature differential (between temp spun down and temp in use), high vs low usage, % of time spun down, and "luck of the draw".

 

Precisely!!!   And as for real-life reliability data, by the time it becomes apparent that a particular make-and-model drive has a high-then-normal failure rate the drive will out of manufacture.  And the same is often true for drives that happens to seem more reliable than the rest of the pack.

 

Are the drives from any one manufacturer any more reliable than the rest of the pack?  That question is virtually moot today as for all practical purposes there are now only two left.

 

And don't believe that length-of-warranty has anything to do with reliability and/or quality!   The warranty period is a marketing tool and the cost of the longer warranty is usually factored into the drive's price.  (It can used the other way, too.  The warranty period can be extended without increasing the drive cost which the marketing department hopes will increase sales (and profits) in the current fiscal period at the expense of increased warranty costs in some future fiscal period.)

Edited August 7, 2017 by Frank1940

[SSD]

20 minutes ago, Frank1940 said:

 

Precisely!!!   And as for real-life reliability data, by the time it becomes apparent that a particular make-and-model drive has a high-then-normal failure rate the drive will out of manufacture.  And the same is often true for drives that happens to seem more reliable than the rest of the pack.

 

Are the drives from any one manufacturer any more reliable than the rest of the pack?  That question is virtually moot today as for all practical purposes there are now only two left.

 

And don't believe that length-of-warranty has anything to do with reliability and/or quality!   The warranty period is a marketing tool and the cost of the longer warranty is usually factored into the drive's price.  (It can used the other way, too.  The warranty period can be extended without increasing the drive cost which the marketing department hopes will increase sales (and profits) in the current fiscal period at the expense of increased warranty costs in some future fiscal period.)

 

Frequently early feedback on problematic drives occurs well before they are out of manufacture.

 

HGST is owned by WD, but from what I've seen, operates independently. Maybe there is some sharing/hand-me-down of technology but I don't tend to think of them as the same. (Case in point, the Helium WD Reds). Seagate is a 3rd completely independent company. So I see it as 3 players. And if you look at the stats, HGST has consistently had the best reliability. I think of them as the Lexus compared to the WD Toyota.

 

I agree with you on warranty. Although the shorter warranty on the externals I think has a lot to do with the unfavorable operating environment (high heat).

[JonathanM]

3 minutes ago, bjp999 said:

unfavorable operating environment (high heat).

And lack of shock protection. No matter how well padded an external drive, it will still typically see more G shock while running than a tower mounted drive. I've seen customers treat external drives quite badly. Teaching your kids about gyroscopic precession is NOT an acceptable use of an external drive. Neither is plugging it in, picking it up to see if it's running, and tossing it back on the desk like it was a paperback book. These are the same people that keep the only copy of their working documents on the external drive because they misunderstood the meaning of backup drive. It's safe there, right? That's what they make them for, so it must be safer than my laptop.

 

And people wonder why computer techs can get in a bad mood when dealing with their data loss.

[Frank1940]

32 minutes ago, bjp999 said:

I agree with you on warranty. Although the shorter warranty on the externals I think has a lot to do with the unfavorable operating environment (high heat).

 

I think it has more to do with the physical abuse that many external drives are subjected to than the high heat.  A lot of them ended up being 'tossed' in briefcases and computer bags to be transported.  Some may end up be 'permanently' connected to a computer but there is no way to determine what percentage ended up being used in either (or even both situations-- both transported and permanent connection).

 

Plus, the fact that many folks base their buying choices strictly on price.  Just look at the number of unRAID users who shucked the 8TB WD externals which voided their warranty.

 

I try to read the user reviews (both Amazon and Newegg) before purchase but that is usually a biased sample as best.  (Unhappy people are far more likely to vent than a happy customer!)  I simply expect that all hard drives will fail.  I have a precleared drive here waiting for the next failure.  I (and probably you as well) have a daily status report sent to me plus notification of the whole of the various conditions that the OS may detect in the servers' day-to-day operations.  If a drive goes off-line, I replace it and rebuilt the data first and then try to figure out if the problem was with the drive.  (My problem is deciding when to get ready to start buying large size drives so that  my drive replacement is based on drive failures rather than having to buy a drive to replace a working smaller one because I needed more storage space.)

Edited August 7, 2017 by Frank1940

[BobPhoenix]

First an explanation: I run same manufacturer on a server. I don't normally like to mix - probably OCD.  I have 2 servers with HGST only, 1 WD only and 1 Toshiba (with some old HGSTs that overheated) and two disassembled Seagate servers.

 

The worst drive I have is the Seagate ST3000DM001 drives followed by the Seagate ST4000DM000 drives both are disassembled from their servers currently.  I have 5-7 out of 17 ST3000DM001 drives in a weakened state - lots of smart errors (no pending or relocated however) and slow access in spots.  The ST4000DM000 drives were in a HP N54L so small drive number only 6 but one is bad.  That is one reason why those servers are disassembled now.  The best drives I've got are my HGST drives.  I overheated (bad fan and parity check) 12 of the HGST drives before I got my Norcos.  The temp history show several > 45, even more > 50 and two > 60.  Of those 12 only one of the > 60s is dead - well significant smart read error counts anyway.  No pending or relocated even on that one.  The others still work and the lowest temp ones are supplementing my Toshiba server.  I mostly retired them because I don't really trust them even though they have given no signs of problems (other than the one anyway). The next most reliable is my WD server followed by my Toshiba server.  So my ranking would be from high to low reliability: HGST, WD, Toshiba, Seagate.  

 

With that said every drive that has failed (except for the overheated HGST drive) was an OEM drive.  All of my HGST were in Retail packaging when purchased.  I suspect if my Seagate's had been in retail packaging they would have fared a lot better.  Maybe not as well as the HGST but probably better then the Toshiba and maybe even better than the WD.