N7700 with WD10EADS

[lexos]

Hi,


Just a feedback. I was quiet afraid to buy seagate drives. Lot of friends bought some (for their PC's) and they still have issue (freeze, restart)...It seems that Seagate is no longer good....

Anyway I bought seven new western digital (1TB Raw), they are recognized with a capacity of 983GB which is not bad.

All the drives are running fine, no issues, they are so quiet..amazing. I have some plans to replace the two Thecus fans with Noctua models...so after that everything will be perfect.

The N7700 is still under formatting process...could we not calculate how many hours it needs?

6 drives Raid5 with one spare (so 7 disks total).....still formatting after 26 hours...guess that will take another couple of hours...I saw around 40 hours for a raid6....


I'm interested as well upgrading the memory from 1GB to 4GB. I saw that someone did it with no issues. Maybe updating the CPU will speed up the process (and the rebuild if needed).


Regards

Vincent

[Pskone]

Are there performance benefits for running a 6 disk raid 5 with 1 spare rather than a 7 disk raid 6?

Raid 5 if you have a disk fail it rebuilds the spare into the array but during that time a second fail causes a damaged array. In the raid 6 scenario if you have a drive fail the second redundant drive is already in the array so you don't have that risk.

[leobg]

I have N8800 here, but I believe hardware is same as N7700.

Regarding build times - here is what I found: no matter how many drives are involved, RAID build always takes same time.

In my case for RAID5 spread on 8 drives WD 2TB it takes 640 minutes which is like 11 hours, file system - XFS (haven't tried with others).
RAID5 with 7 drives x 2TB - exactly the sam time. I can confirm same with 6, 5, 4 and 3 drives too.

RAID5 with 8 Seagates ES.2 1TB take 310 minutes or like 5 hours (1/2 of the 2TB)

BTW - I am experiencing some serious issues with my N8800 - drives drop from RAID randomly! I thought I got bad batch of 2TB Western Digitals after losing 4 of the 8 drives, often not been able to complete initial RAID rebuild. I had some spare ES.2 1TB seagates and decided to test with them to rule out the 'bad batch WD'. Unfortunately my N8800 dropped one of the seagates out again! Tested the Seagate with SeaTools - no issues (same about WD btw - tested with WD Diag - no issues).
I guess I got bad N8800. Very rough initial experience! Seriously started regretting my decision to go with Thecus...

[Pskone]

You know what, I'm having a similar experience on the N7700 with the 1.5TB WD green drives. I got 7 and had one DOA, which was confirmed by the WD tools run from a desktop on the drive. So I built a 6 drive Raid 6, 2 weeks later I had a drive go bad, rebooted, had another drop, rebooted and had a 3rd drop damaging the array. Factory reset of N7700, only the first one was showing bad and the 2nd and 3rd to fail were all reading good. Array was back to degraded state (3rd drive back inside), rebuilt with the 2nd dropped drive and went fine. Tested the first drive on desktop and it is in fact showing bad by WD tools. Pretty sure there's something wrong with either N7700 firmware or hardware.

Look around the forum, a user named Manni had a very similar experience to you. Randonmly dropped drives, test fine using WD tools, eventually Thecus took a look at his config.bin and said they think the controller is faulty and to get it replaced under RMA. I currently have Thecus looking at my config.bin, if I were you I'd do the same. There is no excuse for a raid to drop flawless drives, and it's extremely dangerous that a drive failure will cause multiple other drive drops since it defeats the whole point behind raid.

[leobg]

Thanks for chiming in! Yes - I actually PM-ed Manni asking about how is it going with his. Haven't heard from him back yet. I have my sales guy help me connect with Thecus support, they will be calling me tomorrow. If no resolution on the issue, I plan to return it and eventually replace with another similar product from the competition.
I plan to use the NAS for my data centre online recurring backups and having unreliable machine for this purpose is simply not acceptable. I really hope Thecus support will kick in and do the necessary to bring the trust in the device back!

[lexos]

So finally ...why it was taking so many hours to format the raid array.....one disk had lot of "pending sectors - around 130 - Smart reported". So I stopped the process and restart with only 6 drives...by the way as I have got time will do some tests with Raid6. For now build is 40% completed.

Raid6 versus raid5...well seems that raid6 will use about 30% more cpu ressources than raid5 and of course less capacity.

Will post feedback when it's completed.

[Manni]

Thanks for chiming in! Yes - I actually PM-ed Manni asking about how is it going with his. Haven't heard from him back yet. I have my sales guy help me connect with Thecus support, they will be calling me tomorrow. If no resolution on the issue, I plan to return it and eventually replace with another similar product from the competition.
I plan to use the NAS for my data centre online recurring backups and having unreliable machine for this purpose is simply not acceptable. I really hope Thecus support will kick in and do the necessary to bring the trust in the device back!

Hi Leobg, I've been away for a week and just found your message. I've updated the thread here if you want to hear about the replacement... http://www.thecusforum.eu/showthread...ted=1#post6235.