uldise

thanks for clarification.. this mean, you have a small benefit when you write a lot of small files that not exceeded 64MB in total in above 4TB drive example. when you write a lot of small files more than disk Cache then this will be slow anyway original question was: "Don't tell me I picked the wrong case ". and i say again, your case will work just fine

Gary, can you explain this in more details? when this buffers can help in everyday usage when you use spinners?

you will be ok, SATA III is for SSD only now..

i think you need both type cables: - simple sata to sata cables to connect disks to motherboard ports - SFF-8087 to SATA breakout cables(two for each SAS card) to connect disks to SAS controllers Edit: More info here: http://lime-technology.com/forum/index.php?topic=20515.msg181975#msg181975 you will need Forward breakout cables, notice length also, 0.5m may be too short..

No, you will need breakout cables - one cable will split one SAS connector to 4 SATA connectors. Edit: No need for breakout cables, thanks Gary for correcting my mistake..

you can choose any of these series: http://www.supermicro.nl/products/motherboard/Xeon3000/#1150 next variant is http://www.supermicro.nl/products/motherboard/Xeon/C220/X10SLM_-F.cfm (and respective newegg link http://www.newegg.com/Product/Product.aspx?Item=N82E16813182823&cm_re=X10SLM-_-13-182-823-_-Product) this board have 3 PCIe ports, so room for GPU and two SAS cards (one with x8, one with x4 bandwith) one more option my production server worked with this (http://www.supermicro.nl/products/motherboard/Xeon/C202_C204/X9SCM-iiF.cfm) board for two years without any clues until i went for dual CPU board.. i still have this board in my test server now, and it works great.. ok, it uses second gen. Xeon, but a big advantage over X10 series is 4 PCIe slots - two x8 two x4. if you plan to use non x16 GPU then you can go this way too.. if you will choose this board, you should change CPU, RAM must fit i think..

you can choose any of these series: http://www.supermicro.nl/products/motherboard/Xeon3000/#1150 next variant is http://www.supermicro.nl/products/motherboard/Xeon/C220/X10SLM_-F.cfm (and respective newegg link http://www.newegg.com/Product/Product.aspx?Item=N82E16813182823&cm_re=X10SLM-_-13-182-823-_-Product) this board have 3 PCIe ports, so room for GPU and two SAS cards (one with x8, one with x4 bandwith)

looked at your MB again - it's wrong and won't fit with your CPU/RAM.. look at this one: http://www.supermicro.nl/products/motherboard/Xeon/C220/X10SAT.cfm

by default, cache drive is not a part of an unRAID array and is not protected by unRAID parity drive. if you wanna protected cache drive, then you can create BTRFS cache pool.. - you can use SSDs or HDDs for that. keep in mind that BTRFS is relative new file system, search forums for other's experience with that. if i were you, i would use mechanical HDD for cache and separate SSD for VM's and schedule mover to run more often than once a day for cache and create backups of VMs every day.. but you always have an option to use hardware RAID for additional protection. see you board specs, as i see, you can create RAID1 array for cache. so, decisions decisions..

18 ports is plenty.. if you wanna more, then add second SAS card, but you can add it later when your 18 ports are all full and for a cache drive - no need for SSD imho.. cos gigabit Ethernet will be a bottleneck for network reads/writes. i have two HDDs in RAID1 for Cache. if you wanna use cache drive for VMs, that's a different case. if i remember correctly, you can use separate SSD for VM's with Unassigned Devices plugin.. search forums for such config, i have no experience with this, cos my unRAID is running at top of esxi.

i'm wondering why you choose an older SAS controller, this is very similar and will provide more bandwidth to you: http://www.newegg.com/Product/Product.aspx?Item=N82E16816101792

I never professed to be a Linux guy, but i crew up on DOS, an Apple II, a TRS-80, and the orignal IBM PC's.... no fear of "the command line" just in absence of a strong Linux background I'm liable to double/triple check around here to make sure I have it before leaping.. Thanks and remember to do full power off after you apply a command above to make it permanent.. i have mine disks with small HPA cos 4TB Raid0 parity is a little bit smaller than 4TB drive. but this depends from controller to controller.

you can go with plopkexec as more faster solution than original plop.. i'm using VMDK method and it works just fine. if you wanna go esxi route keep in mind usb reset bug with unRAID6, more info and possible workarounds here: http://lime-technology.com/forum/index.php?topic=40605.0

Just added iperf to Nerd Tools. Enjoy! i'm just wondering why iPerf, it's very old tool! use iperf3 instead IMHO

just tried on my Win10, i can access unRAID with no problems.. i switched to local user after upgrading to Win10..

mine are very old too - usb but without LCD (2006 year may be), i got it used for free, replaced batteries and it just works. i have a setup with APCUPSD on Windows VM, and all values are displayed ok, Linux VMs are missing some values..

are your Smartups 750 working well with unRAID apcupsd? i have the same model, but a few variables are missing, like temperature, voltage... this maybe esxi related, i'm not sure..

Error -110 usually means too much power on USB stick - USB 2.0 stick on USB3 slot or similar..

Yes, may be but then you have to wait somewhere till the end of the year when more Linux distros(Ubuntu for example) start shipping kernel 4.

Agree, no need for xeon with integrated GPU, since X10SL7-F have one. and for PSU look there: http://lime-technology.com/forum/index.php?topic=12219.0

i think you can use an expander that don't need to be installed on PCIe slot(Intel ?) - it need only power from molex. then connect onbord SAS to expander and you have a bunch of HDDs if you need more PCIe slots, you can go with older x9SCM.. i had x9SCM-iif(with 4 PCIe) for two years on production, and it worked great, just switched to older x8 dual CPU board cos i needed more PCIe slots. see my rig.

look there: http://lime-technology.com/forum/index.php?topic=41169.0

Hi jonp, some of us running unRAID as esxi VM cant upgrade to v6 final cos this bug: http://lime-technology.com/forum/index.php?topic=40605 and don't say bare-metal only, that's not a case, esxi route is quite different than unRAID KVM route with some pros and cons of course - you cant compare them at this point now..

so, had some more time for testing today, some new observations: - since i added separate UPS to my test server, i plugged it in usb too, and added both unRAID flash and UPS to my Ubuntu test VM. and surprise, no more usb resets, leaved it for and hour, still no usb resets. then i restarted my Ubuntu VM with unRAID usb flash only and resets are back.. - next i tried unRAID v6 final VM with both usb's - booted and noticed resets started happening. then i setup UPS daemon, and voila, no more usb resets.. leaved it for an hour and half and no single usb reset. Have anyone some comments and/or ideas what to try next?

as i wrote above, my Sundisk USB2 flash works without reset errors for me too.. that's really weird if it's flash manufacturer dependent...