Large >1TB drives in GOS unit

[Max8]

great feedback, many thanks... looking forward to others take on this subject..

[Max8]

a 10x 2TB HDD Review;

http://www.engadget.com/2010/04/16/2...-wd-throw-dow/

2TB hard drive review roundup: Samsung, Seagate and WD throw down
By Darren Murph posted Apr 16th 2010 1:19PM

Its ashame that Hitachi was not present...

[HarryC]

It's been 10 months now with four 550's 4.3.007 code and one 520 all running WD caviar black drives, The FALS model. All running Iscsi and/or NFS with VMware servers, probably about 80 virtual machines and so far ZERO failures or problems. we also have five HP arrays running the Black dirves without problems. Maybe we're just lucky. I personally think a lot of the raid problems with particular drives may be the controllers rather than the drives.
Tried an 18000 with the 1Tb black drive and all was fine. Then tried some 1.5Tb seagates in the 18000, It recognizes them loads the boot to them and will boot from them. They appear as the correct drives but they don't appear in the list of drives you can use in a raid 5 array. I'll update when I get a chance to look into it

[Phoenix32]

I have been reading a lot on this subject. It seems hit and miss about 50-50 with people using the Non RE WD drives in RAID 5 arrays.

I myself am batting about 50-50 also.

I have/had a set of WD Non RE (not green either) drives that would not works in RAID 5 on any machine I had, and this means not just snaps etc. Also meaning a bunch of different controllers. They tested fine and would work solo etc just fine, but try to RAID 5 them and all would go to hell quickly. Other drives in same machines had no problems.

But, I also have a set of WD Non RE drives right now that work fine in RAID 5 in a SNAP 4400 but will puke in RAID 5 in any SNAP 4200/4500/15000 I put them in.

From my reading, various theories, etc, the consencous seems to be that WD changed their firmware a while back and this newer firmware causes problems in TLER and is why they will fail out and have issues with RAID 5, but the RE drives do not have this same firmware problem because their firmware handles TLER differently than the Non RE drives do. Its all about timing issues.

This seems to be confirmed in a round about way by the fact there used to be a downloadable firmware patch on the WD web site that corrected TLER issues in RAID arrays. But, they stopped updating the firmware and it doesn't work on the newer drives. That, and the fact they removed it from their site last year.

When people complain to WD about it, the WD reply is that they are not drives certified for RAID and thus do not go any further than telling you to buy their premium RE drives at a very premium price.

Is this a conspiracy and intentional by WD, that is for each person to figure out for themselves on what they think. Who knows for sure.

Bottom line is, they work for some, they don't for others, and as of now, WD has no intention of addressing it.

50-50, hit and miss, there is no one right answer. i.e. they do or don't work. At least not from my experiences and my research on the subject.

[HarryC]

I guess I would agree I've had perfect success with the WD "FALS" 1tb models so I've stuck with them. Unfortunately they have been discontinued in favor of the models that you can't patch the TLER. I haven't tried the other models. The "Black FALS" have worked perfectly in 550's 520's and 18000's as well as Promise 14 disk arrays and HP arrays. On the other hand we've got IBM x3650's that will only run with specific models of Seagate and Hitachi drives. The controller is IBM serverraid (Adaptec). IBM finally determined that it was a timing problem between the controller and backplane, they issued firmware fixes for the controller and backplane. Basically it locked the drive in sata 1.5gb mode. I wonder if thats the same problem some drives have in Snap Servers?

[Phoenix32]

Originally Posted by HarryC

Unfortunately they have been discontinued in favor of the models that you can't patch the TLER.

E X A C T L Y !

sadly

[achapman]

I am using Seagate 1Tb ES.2 drives in a Synology unit. After a disk upgrade this crashed catastrophically. From the Synology forums it seems that a significant number of people are having problems with 1Tb plus drives in certain Synology units. Seagate have released new firmware for the ES.2 drives seemingly because of problems with these drives in Raid setups. WD users have also reported problems with certain WD drives in Synology units.

In short it seems that it isn't just with GOS units. There seems to be general issues with large drives and raid arrays and the manufacturers of Raid systems and disk drives seem to still be working through these. These problems aren't down to hardware issues, but seem due to software issues and issues that cause controllers to think a disk is faulty when it is not.

[HarryC]

It's been quite a while but in the last three years with WD black drives running in raid 5/6 arrays 40 drives no failures. Seagates (no I don't know the models ) 8 drives with 3 failures. IBM's 2.5" sas server drives 24 drives with 3 failures. HP 2.5" sas server drives 24 drives with 2 failures. On the other hand WD green drives die regularily in desktops that came equiped with them. As I said I guess I'm just lucky with the black drives.

[eschw95458]

It would be my suggestion to HarryC and anyone using the WD green drives to turn off the intellipark feature on these drives. It seems to kill them quick. Search the web for wdidle3.exe.

[SaFeHeX]

Originally Posted by eschw95458

It would be my suggestion to HarryC and anyone using the WD green drives to turn off the intellipark feature on these drives. It seems to kill them quick. Search the web for wdidle3.exe.

Thanks for the tip!

Although i don't use any WD Green drives at this moment in time (i have a few RE4's etc) but having searched for some info on this "wdidle" issue, i found a lot of interesting and usefull info and links that will most likely come in very handy in the future.

From what i have just read (all be it in no great detail yet) it does appear to be a contributing factor to premature failure.

As much as it's nice not to waste power, i value my data and the longevity of the drives more.

Thank you for the info.

Also, my thanks to HarryC for the update on how his WD Black drives have been performing over the last couple of years. I've never been much of a fan of WD drives but i do own some. For the past few months (maybe years now, time flies), i have aquired a few more WD drives in the collection and admittedly i have had no issues with them. Mostly they have been RE versions however. I'm liking the build quality of the newer WD's more than the older ones but i have read good and bad for most drive manufactures so at the moment i'm keeping an open mind for all makes and models.

I do think it's good that you have come back and kept people updated. The proof of the pudding is always in the eating they say!