Green vs Red redux.

[HellDiverUK]

So, I see WD have released their new 5TB and 6TB Green drives. I'm still convinced these drives are the same mechanically as the Reds, and a few minutes with WDIDLE3.exe turns off/down the idle timer. In a pair in RAID1 I don't see any reason why I should spend £20-25 extra on the Red.

I've had decent performance and reliability from the Greens, basically as good as the Reds (both perfect).

Anyone else feel the same way?

 

[Chuckleb]

I think the new NASware has some vibration compensation logic in the reds. It depends on how many you have in your stack and other conditions.

 

[Mike]

I'm with you. No one can say anything about either one of them being better or more reliable, unless they have a couple thousand of them.
I've had issues with the headparking and some sas-expander chips though, so i've disabled it like so many others.

 

[MiniKnight]

It may not be scientific, but green failure rates are much higher than reds for us.

4TB Red - 3/251 drives in the first six months
4TB Green - 17/212 in the first six months

Sure there's warranty but the gap is too big. I remember at 2TB this was similar.

 

[HellDiverUK]

It may not be scientific, but green failure rates are much higher than reds for us.

4TB Red - 3/251 drives in the first six months
4TB Green - 17/212 in the first six months

Sure there's warranty but the gap is too big. I remember at 2TB this was similar.

Was the idle timer turned off on the Greens? I think that's the key to a long living Green.

 

[MiniKnight]

Was the idle timer turned off on the Greens? I think that's the key to a long living Green.

It was not.

 

[RTM]

I think the new NASware has some vibration compensation logic in the reds. It depends on how many you have in your stack and other conditions.

I believe the other reason why people buy the RED drives, is that the firmware supports TLER.
Which as far as I know limit problems with disks falling out of RAID, due to disks not responding to RAID controller because of internal recovery process.

 

[HellDiverUK]

TLER doesn't help unless you're using hardware RAID, which in the case of commercial NAS units is no-one. Any commercial NAS I've used has been using good old linux mdraid which will kick a drive if it's got TLER or not.

See: How to use "desktop" drives in RAID without TLER/ERC/CCTL - Home Storage and Computer Hardware - Storage Forums


The responses I received from Synology, QNAP, NETGEAR and Buffalo all indicated that their NAS RAID controllers don't depend on or even listen to TLER, CCTL, ERC or any other similar error recovery signal from their drives. Instead, their software RAID controllers have their own criteria for drive timeouts, retries and when a drive is finally marked bad.

So, basically, another reason why Reds are kinda pointless compared to Greens in the devices the Reds are marketed and "optimised" for.

 

[mervincm]

I built an XPenology NAS in an old upgraded intel SS4200NAS using 4 x 2TB Seagate Barracuda Greens. I have not found a way to disable the idle timer like with a WD green series. Any suggestions? To be honest, I pretty much only turn it on for a few days every week or two (2nd copy offline backup) but I do want to squeeze as much life as I can out of them.

 

[capn_pineapple]

I get reds instead of greens for purely superstitious reasons... I've had 6 green drives fail within 2 years, yet to have a red fail at all...

 

[HellDiverUK]

I have not found a way to disable the idle timer like with a WD green series.

There's a boot CD available with DR-DOS and wdidle3.exe on it, just boot that and disable the timers.

 

[mervincm]

That's for WD green, not Seagate green.

 

[HellDiverUK]

That's for WD green, not Seagate green.

Who mentioned Seagate Greens, apart from you?

This is a thread about WD Greens vs WD Reds. Nothing else. Derp.

 

[mrkrad]

You have to remember with TLER=7 or TLER=0 (AV) - you don't have long hangs on your storage subsystem when the drive hangs for 100 seconds holding up all the applications which are trying to read/write to the drive! It is more of an application layer failure since SQL SERVER or whatever you are running on the storage will balk at such high timeouts and disconnect the share with high latency. Why AV drives run TLER=0 - they would rather have a corrupted sector than video loss due to buffer overrun!

 

[HellDiverUK]

Well, I eventually ended up with a pair of 5TB Reds. Both are spewing bad sectors with less than a week running time.

So, much data shuffling required and I'll get the 4TB HGST drives back in to service until HGST bring out their (cheap) 6TB drives.

 

[Patrick]

Well, I eventually ended up with a pair of 5TB Reds. Both are spewing bad sectors with less than a week running time.

So, much data shuffling required and I'll get the 4TB HGST drives back in to service until HGST bring out their (cheap) 6TB drives.

Wow! Where did you get them from? Packaging/ shipping?

 

[HellDiverUK]

Scan.co.uk, who are usually OK. Each drive was wrapped in a good 2" thick ball of bubblewrap, in a box with bubblewrap in the bottom of it. The 500GB Blue that came in the same box is running fine. I suspect the damage may have been done pre-shipping, because the shipping looked perfect.

RMA logged, syncing data back to my trusty Hitachis at the moment.

 

[HellDiverUK]

So, currently I have a full backup on the 2x4TB HGST. The 2x5TB Reds are still showing as ill. I've ordered a pair of 5TB Toshiba (AKA Hitachi) MC04ACA500E drives.

Once they arrive they'll replace the dead/dying Reds. I will probably then replace the 2x500GB Blues with the 2x4TB HGST drives, and hopefully with 9TB mirrored space that'll do me for a while, and I can safely ignore the NAS for a few years.

At least, that's the current plan.

 

[HellDiverUK]

14/10/2014 PM 06:04 ERROR SYSTEM [Disk] There is a media error on sector 1107888400 of the slot 3 disk!
13/10/2014 AM 08:05 ERROR SYSTEM [Disk] There is a media error on sector 2986871144 of the slot 3 disk!
13/10/2014 AM 04:33 ERROR SYSTEM [Disk] There is a media error on sector 1459432 of the slot 4 disk!

Oh dear. Disk 3 is starting to make some terrible noises.

 

[HellDiverUK]

The plot thickens. The Reds are out of the Asustor. I've run a full test on both with WD's Data Lifeguard which they passed. I have the two drives in a Synology DS214+ and so far so good.

The Asustor has a pair of shiny new Toshiba MC04ACA500E drives (5TB Cloud Storage) which are very fast, but are incredibly loud during seeks. Too loud. The Asustor with 4x 7200rpm Enterprise drives in it (the 2x Toshibas and 2x HGST) is using nearly 60W of power.

So, if the Reds behave, I'll probably keep them in the Synology and sell the Asustor and the Toshiba drives. The Synology uses a third the power, runs just as quickly, and is much quieter.