LTT Storage Rankings

[Brandonberg]

But even after I uninstall and re-install plex I have issues. 

 

Not that it really matters too much....the only reason I'd use Plex is to stream over a VPN since I'm having serious issues trying to stream normally over a VPN and for some reason I keep running into firewall issues when trying to access plex over the VPN even though there are no firewalls enabled. 

 

Im no expert on FreeNAS or plex but in my experience it is still a very buggy system. Now i have it working, i wont be touching anything... ever again

 

I did alot of troubleshooting online to try and get Plex to install properly. I came accross the following which worked for me in one instance but may or may not work for you;;

 

When you delete the Plex plugin (assuming this is the only addon you are running), go to the Storage Manager and delete all of the jail template datasets so they are recreated when you reinstall Plex (you may need to restart the server to unmount the datasets so you can delete it). This worked for me after i failed to successfully install Plex about 5 times. 

 

Given how FreeNAS is temperamental with plugins, it could be like 1 of 1000 different things that is going wrong though =/

[79wjd]

Im no expert on FreeNAS or plex but in my experience it is still a very buggy system. Now i have it working, i wont be touching anything... ever again

 

I did alot of troubleshooting online to try and get Plex to install properly. I came accross the following which worked for me in one instance but may or may not work for you;;

 

When you delete the Plex plugin (assuming this is the only addon you are running), go to the Storage Manager and delete all of the jail template datasets so they are recreated when you reinstall Plex (you may need to restart the server to unmount the datasets so you can delete it). This worked for me after i failed to successfully install Plex about 5 times. 

 

Given how FreeNAS is temperamental with plugins, it could be like 1 of 1000 different things that is going wrong though =/

Well, I tried that hopefully it will work without any issues now

 

On a slightly related note....I'm tempted to throw in a 60gb hdd I have lying around for the sole purpose of handling the jail dataset...I don't like seeing it every time I open up one of the drives to access files. 

[d3sl91]

ZedNAS

 

 

Hardware

 

CASE: Fractal Design Node 304 (Black)

PSU: Corsair HX750

MB: ASRock Rack E3C224D2I

CPU: Intel Xeon E3-1231V3

COOLER: Noctua NH-L9i

RAM: 16GB Kingston KVR16E11/8I 1600Mhz DDR3 Unbuffered ECC (2x8GB)

BOOT: Fully sick 8GB Sandisk thumb drive

HDD: 6 x 4TB HGST Deskstar NAS Drive (Total Drive Capacity: 24TB)

 

Software & Configuration

 

FreeNAS 9.3 running ZFS. Raid-Z2 resulting in total pool size of 14TiB or ~15.4TB. Using Plex media server plugin and still experimenting with other plugins at the moment.

 

Usage

 

Home storage server for personal files, movies, other media, backup for parents business PC, photos and girlfriends iPhone photo backup (which almost needs its own raid.. seriously she is obsessed..)

 

Backup

 

I have an external 2TB Drive and two 1TB Drives which important stuff is backed up to. Also have a 1TB cloud storage account for important files that need to be accessed on the go.

 

Additional Info

 

The NAS uses about 85-90W when idle, 100-110W when under full network transfer to RAID and 120-130W when transcoding media via Plex to 3 different devices at once. The Xeon is probably overkill but opted for a Xeon over my previously planned Avoton C2750D4i due to similar pricing and sourcing of components in Australia. I also opted for the 1231 Xeon as it offers Hyperthreading for only a few dollars more (Again in AUS). The HX750 PSU is way more power than i need but i had this lying around so decided to put it to use. I may re purpose this PSU for another build in the future in which case i would probably run the NAS on a Seasonic or Corsair 400-500W PSU.

 

Photos   (More photos below in spoiler tag)

 

Duff Beer added for extra Awesome (and size reference)

 

2015-09-01 11.05.12.jpg

 

2015-09-01 09.06.37.jpg

 

 

 

 

Windows7.png

 

FreeNAS.png

 

2015-09-01 10.54.03.jpg

 

2015-09-01 09.07.07.jpg

 

2015-09-01 09.06.54.jpg

 

2015-09-01 09.07.29.jpg

 

2015-09-01 09.07.35.jpg

 

How do you like the 304 with that many drives? How is heat/sound?

 

I am using an older "silent" antec case... but cooling is horrible and I am trying to downsize.

[alpenwasser]

Thanks, Can it count if it's used as a media server for the other 4 people in the house and for streaming to my HTPC?

Will have another 8TB drive by the end of the year.

 

Oh, sorry, forgot about you. Yeah, I think we can allow it.

 

- snip -

Nice box of terabytes you got there, added to list.

[MG2R]

 Nice box of terabytes you got there, added to list.

 

> Box of terabytes

> BOT

 

 

*runs*

[AnonymousGuy]

#5 reporting in.  Seagate 3TB drives are complete shit.  They're literally dying faster than I can even warranty replace them.  I have a stack of 4 of them dead on me right now and have had to shut down my server because my Raid 60 array is now critical where 1 more drive failure on one of the raid 6 sections (raid 60 = raid 6 + raid 6 in raid 0) will cause data loss.

 

I'm taking a "fuck this noise" approach and replacing them now as they die with 3TB HGST Ultrastar drives.  2M hour MTBF enterprise drive and HGST is what Backblaze found is reliable as the sun.  Not cheap but neither is data loss.

[79wjd]

#5 reporting in.  Seagate 3TB drives are complete shit.  They're literally dying faster than I can even warranty replace them.  I have a stack of 4 of them dead on me right now and have had to shut down my server because my Raid 60 array is now critical where 1 more drive failure on one of the raid 6 sections (raid 60 = raid 6 + raid 6 in raid 0) will cause data loss.

 

I'm taking a "fuck this noise" approach and replacing them now as they die with 3TB HGST Ultrastar drives.  2M hour MTBF enterprise drive and HGST is what Backblaze found is reliable as the sun.  Not cheap but neither is data loss.

Out of idle curiosity were the 3tb drives also enterprise/NAS drives or no? 

[AnonymousGuy]

Out of idle curiosity were the 3tb drives also enterprise/NAS drives or no? 

 

 

Nope.  The regular Seagate 3TB consumer drives that Backblaze found died incredibly fast: https://www.backblaze.com/blog/best-hard-drive/

 

"ST3000DM001"

 

I was like "oh they're cheap I can just replace them".  Nope.  My raid card screaming at me that a drive has failed is starting to haunt me and I hear it everywhere....

 

I think they're incredibly sensitive to shock / vibration because it's the ones in the external cages that are dying the fastest.  My guess is the cage has it's own resonance frequency the ones in my Node 804 don't have and they're vibrating themselves to death.  Enterprise drives should have additional mechanisms to be immune to this sort of thing.

[handruin]

Nope.  The regular Seagate 3TB consumer drives that Backblaze found died incredibly fast: https://www.backblaze.com/blog/best-hard-drive/

 

"ST3000DM001"

 

I was like "oh they're cheap I can just replace them".  Nope.  My raid card screaming at me that a drive has failed is starting to haunt me and I hear it everywhere....

 

 

I bought four of these Seagate 3TB drives back when Amazon was having sales on them.  Two have died and one of them in my current workstation shows signs of sector relocation in the SMART data.  I never used them in a NAS, they've all been deployed in workstations for general use.  All the data is backed up in two locations for when it does eventually die.

[handruin]

Well, the timing on this is rather interesting.  I lost the 3TB Seagate I just mentioned in my post earlier today.  Windows took forever to boot back up even with a power cycle and now my G:\ drive is missing.  lol  Time to get a replacement that doesn't suck...

[handruin]

So...actually all four 3TB Seagate drives are now dead.  After removing the bad drive from my workstation I brought it downstairs to where my fiancee was working on her workstation and we were joking for a bit.  I asked her jokingly to check her drive and it was missing from the list.  Turns out hers died with bad blocks back in June and she never noticed it missing.  In her defense it was simply and archive drive that was full of her photos from her business so she rarely accessed it.  Both systems are now restoring from my NAS via local Crashplan client.  There is also a copy on the Crashplan cloud in case the NAS dies...eeek. :unsure:

[AnonymousGuy]

I have online backups to both Crashplan and Google Drive for Enterprise (unlimited storage instead of 1GB or whatever) but with 20Mbps upload ....fastest Cox has currently that should get upped to 30Mbps this month...I can't go 100% cloud based.

 

Personally I've never had a WD drive die on me.  But I don't have 24 WD drives so my sample size of them is more limited, but everyone I've ever talked to has preferred WD over Seagate.

[MG2R]

Well, the timing on this is rather interesting.  I lost the 3TB Seagate I just mentioned in my post earlier today.  Windows took forever to boot back up even with a power cycle and now my G:\ drive is missing.  lol  Time to get a replacement that doesn't suck...

That timing is rather interesting indeed. Seagate does know when to strike, it seems

[alpenwasser]

Personally I've never had a WD drive die on me.  But I don't have 24 WD drives so my sample size of them is more limited, but everyone I've ever talked to has preferred WD over Seagate.

I bought four WD RE4 enterprise drives in 2011 and this summer, two of them started

producing bad sectors. They're not completely dead yet, but I've replaced them anyway,

just to be on the safe side. They both have ~30,000 power-on hours on them, so I'm

not too bothered by it, these things happen after all.

More annoyingly though, I have three WD Red drives which have a ridiculously high

load/unload cycle count (as in: they're about a year old, are rated for 300,000

cycles, and all of them are now above that limit). No actual failures yet, but I'm

not happy about it I must admit. HDDs are mechanical drives, so on occasion they

will fail, but this doesn't really seem like that to me.

The rest of my WD drives are running okay though.

EDIT: As a comparison: The load/unload cycle count on one of my RE4 drives is

somewhere above 60 for 30k power-on hours, whereas one of the Reds has ~11k

power-on hours and a load/unload cycle count of ~330k. So yeah... not good.

Edited September 3, 2015 by alpenwasser

[beach_boy98]

Thanks guys. You have inspired me to build a server. Only 2 1tb in raid1 for now though but thanks.

[alpenwasser]

Thanks guys. You have inspired me to build a server. Only 2 1tb in raid1 for now though but thanks.

Excellent, that's certainly part of the purpose of this topic. Have fun!

[AnonymousGuy]

I'm now shitting bricks:

 

 

Span 0 is about to have PD 6 (aka slot 6) die, the two drives I just replaced are rebuilding, and 2 more drives decided to die in the other span, probably from being power cycled when I turned my server off.

 

I have another Seagate 3TB that arrived today from RMA so I'll swap that in to Slot 14 once the rebuild (if the...) finishes on Slots 10 and 11.  I have 2 more HGST drives on order to fill in Slot 15 and one to be on standby if another drive fails.

[alpenwasser]

I'm now shitting bricks:

 

Span 0 is about to have PD 6 (aka slot 6) die, the two drives I just replaced are rebuilding, and 2 more drives decided to die in the other span, probably from being power cycled when I turned my server off.

 

I have another Seagate 3TB that arrived today from RMA so I'll swap that in to Slot 14 once the rebuild (if the...) finishes on Slots 10 and 11.  I have 2 more HGST drives on order to fill in Slot 15 and one to be on standby if another drive fails.

Damn dude, that does not sound healthy for one's nerves at all.

I'll knock on some wood for you and cross some fingers that things don't go

completely belly-up.

[beach_boy98]

Excellent, that's certainly part of the purpose of this topic. Have fun!

Thanks. Now to not go out for like 2 months but oh well. 

[b348xyx]

This PC is my all in one Workstation. I run multiple VMware machines, Stream films to house, Storage Server, and mainly while doing all that has the power for me to work for my day to day as well without slowdown issues. Best system i have owned.

 

So this system runs 24/7 and never gets turned off unless windows updates or upgrades so on.

 

Only upgrade i want is a Nvidia Titan, want the 12gb ram it has !

 

Hardware

• CPU
  2x Hex-Core Xeon X5650 12 Core 24 Threads

• Motherboard
  HP Z800 Motherboard
• RAM
  48GB DDR3 ECC @ 1333mhz
• GPU
  ASUS OC Nvidia 960 GTX Mini
• Case
  HP Z800 Workstation
• Storage
  1x Samsung 850 EVO 500gb
  3x256GB Crucial M4 SSD Windows Storage Spaces Raid 0  
  9x2TB Westren Digital Green 7200RPM Running In Two Way Mirror Windows Storage Spaces (Icy Dock MB975SP- B)
• PSU
  1110 Watt HP Z800 PSU
• Display(s)
  3x 24inch Dell U2410 1x 24inch HP LP2475w
• Cooling
  Stock
• Keyboard
  Logitech G15
• Mouse
  Logitech G502 Proteus Core
• Sound
  Creative Sound Blaster Z PCIe (OEM Version)
• Operating System
  Windows 10 With Start Is Back !!!!

Software and Configuration:
My Workstation is running Windows 10 Pro and i am using onboard SAS and Sata Ports with Windows Storage Spaces.

 

All Bitlockered at 256bit, and performance on the 9x2tb is around 100-120mb write. which is not to bad.

My main array consists of 9x 2TB in two way mirror, so I end up with 8.17TB actual storage space on that array.

 

Second array Also Bitlockered is 3x256GB SSD's which are in raid 0 and i use these for running my VMware machines as they need the performance.

 

Windows OS is just a 500GB Samsung 850, which i have a Full OS backup which runs automatically weekly to the large array.

 

 

 

Backup:
I have a APC Smart-UPS 1000 (Incase power failure)

Photo's:

 

 

 

 

 

 

My SSD Stack and love how it looks !

 

[alpenwasser]

12 core goodness

Ah yes, I think I saw this in the systems showoff thread a few days ago. Always

great to see a fellow LGA1366 2CPU aficionado (I have two dual-CPU LGA1366 machines

myself, one with two L5630s and one with dual X5680s).

Welcome to the list!

[b348xyx]

Ah yes, I think I saw this in the systems showoff thread a few days ago. Always

great to see a fellow LGA1366 2CPU aficionado (I have two dual-CPU LGA1366 machines

myself, one with two L5630s and one with dual X5680s).

Welcome to the list!

Thanks ! I love the dual socket boards and as system is on 24/7 I wanted server grade and ECC memory. Which reason picked it.. And HP build quality is amazing.

Only thing I wish there were some options to over clock a little. As its well know these Xeon chips overclock very well.

Will be adding 10th drive most likely as I hate odd numbers ! And with windows 10 the storage spaces has option to rebalance when adding new drive which is wicked !

[alpenwasser]

Thanks ! I love the dual socket boards and as system is on 24/7 I wanted server grade and ECC memory. Which reason picked it.. And HP build quality is amazing.

Only thing I wish there were some options to over clock a little. As its well know these Xeon chips overclock very well.

That reminds me, I still need to overclock my Xeons. After all, what's the point of an

SR-2 if you're not going to overclock your Xeons?

Although I probably won't overclock them by much. The rig runs 24/7 at full load, and

I don't want to increase the power bill too much.

[Ramaddil]

Total Storage 

75.0TB

 

 

Server Name

Titan

 

Hardware

CASE: Lian Li 343B

PSU: Corsair TX850

MB: Asus TUF Sabertooth 990FX

CPU: AMD FX - 8120 @ Stock

HS: Noctua NH-D14

RAM: 32GB DDR3 1600 GSkill Ram

RAID CARD 1: Adaptec 52445 28 Port Raid Card

SSD: Samsung 840 EVO 120GB

Enclosures: Startech 2/5.25 for 3 HDD (8) 

HD's: 

15x 2TB Western Digital (Red) WD20EFRS

9x 2TB Western Digital Greens (Various Models) 

1x 3TB Western Digital WD30EZRS

4x 6TB Western Digital WD60EFRX

 

((( Update Added 7TB to Server )))

((((( Update # 2 Added 10TB More To Server ))))) 7/12/14

((((((( Update #3 Replaced a Failed 2TB//Added Additional 2TB//Changed Operating System//CPU, Motherboard Upgraded)))))))

((((((((((Update #4 Changed Raid card//Added Additional 20TB of Storage//Added 16Gb more of Ram (total 32gb) ))))))))

((((((((((((Update #5 Added an additional 2TB/Reconfiguring Fans for external Fan controllers (Pics to Follow)  ))))))))))))))

(((((((((((((((Update #6 Added an additional 6TB/Replaced MB/Scrapped outside Fan controller/Added 2U for future Expansion)))))))))))))) 

 

Software and Configuration:

My server is running Windows 8.1, I have one Raid 6 Array with 23 drives (23 X 2TB =46TB).  This array holds all my movies/tv shows and data.  I have one 3TB drive that I use for backup for my software that is synced to nightly with Syncback Pro on top of my cloud storage which is mentioned in the backup section.  Second Array is 4 x 6TB (24TB) for Backup Purposes). 

 

Usage:

I use my server as my media hub for my entire house and beyond.  I have 2 WD TV Media players in different rooms in the house which can remotely access my server.  I also utilize my server to access Air Video remotely on my Idevices, and it also serves as my security system monitoring cameras placed around the house.

 

Backup:

My Server is connected to Crashplan which stores my data off site.

 

Additional info:

Server now running 8.1 after some tweaks to add start menu and better functionality, looking forward to Windows 10!

 

 

 

Photo's:

 

 

 

 

 

 

 

Update: (((((((((((((((Update #6 Added an additional 6TB/Replaced MB/Scrapped outside Fan controller/Added 2U for future Expansion))))))))))))))

[AnonymousGuy]

We survived!  I now have 3 HGST drives, 21 Seagates, with 1 HGST spare on my desk.  4 more Seagates are still working through RMA.   I'm eyeballing adding all 4 of them as hot spares via the SFF-8088 connector, but need to consider if it's really worth the effort to do that.