WD Blue failing soon?

[Geralt]

Just ran a HD Tune benchmark on a 320GB WD Blue I got from a friend and it seems like there's a sudden drop in read speeds at 4 places. Could the drive be failing? This is currently used in my server and has chalked up 329 days of runtime.

 

 

Any help is appreciated!

[failblox]

Download CrystalDiskInfo and post the results here please so that I can take a look at them.

[Geralt]

3 minutes ago, failblox said:

Download CrystalDiskInfo and post the results here please so that I can take a look at them.

 

Here we go

[You_are_a_cunt]

A couple values seems a bit screwy to me, but it should still work for a while (I'm running drives in worse shape so yeah).

 

I SUMMON @Captain_WD!

 

[failblox]

8 minutes ago, Geralt said:

 

Here we go

Thanks! Looking at the data, your drive seems to be fine for the time being because the raw values for reallocated sectors count and current pending sector count are both 0.

[Syntaxvgm]

That might be data fragmentation- to do the test write, it had to move around a few times likely. (Also why copying a ton of small files on a mechanical hard drive is much slower than a couple large files ). Towords the end of the write, the access times approached 30ms, and for a busy hard drive that's just fine. What I'm more concerned about is the slow drop to 20MB/s

 

edit- that's read. Those gaps shouldn't be happening.Hmm 

[Geralt]

15 minutes ago, revsilverspine said:

A couple values seems a bit screwy to me, but it should still work for a while (I'm running drives in worse shape so yeah).

 

I SUMMON @Captain_WD!

 

 

14 minutes ago, failblox said:

Thanks! Looking at the data, your drive seems to be fine for the time being because the raw values for reallocated sectors count and current pending sector count are both 0.

 

One thing to note is HD Sentinel states that " Problems occurred between the communication of the disk and the host 585 times. " So I dunno if that has to do with the graph.

[Captain_WD]

1 hour ago, Geralt said:

~snip~

 

Hello

This does look like the drive may have some problems. I would check the raw values of the S.M.A.R.T. status as well as run WD Data Lifeguard Diagnostic tool and see if the drive passes both the quick and the extended tests. 

The raw values don't show anything abnormal so I wouldn't think the spikes are coming from the HDD itself. 

I would try another SATA cable and another SATA port and re-run the tests and see what are they going to show you. 

Thanks @revsilverspine for mentioning :)

Captain_WD.
 

[Geralt]

12 minutes ago, Captain_WD said:

 

Hello

This does look like the drive may have some problems. I would check the raw values of the S.M.A.R.T. status as well as run WD Data Lifeguard Diagnostic tool and see if the drive passes both the quick and the extended tests. 

The raw values don't show anything abnormal so I wouldn't think the spikes are coming from the HDD itself. 

I would try another SATA cable and another SATA port and re-run the tests and see what are they going to show you. 

Thanks @revsilverspine for mentioning :)

Captain_WD.
 

Ran both the quick and extended and they passed the test. Also swapped out the cable and even put the hard drive in another system to run the test and they produced the same drops. I did notice that the drive tends to heat up more than usual during normal usage tho. At first I thought it was the SFF of my server but after putting it in open air, it still runs in the 40C range.

 

Should I run out and get a new drive? This drive definitely has seen better days with almost 8000 hours under it's belt

[Captain_WD]

2 hours ago, Geralt said:

~snip~

This may be cause by the drive's age or a simple malfunctioning but I'd consider swapping it or at least having backups of the data on it. 
Since the drive passes both tests it should still be safe to use and the S.M.A.R.T. doesn't indicate any issues.
I'd check the drive with other benchmark tools and see if the results are similar.

Captain_WD.

[Geralt]

3 minutes ago, Captain_WD said:

This may be cause by the drive's age or a simple malfunctioning but I'd consider swapping it or at least having backups of the data on it. 
Since the drive passes both tests it should still be safe to use and the S.M.A.R.T. doesn't indicate any issues.
I'd check the drive with other benchmark tools and see if the results are similar.

Captain_WD.

Alrighty then. I got a failover server which mirrors my current server so I have a back up even if this hard drive dies.

 

So which hard drive should I go for next? A red drive?

[d3anio97]

Absolutely go with a red!! I've been using reds in my gaming rig as a shared pool and they're extremely reliable!! You shouldn't have issues with them at all! The best bit is the addition of the TLER function, which stops the disk from dropping out of the array when it encounters a read/write error or bad sector! And if I recall, they can also alter their platter speeds to conserve power when the disk isn't being heavily used... You'll have to check WD's datasheets before quoting me on that bit, but I recall reading something like that on the marketing w4^k pages...

[d3anio97]

6 minutes ago, d3anio97 said:

Absolutely go with a red!! I've been using reds in my gaming rig as a shared pool and they're extremely reliable!! You shouldn't have issues with them at all! The best bit is the addition of the TLER function, which stops the disk from dropping out of the array when it encounters a read/write error or bad sector! And if I recall, they can also alter their platter speeds to conserve power when the disk isn't being heavily used... You'll have to check WD's datasheets before quoting me on that bit, but I recall reading something like that on the marketing w4^k pages...

EDIT: The power saving function is called IntelliPower... and I was mostly right... WD's power saving disks (WD Green, WD Red, WD Enterprise capacity and WD Purple) use caching algorithms as well as altering the rotational speed (On IntelliPower ready disks, it's up to 5400RPM nominal) to conserve power and maintain their performance... 

Here's WD's KB answer about it: http://support.wdc.com/KnowledgeBase/answer.aspx?ID=3665

[dlink377]

Juy buy WD RED, it is rock solid until now. I already bought 3 x 6TB drives.

 

Just remember to use WDIDLE3 tool and set the sleep timer to 300 seconds (5 mins) to prevent very high load cycle count. I have 2 4TB RED that nearly fails because of this.

[Syntaxvgm]

20 hours ago, dlink377 said:

Juy buy WD RED, it is rock solid until now. I already bought 3 x 6TB drives.

 

Just remember to use WDIDLE3 tool and set the sleep timer to 300 seconds (5 mins) to prevent very high load cycle count. I have 2 4TB RED that nearly fails because of this.

I wonder if that's why I've had such a bad experience with reds? 

[Captain_WD]

On 23.06.2016 г. at 5:35 PM, Geralt said:

~snip~

 

WD Red is indeed a good option for a server regardless if you use it in a RAID array or not. It does feature shorter times between spindowns but as the guys mentioned you can alter that with WDIDLE3. That shouldn't be causing problems as WD Red is designed to work like this and save power while delivering good performance and reliability in server builds. 

Another good option would be WD Black as it is designed for durability and heavier and more extensive workloads. It comes with a longer (5-year) warranty and should also serve well in your build. :) It's up to you to pick one. 

Feel free to ask if there are any questions regarding any of these two drives.

Captain_WD.