Search results

-> PCIe Card Slot: agreed. I will probably create room via half height or 90deg with riser. -> True: I suspect I'm going to need some more room in general in the bottom section. I don't mind making it taller, but i'm trying to keep the footprint about the same. -> mSATA would be nice but...

Very nice. Pretty much exactly what I'm looking for. Havn't found it available anywhere yet though. Amazon.com has it, but would be nice to purchase up in Canada.

Haha, possibly. At the moment i'm not too worried about that. There is 10mm clearance on each side of the drive and my fingers aren't that fat (yet). Depending on the final dimensions of the case, I might be able to push that to 15mm each side.

Power supply wise, assuming Mini ITX i'll probably go with a PicoPSU-160-XT and a custom 12v/5v 300W power supply. I'm trying to keep the cabling to a minimum here. An alternative would be some of the 300/350W MiniITX power supplised (FSP for example). I'm open to ideas from the community...

Cables are not part of the design just yet. I'm still trying to figure out if I go with backplane technology or if I go the same route as Backblaze with SATA fan out cables. Once I've decided this, i'll have a better idea on what cabling requirements I have.

I have looked at this board, and kind of designed this case with this board in mind. It's an awesome board; 4 x SFF-8087 connectors on a Mini-ITX board? Sweet! However I've subsequently discovered hardware driver compatibility issues with Illumos based distros and FreeNAS that make this board...

Interesting board, that might be exactly what i'm looking for, though it would only support 8 disks natively. I'm curious what that extra SFF-8087 port is (there appears to be 3 on that board?). Intel VT-D is not really an issue. I plan to make this a native ZFS box. I already have a...

So I've been looking for a decent NAS case for quite some time now that can push the limits on disk count. I've come across cases up to 8 drives or so, but these cases have a lot of wasted space internally, and are expensive, noisy and leave me wanting. So, I've started putting together some...

pretty standard for ZFS; expensive higher IO disks for production, larger capacity slower disk for backup. You'll find this to be a "run of the mill" use of ZFS. I have several SAS based production ZPOOL's (mirror based) shipping snapshots via zfs send/recv and netcat to a SATA based (raidz)...

Hi houkouonchi, I noticed you had a smartinfo.pl scrip that you used to agregate all the critical SMART stats for your disks. Is this publicly available, or was it something you wrote yourself? I'm looking for something exactly the same, and at the point where i'm going to write the script...

Just ordered up 8 of the 4TB Seagate ST4000DM000's to replace my 2TB Samsung HD204U's (need more space!). It'll be interesting to see how they perform. Now I just have to migrate data and sell the 8 x 2TB disks before my wife finds out I bought $1600 in disk drives:D

CPU discussions only really come into play with large ZFS systems with have enough disk to saturate controller and CPU I/O. I've noticed this recently when performing "zpool scrub" on a system with an older Xeon Quad Core. It was hitting 800+ MB/s, but running "top" revealed 100% cpu...

*subscribed* This looks interesting guys. I'm interested in the Solaris side as well as I'm having backup time window issues with ZFS send/recv's in our production environment. Looks promising!

Are you sure? It's entirely possible to accidentely fill your root pool by doing something stupid like a zfs send/recv to it, or something similar. I'm with brutalizer on this one; boot the VM with the OpenIndiana CD ISO image and see if you can mount the zpool. At that point, do a 'zfs list'...

Light fragmentation will rarely cause increased login times and degraded user performance on MS Terminal Servers. That being said, heavy fragmentation certainly can (ie: upwards of say 50%). Unfortunately this is a contentious topic as it's difficult to accurately measure the effect that...

Most modern raid controllers allow you to move your RAID arrays (ie: disk groups) between controllers of the same model or family and will simply alert you on boot that they have detected a foreign array that they would like to import. Most of them are smart enough not to wipe out your data...

Which raid controller are you using? In most cases, you can move the raid controller and disks from one computer to another while preserving the RAID layout, and thus, preserving your ZFS layout.

out of curiosity, why are you using an iscsi volume instead of just sharing a folder through SMB/CIFS?

If you've published an iSCSI volume, depending on how the OS/web interface is configured, this can mean one of two things: 1) a ZFS ZVOL is created and shared out using comstar or the BSD equivalent 2) a flat file is created within a standard ZFS file system volume, which is then shared out via...

Interesting idea. Is there software available to do this? I would think some kind of "proxy" with a time slider capability would make sense. If it already exists, I would probably use it as a lot of sites i've book marked say, 5 years ago, no longer exist or have been restructured and thus...

VT-d doesn't so much work on a "PCI Slot" basis, so much as a "PCI device" basis. To illustrate, take a look at the vSphere console to my ESXi server (Supermicro X8SIL): You can see that I have passed through the onboard USB devices, as well as an LSI SAS card (LSI 2008 aka 9211-8i). You...

A distinction should be made between "Solar Flares" and "Cosmic Radiation". While one does not necessarily preclude the other, Solar Flares are a little more dangerous due to their potential strong EMF/EMP properties which can theoretically fry the actual circuitry of a disk, or whole computer...

Cosmic radiation: yes. Solar flares... m-m-m-maybe-not. This is my point though. Many of us have invested some decent cash in ZFS rigs that are able to detect silent corruption over time. What about protecting the data/system itself from the effects of a decent sized spike like that...

Looking for something a little more "home power user" friendly :D Although, if we get enough people together, we can probably fund our own "Hard|Forum Community DC" somewhere!

A word of caution as well: always make at least two copies when converting to new media, or at a minimum, retain the original copy for a time. Issues with media (CD/DVD degradation) and the like only seem to become apparent after the media has been around for a while. Remember when we used...

As for free programs, I've had good success with "Recuva" http://www.piriform.com/recuva It's lightweight, it's free, and has been able to recover file from different "failure" scenarios. It's creators are also the creators of CCleaner and Defraggler.

Agreed, not good. Check SMART and also check your systems's uptime ('uptime' command). Either you have a bad disk, or perhaps the systems rebooting on you. Check /var/log/messages for any messages relating to md0 or your physical disks (sda/sdb/etc): cat /var/log/messages | grep md0 The...

Because I'm in the market for SSD's right now, and will be buying a spread for testing :) ZFS ftw! Corsair FTW :)

Oh, and I'm not saying that my previous "method" of "physically connecting, powering on, backup, power down and physically disconnecting" and stashing in a Faraday cage is necessarily bad either. Lets face it, having a backup that is 30 days out of date is better than losing everything, period...

Brutalizer, agreed. Though you gotta admit, it's a thread to have a bit of fun with as well. On a serious note, how would we propose to ensure our data is safe from the potentially massive EMF that could be created by such a large scale flare? I've been doing some research on the topic...

So if a Faraday cage will help to diminish the effects of a severe solar storm, would not a metal server rack operate in much the same way. Assuming the doors, walls, ceiling and bottom are all grounded to each other and, well, the ground, It should operate essentially like a Faraday cage! To...

As I understand it, ESXi has had support for USB 2.0 devices, and with ESXi 5.0, it now has support for USB 3.0. While I can get my VM's to see external USB devices (hard disks in this case), the performance sucks! Typically I get transfer rates of around 7MB/s to/from the disks, which really...

Just found this: http://www.exactfile.com/ Looks exactly what I was after.

But derailing is fun :D Back on topic: speaking of checksuming, is there a good windows based util for creating MD5's for reverification? Whilst I use rsync for large copies, my wife could use something more pointy-clickie.

A very wise practice. md5sum or using "rsync --checksum" is very useful in this regard.

Yes and no. File's in and of themselves simply contain data. Some applications do have the ability to detect corruption in their data streams (ever come across a corrupted ZIP file?), however the vast majority do not. There are different types of checksum/verification that can apply here...

Actually the number of disks can be important in several differing ways: 1) Performance - higher disk count, higher potential IO (to a point of course... controller needs to be capable!) 2) Power consumption - fewer disks, lower power consumption 3) Higher capacity disks = more expensive...

Looks like a re-brand of this one: http://www.dataonstorage.com/dataon-products/6g-sas-jbod/dns-1660-4u-60-bay-6g-35inch-sassata-jbod.html

Correct. Unfortunately there will probably never be an official ZFS port to linux due to license incompatibilities. However, there are some non-official ZFS ports to linux, one of the major ones being: http://zfsonlinux.org/. I cannot vouch for their stability though.

Oh ZFS.... oooooh..... :eek:.....:p Now, back to reality. One option for being able to guard against bit-rot without using ZFS could be using QuickPar (http://www.quickpar.org.uk/) It's able to calculate file level parity (instead of disk level), generating some additional "parity" files. In...