VM side of things is more complicated. As you can tell I stopped trying to make the diagram pretty
Red is the WAN network
Black is the LAN network with a route to the internet
Blue is the internal network for accessing storage
The docker network is actually a separate natted network and should be a different color
Why learn Linux? Because, quite simply, it is the future.
https://www.youtube.com/watch?v=6TBRTLsCNXs
I truly believe that Linus and Slick, being sources of knowledge for many in the tech/PC community, should feel not only obligated but that they owe it to them selves and to their audience to start learning Linux. Ignorance will only hold back what is to come. I know LMG is a busy crew and I don't think they're being intentionally ignorant, but rather I don't think they realize the scope and potential of what they're ignoring.
Regardless, the more of us (the members of the community) delve into the world of Linux, the better it will get.
Thus, I'll keep on advocating for Linux. Check out the video from Wendell from TekSyndicate. This is just the beginning of a whole video series (they've devoted a YT channel specifically for it) on Linux. He'll be exploring the ins and outs, how to install, what you can do with it, how to diagnose and fix problems, and much more.
I cannot and will not recommend something that isn't worth the money. It's a horrid practice to appease a bad decision, especially when it's someone else's rig. Give them the slightly higher product and if they are in THAT MUCH of a money crunch, tell them to use stock cooler and not OC until they can afford it.
Ah that was another unique show I enjoyed. I don't remember too much about it unfortunately.
It may just be me but I felt Kamisama no inai nichiyoubi had a similar "feel" to it and I also liked it.
I watched these a while ago. Really enjoyed Humanity has declined. Very different and it helped that I liked the main character's voice.
I liked Maoyuu at first but I felt the plot was moving and changing so rapidly towards the end that I could not follow. Also felt really weird that not a single character had a name in this show. I wonder what they were going for with that.
I watched these a while ago. Really enjoyed Humanity has declined. Very different and it helped that I liked the main character's voice.
I liked Maoyuu at first but I felt the plot was moving and changing so rapidly towards the end that I could not follow. Also felt really weird that not a single character had a name in this show. I wonder what they were going for with that.
Wendell over on Teksyndicate recently posted this thread and I highly recommend you all check it out: https://teksyndicate.com/forum/linux/community-linux-content-ideas/191796
Linux is so amazing in so many ways and especially among the tech community, it needs (and deserves) FAR more attention than it receieves. Logan and the gang are in the process of planning and creating content specifically aimed at Linux and to help those who aren't familiar get more interested/informed about Linux. There is a stronger presence of knowledgable Linux users on Teksyndicate, who've been spurring-on and suggesting they put out some Linux-specific content. This will help spread knowledge and push more people to start using Linux (or at least try it), but we need more people on-board. The more people use it, the better it will get.
Linus and Slick, I would like to put forward a request and challenge to one or both of you to start using Linux on a daily basis for a variety of tasks (email, work, gaming etc.). The thing is - it can't just be for a 2-week period, it has to be for long-term use. I would suggest building a Linux-specific PC, install steam and some other programs you guys use on a regular basis and give it a go. I understand this will require extensive research on your behalf as you guys don't have much prior knowledge with Linux (I believe, correct me if I'm wrong), but the fact remains - Linux is fast becoming an important part of the tech community and is slowly but steadily gaining traction/popularity. More and more games are being natively supported and more and more programs are becoming compatible as well.
Many of us in the tech community are not "the typical user", which is partly why I hate it when people from the tech community use that argument against Linux saying; it's not for the typical user. We overclock, tinker and experiement with our PCs and equipment, and we enjoy doing it. Linux is perfect for people like us. It holds so many benefits for the power-users and even for "the typical user", but people don't know this and will never realize it unless more recognition is given and supported by the community. It's not just for the "elite few", it is and can be for everyone - depending on what your needs, preferences and interests are.
Linus, Luke, what do you say? (Obviously you don't have to give an answer here and now, but at least give it some serious thought?)
What do the rest of you think? Should LTT put out some Linux content? Go over some of the more popular distros, explain where Linux came from and how it's different *cough* and better *cough* than windows... Take a look at which hardware is best to use with Linux and why... Show how easy it can be to adopt a newb-friendly distro like Ubuntu or Mint for daily use and share that experience... There's a lot they could cover and I firmly believe this could open a lot more eyes because people who normally avoid the Linux sections/discussions among the forums would more likely watch these videos and perhaps develope on a new appreciation for Linux. Like I said earlier; the more users, the better it will become.
You won't be able to keep your data and move to motherboard raid (or any firmware/hardware raid). This goes for all raid levels including raid1 and even raid0 or JBOD of a single disk. Some software raid options may allow you to keep your data. Not sure.
4 disk raid10 sounds like a good option to me but you will need to back up your data elsewhere for migration.
That config was using ZFS with all 8 disks in a single pool with no RAID or compression so I admit it's a bit unrealistic. I'll rerun them with Raid Z-2 possibly with compression which is what I'll be using long-term. I have one SSD for SLOG and L2ARC that I'm experimenting with. I mentioned the test being sustained transfer using linux dd. I've also begun running other tests under another set of benchmark tools (fio) to check for random and sustained for a better measurement. No matter what speed they end up, I'm still limited by the 1Gb NICs until 10Gb becomes reasonably affordable to upgrade to.
My hardware is as follows:
SUPERMICRO MBD-X10SL7-F-O
1 x Crucial 16GB Kit (8GBx2) DDR3L 1600MT/s (PC3-12800) DR x8 ECC UDIMM 240-Pin
Intel Intel Xeon E3-1270V3 Haswell 3.5GHz LGA 1150
8 x HGST Deskstar NAS H3IKNAS40003272SN (0S03664) 4TB 7200 RPM 64MB Cache SATA 6.0Gb
1 x SAMSUNG 850 Pro Series MZ-7KE256BW 2.5" 256GB (SLOG & L2ARC)
1 x SAMSUNG 850 Pro Series MZ-7KE128BW 2.5" 128GB (boot drive and other processing)
i have some i have rewatched tons, a few i would never, and quite a few that i have watched twice over the past 8 years, for example i rewatched the entire Aria show 3-4 times in a ROW.. last time i just had it in the background on another monitor but its so freaking relaxing thats is amazing, also rewatched One Piece from 1 to 250 then 1 to 350 then 1 to 400 something nearly in a row...
There's a lot of people here that would rather maintain their own ignorance/intuition instead of trying to learn something that invalidates their own ignorance/intuition.
Edited 8/24/2014: Section headers added and essay cleaned up.
BEFORE YOU READ: This thread is intended to serve primarily as a vehicle for a brief history of how we got from UNIX to OS X as it stands today, secondarily for discussing why I use OS X and by extension, why I use Macs, and as a tertiary intention, for the discussion of the pros and cons of various UNIX-like or UNIX derived operating systems, or other operating systems. It will also highlight Apple’s rich history of fostering open development processes and open source software projects. Yes, you could find this information elsewhere, but it’s in front of you right now, carefully organized for your reading pleasure. If I correct some misconceptions along the way then so be it.
I’m not perfect, a lot of this information is shrouded in myth and legend, TL;DR if I get something wrong please don’t hesitate to say so and I will correct the OP with it. One cannot possibly recount the entire history of UNIX in a single forum post, so if you have any other questions, again, don’t hesitate to ask them. I will do my best to answer them.
Sources: Mac OS X Internals, A Systems Approach (osxbook.com) Wikipedia (wikipedia.org)
Knowledge accumulated over the years that I can’t really cite because I forgot where it came from but it’s been verified to be true (myself)
Sit back, relax, grab a coffee, and ponder life’s big questions as I take you on a guided tour through history of the most important software ever wrought of humble code by all humanity.
"Unix (all-caps UNIX for the trademark) is a multitasking, multiuser computer operating system that exists in many variants. The original Unix was developed at AT&T's Bell Labs research center by Ken Thompson, Dennis Ritchie, and others." (http://en.wikipedia.org/wiki/Unix)
Everything is a file: UNIX is created.
UNIX was an advanced multi-user operating system whose first release occurred in 1973. It was originally intended to be a "programmer's workbench" for writing code that was compatible on many platforms. Designed primarily by the now famous Ken Thompson and Dennis Ritchie, UNIX had its roots in the MULTICS (soon to become UNICS as a joke on UNICS being a castrated version of MULTICS after Bell Labs pulled out) project started at MIT in conjuncture with AT&T Bell Labs and General Electric in the mid 1960s. At Bell Labs, UNICS was rewritten in the C programming language, which had been designed specifically for the task. The idea behind UNIX was to develop an operating system that used a minimal amount of machine language (assembly) code to ensure that porting it over to new platforms would require minimal effort, often only porting the assembly code base and writing a C compiler. This may sound elementary now, but at the time it was revolutionary to create a system that could very easily be ported to many other computing platforms with minimal effort. Traditional efforts were often assembly language only. At this point in time, UNIX was known as Research UNIX, and was completely proprietary, with source licenses being available at nominal cost.
One of these source license was issued to the University of California Berkeley, where a graduate student named Bill Joy began compiling his own version of UNIX for distribution. The first release of the first Berkeley Software Distribution (BSD ) occurred on March 9, 1978, and in edition to UNIX’s core codebase, also contained a Pascal compiler and the line editor ex, of Joy’s own design. This release was known as 1BSD. 2BSD would be released a year later in 1979 to include two programs that remain at the core of programming on UNIX systems to this day: the famous (or infamous, depending on who you ask ) vi text editor and the C shell.
The Free Software Revolution and the Birth of GNU/Linux
Though now in widespread usage, (and long before Windows took over) BSD was subject to licensing fees from AT&T, which had grown quite large. In response to the growing price of a BSD license, a man by the name of Richard Stallman founded the GNU Foundation to start an effort to create a free software version of UNIX that would be both free as in beer and free as in speech. The GNU Foundation also authored a new software license known as the GNU GPL, which introduced the concept of “copylefting” (a play on words of copyrighting) to the world. Copylefting meant that modified version of the source code had to be distributed containing the same license, and it served in the GNU license to ensure that a company could not leverage GPL licensed software for their own purposes without giving their modifications back to the community. It restricted the distribution of binary-only sources, that is, sources which can be executed but not read by humans. The GPL is actually a lot more complicated than this, and if you are interested in software licensing and the GPL in particular, I urge you to read http://en.wikipedia.org/wiki/GNU_General_Public_License.
As part of Stallman’s project to create the ultimate free software distribution of a UNIX-like system, a project was launched to create a revolutionary new microkernel on which to base the system. This project is known as GNU Hurd, and to this day they have not produced a kernel suitable for consumer or production usage. Never let it be said that Stallman is a quitter though, as Hurd is still in active development well over its 30th birthday.
Obviously, those impatiently awaiting Stallman’s halo UNIX derivative grew more impatient. Seeing the problem with the Hurd and fearing (as is now obvious) that it would never get completed, a brilliant systems programming by the name of Linus Torvalds decided to come up with his own solution; a GNU GPL licensed kernel compatible with the rest of the GNU software library (which had been completed in spite of Hurd’s perpetual state of almost-doneness, which, as a side note, Hurd has been in since Stallman stated it would be done by fall 1990) now known as the Linux kernel. Suffice it to say that the Linux kernel took over, despite it being a monolithic kernel while Hurd was intended to be a microkernel.
Chart of UNIX, UNIX-derived, and UNIX-like Operating Systems
At this point you might be wondering, what the fuck does this have to do with OS X? Don’t worry, we’re getting there. The history of free software to this point is important to understand, because it was pivotal in a 1985 startup’s history and in general helps in understanding what happened in the next 12 years before OS X was released. If you’re looking for a tree of all of this stuff, brace yourself:
Five years ahead of its time: The 3M challenge and the founding of NeXT.
Ousted from his position at Apple after working the Macintosh and Lisa teams into the ground in 1984, Jobs set out to create the perfect “3M” computer at his new company NeXT Inc., with at least a megabyte of RAM, a megapixel display, and a processor that could process one million instructions per second, (MIPS). His goal was to create the ultimate research computer for universities and research centers. When asked how he felt about the NeXT computer arriving months late, Steve Jobs (in typical Jobsian fashion) replied, “Months late? This computer is five years ahead of its time!” And on this occasion, most people think he was right in saying so. Tim Berners Lee would later use a NeXT “Cube” (named affectionately for its cube like 1’x1’x1’ chassis designed by Apple Ilc case designer Frogdesign) and its corresponding NeXTSTEP operating system to create the World Wide Web, attributing the speed at which he did so to the operating system’s object oriented capabilities.
While NeXT’s early hardware failed to amaze, NEXTSTEP was a fully object-oriented operating system, a product that had previously been the subject of many years’ research at Apple.
NEXTSTEP and the Object-Oriented Craze
NEXTSTEP was composed of a hybrid kernel containing the Mach 2.0 microkernel and the 4.3BSD (sound familiar?) UNIX environment. NEXTSTEP introduced a large array of revolutionary user interaction paradigms, including ubiquitous drag and drop and an object-oriented device driver framework known as Driver Kit. NEXTSTEP used Objective-C as its core programming language. Objective-C is an object-oriented C variant containing syntax inspired by Smalltalk to which the comprehensive rights were purchased by NeXT. There is a joke among Objective-C programmers that Objective-C contains the syntactical efficiency of C with the compiled speed of Smalltalk, i.e. taking the worst points of two languages instead of the best ones. It was at NEXTSTEP release 1.0 that the first iteration of Interface Builder appeared, a feature that is now part of Xcode in OS X.
Despite NEXTSTEP’s resounding success and wide appreciation as an object oriented programming environment, it was unable to sustain the company’s lackluster hardware business and so in 1993 the hardware business was dropped in order to focus efforts on developing NEXTSTEP for the x86 platform.
OpenStep and OPENSTEP
NeXT developed a partnership with Sun Microsystems to release the specifications for OpenStep, a completely open platform containing several frameworks and APIs that could be used to create one’s own implementation of an object oriented operating system on any underlying core system. This specification was implemented in HP-UX, SunOS, and yes, even Windows NT. NeXT released its own OpenStep compliant operating system known as OPENSTEP in 1996. OPENSTEP was essentially a version of NEXTSTEP fully compliant with their OpenStep standard. (speaking of that bad product names thread, whomever thought it was a good idea to use all of these similar names and capital letters in NeXT’s product lineup should be dragged out into the street and shot. If these names confuse you, you’re not the only one.)
Meanwhile at Apple, Microsoft was beating them to a pulp. After establishing complete operating system dominance, it became clear that Apple’s floundering strategy of releasing half-completed products that barely worked and reeked of desperation, (such as a licensing program allowing their System series of OSes to be installed on any hardware, much in the same fashion as Windows though without having the critical advantage of actually being any good) would not work.
WHERE IS OS X?
Whew. 1500 words in and OS X hasn’t even been created yet. I promise, it’s coming. We just have another new kernel ( ) to cover before we get to the really good stuff.
Apple System OS and the Search for a New Beginning
The Apple System series of operating systems was looking long in the tooth. Being the default method of interacting with a Macintosh since 1994, Windows was dominating the market, releasing improved technologies year after year that further extended their lead. Apple began a frantic search for their next generation operating system, at first partnering with the Open Software Foundation to port Linux (see, I told you it was important!) to the Power Macintosh. The result was known as MkLinux, running on the OSF’s Mach implementation, and was completely open source. It featured the Linux kernel running as a single process in Mach. The core system, osfmk, would eventually become part of the OS X kernel base. Despite yielding osfmk, MkLinux was unsuccessful in providing what Apple wished for in their next-generation OS.
Apple announced the purchase of NeXT Inc., in late 1996. What would have been called System 7.6 was released afterwards bearing the new moniker, “Mac OS.” Showing how the tables have turned since, Apple focused on marketing Mac OS 7.6 as being very compatible with Windows 95 and the internet. Mac OS 8 would bring the Copland API to the Mac, which later evolved into the Carbon API, (now defunct). It also introduced the brand-new Platinum user interface and a brilliant little search engine for local drives, network servers, and the internet, known as Sherlock. Mac OS 8.1 sought to apologize to Apple’s power users for OS 8’s ruinous changes to the classic Mac user interface by adding back in a start button and allowing the user to choose whether to boot to the Metro UI or the desktop. (…) In all seriousness, 8.1 introduced such important under-the-hood advances as the HFS+ file system, the Macintosh Runtime for Java, as well as bringing in Internet Explorer and Netscape Navigator for browsing the web. Here ends the changes to the original System series that were eventually implemented in OS X.
Power Meets Usability: The development of Mac OS X begins.
Apple’s acquisition of NeXT was looking pretty bad at this point, having spent a whole paragraph improving their old codebase and not doing anything whatsoever with NeXT’s vast software portfolio. Just kidding. You knew I was kidding, right?
After buying NeXT, Apple began creating an entirely new operating system based on OPENSTEP’s codebase. The first step was creating a similarly awesome object oriented programming interface to OpenStep’s, which is now known as the Cocoa API. The first version of what would become OS X was known as Rhapsody, an effort to combine the intuitiveness and usability of Mac OS with the true grit power user features and standards establish by NeXT and OPENSTEP, and it was completed in September 1997 comprised of OPENSTEP, 4.4BSD and Mac OS, of course. Already we have a lot of open source software coming in, with 4.4BSD being totally open source and OPENSTEP’s OpenStep collection also being open source. In Rhapsody 2, more code, this time from the FreeBSD, NetBSD, and Mach 3.0 projects, was added. From there, Rhapsody 2 was forked into OS X Developer Preview 1, OS X Server 1.0, and Darwin 0.1. In DP2, the OpenStep-derived Cocoa API was implemented. DP3 introduced Aqua, which is the glowingly blue user interface that has been a hallmark of OS X design, (and dare I say an inspiration to Aero?) ever since. The OS X Public Beta followed the incremental DP4, featuring the open source Darwin core, the Dock, Aqua, and a new version of Sherlock. In 2002, Apple and the Internet Systems Consortium founded the OpenDarwin project in an attempt to create a vibrant open source community around OS X’s open source core.
And finally, on March 24, 2001, Apple released the first version of Mac OS X, 10.0 “Cheetah.” Drawing on nearly 30 years of UNIX heritage and open source community-based development, combined with the power and efficiency of the OPENSTEP environment, and the usability and friendliness of Mac OS, Mac OS X is truly “the world’s most advanced operating system.”
I run 8 Greens in RAID Z (I know it's not the most efficient number of drives for Z)
You can enable TLER on greens
You can have them not spin down/park at all
Speed is more than good enough for media storage and backups
Red's are the "NAS drive" but not everyone needs them.
EDIT: This is a good read... http://forums.freenas.org/index.php?threads/hacking-wd-greens-and-reds-with-wdidle3-exe.18171/#post-98858
It was pretty outdated. Please update the data when you get a chance
Hardware:
Main
See this post.
Backup
CASE: Supermicro SC846BE26-R920B
PSU: Built-in redundant 920W
MB: Supermicro X9SCM
CPU: Xeon E3-1220
RAM: 16GB ECC
RAID CARD: LSI 9211-8i flashed to IT mode. Chassis backplane has built in SAS expanders.
SSD: Intel 710 100GB (OS)
HDD: Seagate 4TB x8, Hitachi 3TB x10 (62TB raw)
Software and Configuration:
Gentoo Linux
ZFS with 2 raidz2 vdevs (about 48TB usable on each box)
Usage:
ZFS, NFS, KVM hypervisor
See for details:
link
Photos:
Main above, backup pulled out below:
Backup:
vm1 ~ # zpool listNAME SIZE ALLOC FREE CAP DEDUP HEALTH ALTROOTz2 56.2T 41.8T 14.5T 74% 1.00x ONLINE -
I need to update my posts at some point. I've done a lot of rearranging of hardware and have had a second box with 62TB of raw storage for some time now. It is mostly cold standby though.
madcow got a reaction from Mattias Edeslatt in 100TB+ Server build log (151TB to be precise)
madcow got a reaction from KYAG in Network layout showoff
madcow reacted to MEC-777 in Free Your PC, Learn Linux, Embrace The Future
madcow got a reaction from raj47i in Mini ITX - Headless Home NAS - with FreeNAS, iTunes, Plex
madcow got a reaction from CostcoSamples in Is Intel 730 Series still worthwhile?
madcow got a reaction from Fred Castellum in Heaven Society - Week 33 (MAOYUU MAOU YUUSHA & Humanity Has Declined)
madcow got a reaction from cae in Heaven Society - Week 33 (MAOYUU MAOU YUUSHA & Humanity Has Declined)
madcow got a reaction from UnknownEngineer in Help for choosing raid controller
madcow got a reaction from LAwLz in What happens if you plug a Ethernet cable into one pc and the other end into another?
madcow got a reaction from JLCitadel in Electronic Ecosystem Discussion (Ex. Apple)
madcow got a reaction from alpenwasser in LTT Storage Rankings
