testdasi

Dont do funky Raid, unless you are 100% sure of what u r doing or just testing a system. In your case, I would throw the 1TB in array, 500GB as cache and mount the 64GB as Unassigned Device. But that is not really how unRAID is supposed to be used, I think.

What storage do u have / plan to have? What is best depends on ur specific plan.

Cache pool can be manually configured to other modes, including RAID0/5/6 Thank you. Silly question indeed.

Silly question: how would having all the drives in the cache pool improve speed? I thought unRAID uses RAID1 btrfs which I can't think of how it would be faster. Maybe I misread something.

Unfortunately my X99 Sabertooth has the same issue (all the buses on the same controller) even after BIOS update :'( lsusb Bus Device ID Device Name Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub 003 00:14.0 USB controller: Intel Corporation C610/X99 series chipset USB xHCI Host Controller (rev 05) Bus 003 Device 002: ID 0a12:0001 Cambridge Silicon Radio, Ltd Bluetooth Dongle (HCI mode) 003 00:14.0 USB controller: Intel Corporation C610/X99 series chipset USB xHCI Host Controller (rev 05) Bus 003 Device 003: ID 05e3:0608 Genesys Logic, Inc. Hub 003 00:14.0 USB controller: Intel Corporation C610/X99 series chipset USB xHCI Host Controller (rev 05) Bus 003 Device 004: ID 8644:800b Intenso GmbG Micro Line (4GB) 003 00:14.0 USB controller: Intel Corporation C610/X99 series chipset USB xHCI Host Controller (rev 05) Bus 003 Device 005: ID 0c45:62e0 Microdia MSI Starcam Racer 003 00:14.0 USB controller: Intel Corporation C610/X99 series chipset USB xHCI Host Controller (rev 05) Bus 003 Device 006: ID 174c:2074 ASMedia Technology Inc. ASM1074 High-Speed hub 003 00:14.0 USB controller: Intel Corporation C610/X99 series chipset USB xHCI Host Controller (rev 05) Bus 003 Device 007: ID 056a:00d1 Wacom Co., Ltd Bamboo Pen & Touch (CTH-460-DE) 003 00:14.0 USB controller: Intel Corporation C610/X99 series chipset USB xHCI Host Controller (rev 05) Bus 003 Device 008: ID 046d:c517 Logitech, Inc. LX710 Cordless Desktop Laser 003 00:14.0 USB controller: Intel Corporation C610/X99 series chipset USB xHCI Host Controller (rev 05) Bus 003 Device 009: ID 10f5:0650 Turtle Beach 003 00:14.0 USB controller: Intel Corporation C610/X99 series chipset USB xHCI Host Controller (rev 05) Bus 003 Device 011: ID 1e3d:2092 Chipsbank Microelectronics Co., Ltd 003 00:14.0 USB controller: Intel Corporation C610/X99 series chipset USB xHCI Host Controller (rev 05) Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub 004 00:14.0 USB controller: Intel Corporation C610/X99 series chipset USB xHCI Host Controller (rev 05) Bus 004 Device 002: ID 05e3:0732 Genesys Logic, Inc. All-in-One Cardreader 004 00:14.0 USB controller: Intel Corporation C610/X99 series chipset USB xHCI Host Controller (rev 05) Bus 004 Device 003: ID 174c:3074 ASMedia Technology Inc. ASM1074 SuperSpeed hub 004 00:14.0 USB controller: Intel Corporation C610/X99 series chipset USB xHCI Host Controller (rev 05) Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub 001 00:1a.0 USB controller: Intel Corporation C610/X99 series chipset USB Enhanced Host Controller #2 (rev 05) Bus 001 Device 002: ID 8087:800a Intel Corp. 001 00:1a.0 USB controller: Intel Corporation C610/X99 series chipset USB Enhanced Host Controller #2 (rev 05) Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub 002 00:1d.0 USB controller: Intel Corporation C610/X99 series chipset USB Enhanced Host Controller #1 (rev 05) Bus 002 Device 002: ID 8087:8002 Intel Corp. 002 00:1d.0 USB controller: Intel Corporation C610/X99 series chipset USB Enhanced Host Controller #1 (rev 05) Bus 005 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub 005 08:00.0 USB controller: ASMedia Technology Inc. ASM1142 USB 3.1 Host Controller Bus 006 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub 006 08:00.0 USB controller: ASMedia Technology Inc. ASM1142 USB 3.1 Host Controller

I read some confusion information on the forum: There's a thread which says unRAID does not have optical drive drivers and thus will never support passing through There's another thread which says it's possible using "by-id" (which I can see on my rig e.g lrwxrwxrwx 1 root root 9 Apr 22 21:40 ata-PIONEER_BD-RW_BDR-207D -> ../../sr0) but then it refers to XEN, not KVM, I think. Someone also said the only way is to passthrough optical drive is the entire sata controller. Confused.com

What did you find?

I think it's the buffer.

Turbo Write (or in 6.2.0 beta 18+ language: "reconstruct write") helps a lot. See my 2 attachments: Turbo Write gives 60% improve in speed (caveat: I'm testing unRAID in VMWare Workstation so the disks are all virtual but each disk is located on a separate 7200rpm physical HDD so performance should be representative of real life). Cache drive, in my opinion, would only work on an established array (i.e. most data is already there, so daily activity is just the "delta"). When you just started, it's likely you will need a massive amount of overnight copying. So 60% diff for me is a 5-hour difference. And it's not like the cache will stop the slowness. It just delay the slowness into the night.

There is a guide in the wiki to convert physical PC to VM which might be useful to you: http://lime-technology.com/wiki/index.php/UnRAID_6/VM_Management#Physical_to_Virtual_Machine_Conversion You can attach NTFS drive but only by using Unassigned Devices plugin (will be mounted as a network drive) or full controller pass-through. Cannot add it to the array. What I have found is formatting a drive as xfs and attaching using Unassigned Devices means it can be written to via windows easily so I'm considering only using NTFS for the OS vdisk.

This happened to me a few times without a VM. I was just trying to do some benchmark and it just hung. What I noticed is that once I increased the minimum share size to larger than the copying file (copying 8GB file, set minimum size to 10GB), the hanging seems to have stopped. Note: the drive has more than 10GB free space so it shouldn't have been caused by running out of free space.

Has anyone tested performance of the NVMe implementation in 6.2.0? I have the Intel 750 and wondering if I should make it cache or make the Samsung SM951 as cache.

I am interested in unRAID for that exact reason!!!! I'm fumbling about testing unRAID now so hopefully can figure out a way to set up Crashplan to back up onto a special share which only Crashplan can access - in that way, even if the ransomware trawl the network locations, it won't be able to get there.

Is this just Haswell or Haswell-E as well? How did you get the information in the "code" section?

No ISO files for unRAID . I think he's talking about ISO's for VM's. Also sure you could do that, or you could put them on some other storage device and use Unassigned Devices to load them. Awesome thanks! I trust tested it out by running 6.2 beta trial on my VMWare Workstation. I tried to "download" the iso from the unassigned / flash drive and save it to the iso share and it worked! I'm now testing out running a Win 7 VM on unRAID on VMWare Workstation.

Can I pre-download the necessary ISOs and place it in the USB flash drive?

On the front post, I noticed this: It only refers to Linux. Will I run into problems if I use NTFS with Windows VM?

All the guides I read always refer to having to download something / copy something onto the iso share. That basically says I need a 2nd PC / laptop to do it (I think). Since I'm upgrading my current rig, is there any way to get around that?

Many thanks - it will be pricey but at least I can save £10 having to upgrade. The card reader has MS, Micro MS, SD, Micro SD, CF, + an eSATA so total 7 slots! I guess it means it alone would count as 7 devices (7 drives!)

I'm considering adding a few parts to my current rig to build a server and have some questions. Please would someone help answer some/any of my questions. I have done a lot of reading (following unRAID since 2015!) but I would really love to get some more specific advice. Hardware (+ means new addition) CPU: i7 5820K Motherboard: Asus X99 Sabertooth Memory 32GB GPU 1: MSI R9 390 for the main Windows VM - this massive monster covers 2.5 slots ! GPU 2: + Zotac GeFo?rce GT 610? 512MB PCI?-Express x?1 for a Ubuntu VM => no other choice I can think of since the PCIe x1 is the only empty slot PCIe storage: Samsung SM951 AHCI M.2 512GB SSD (on the M.2 slot, current Windows boot drive), Intel 750 1.2TB NVMe 2.5" SSD: Kingston 128GB, Samsung 850 Evo 1TB, Samsung 850 EVO 2TB 7200rpm HDD: 3TB Toshiba, 4TB Hitachi, 6TB WD Black, + Hitachi He8 8TB 5900rpm HDD: 8TB Seagate Archive 7-slot 5.25 internal card reader (i.e. it shows up as 7 drives) + 1 Bluray drive The plan: Use the Hitachi He8 as parity (for maximum future expandability) The rest of the 7200rpm in the array, total 13TB Samsung 1TB as cache Ubuntu VM will have exclusive use of the Kingston 128GB SSD (boot) + Seagate Archive 8TB Windows VM will have exclusive use of the Samsung SM951 (boot) + Intel 750 + Samsung 850 EVO 2TB Questions: [*]Is the above plan realistic? How would you rate the level of difficulty (I'm comfortable with command lines and following instructions)? [*]Any potential issue you think I might run into? Any additional reading you recommend before I start? [*]Which license do I need to buy? It's not very clear on the website whether passthrough / outside array devices would count toward the license device limit. If everything counts then I would need a Pro license - if only array devices count then I only need a Basic license. [*]For the M.2 and PCIe SSD, would you recommend me to go on the PCIe-passthrough route or the outside-of-array route? I think I read on the forum that the latter is more recommended but then it wouldn't work with NVMe and I'm not sure if M.2 PCIe slot (the PCIe 3.0 x4 slot is disabled if the M.2 drive is installed) can be passed-through. [*]Do I need to get 6.2.0 beta 18+ in order to have reconstruct write? The forum posts I read seem to say it could be done as far back as 2014/2015 but then it is a feature of 6.2.0 beta 18 (March 2016) so I'm not too sure. I must misread something somewhere.