5 paragraphs of stage setting:
Since the late 1950’s, we’ve seen communications, computing power and memory per square inch steadily improving. While the rate has varied, it remains positive and is likely to continue for a few decades. Even the largest servers costing $10k/each new, two or three years later fetch $1K while having a decade of life left.
Today, most business locations have access to at least two physical internet service providers (cable and fiber) and now with radio-based providers at 4, 5 and 6G offering speeds more than sufficient for business purposes (other than those akin to providing large catalogs of movies to a large audiences or with rare extreme demands / low everyday demands).
Internet security has become a ‘cat and mouse’ game of ever greater assurances of new encryption methods punctuated by news of devastating breaches at data centers– even among our most technically astute private and government organizations.
The rise of “Software as a service” is driven by two things: ‘providers’ strong desire to generate a recurring revenue model, (and in the case of large market sectors of ‘canned’ software like accounting: fees not really justified by year-on-year gain in valuable function), and end-users willing to put up with that arising from concern about hardware downtime, software/system maintenance/upgrades, internet outage and etc.
We’ve all experienced high stress at business locations with variations on the theme “the internet is down” or “we can’t take credit cards or make appointments now” or “the server is down”. That forces service teams, whether in-house or outsourced, to provide highly capable staff 16x5x50, sometimes 24x7x365. That generates the expense of retaining competent staff at off-hours — and costs relating to keeping sufficient labor capacity largely unused during intervals of slack demand so as to be able to handle peaks. In a fluke of statistics, we know failures tend to clump, which adds stress both on service arrival time and staff taking ‘shortcuts to partial fixes’ and related blunders.
That’s it for stage setting. With that: A long term new effort, to succeed in a very large scale, global way, in accord with the above realities:
- must not ‘row upstream’ against the improving speed and lower cost of computing and communication.
- has to take advantage of multiple internet on-ramps at most business locations, and provide everything that can be done without the internet — without need for the internet. To include company email, websites, databases, phones, custom/local software already running but at risk because ‘one server’, etc.
- has to provide a ‘game changer’ approach to internet security, as better encryption vs. crackers feels like having to defend a large balloon 24x7x365 against a small pin one time.
- must not pick favorites among operating systems or software language or delivery styles, but make ‘best of breed’ in the client’s area of concern available: whether that’s open source software, purchased software, local custom software, connects via pad, phone, laptop, IoT device, industrial/medical/security machine interfaces, even making practical the current software-as-a-service vendors locating their proprietary product in a ‘locked set of machines’ at the customer’s shop.
- must make it possible for highly competent technical service staff to do all their work during daytime weekday hours– no matter the nature or timing of the client hardware/software fault, so freeing their employer to shift money from second and third shift staffing to attract more capable first shift talent. Ideally, a repair staffer arrives at a client shop with a replacement server, the client staff looks up in surprise “What are you doing here? Did something go down?”
That’s the problem I think “Rock Stable Systems” solves.
RSS is for teams wishing a cost-effective fault-tolerant platform they can integrate, locally extend and maintain among their locations; for teams aiming to ‘provide the cloud’ to their staff and others; for teams who see value in avoiding loss and risk associated with every-minute dependency upon cloud-distant service providers. RSS provides a robust security infrastructure, yet can be integrated into pre-existing ‘cross-domain trust forests’. RSS operates at the sweet spot of sturdy, performant, affordable, locally maintainable and extendable performance as hardware and software technology allows. It is for those who want to be ready for “Web 3.0”. RSS ‘checks all the boxes’ in the service areas we support. RSS scales from a little batch of used servers in a dishwasher-sized cabinet to a room full of server racks cross-linked over a dozen+ locations.
“Two is one, and one is none” means all your internet on-ramps need to be down and, for local capabilities: more than one major server (not just one storage drive) dead before your business suffers. If you have RSS at several locations, so long as one is ‘up’ your information is available.
Those with new high speed internet connections are right to think “RSS will make me more valuable to my market and be valuable for me internally as well.”
It ‘just works’ and because we know equipment will break, it’s designed to keep working long enough for repairs to take place during normal business hours in nearly all cases. No more ‘everything stops and everyone is sweating because the network computer thingy is down and we can’t get a part installed for hours, days or whatever.’
Technology “power users” who feel they can maintain something well, so long as it comes already set up and all the information about everything (from overview, to manuals, even source code in most cases) is online. “No black boxes”. Learn as much as you like about the parts you care about most, at your pace. Specialists are available the world over in each of the major subsystems. We’ve got the complex stuff that’s not in your wheelhouse covered, you can spend your time where you find the value.
- Organizations led by those who know they could boost both customer and staff morale, and afford and retain the most capable staff– if only they could promise an end to routine evening, night and weekend work; who would appreciate an end to stressed out customers breathing down the necks of the repair staffers.
- Organizations that specialize in some aspect of the information world, looking for a ‘stable platform’ with all the interfaces accessible, ready and waiting for them to build upon. A platform compatible with all known ‘ecosystems’, one that does not ‘lock you in’. Like an ERP or CRM solution and know folks who can install it? This is the base that could run it and do ‘everything else’.
- Those looking for a way to know what’s broken and before the first visit to the customer site — have all the parts in the trunk so repairs happen in one trip, and usually before the customer notices harm.
- Internet service providers looking to provide backup solutions to big customers, so routine outages don’t cause insurance claims.
Elsewhere in the menus of this website you’ll read of the capabilities the various RSS subsystems, a bit about related structure and sub-components, rationales and history. But RSS itself, what is it beyond a collection of the work of so many other teams (all of whom we honor greatly)? A few things. It is:
- An AI trying to do most of what a “Ph.D in everything” paid to live and sleep next to your server closet would be hired to do, except for using a screwdriver.
- Over 10,000 lines of software, mostly written in the Python programming language but with small parts in several other languages. Python positions itself as “a programming language that lets you work quickly and integrate systems more effectively” — which is precisely the use RSS makes of it. “A Force Multiplier” — who else uses Python? YouTube, Industrial Light and Magic, Google, and ten or so thousand others. Does that mean clients can edit and change RSS on their own? Short answer is similar to the answer given about airplane pilots: “There are old pilots, there are bold pilots, there’s a reason there are no old, bold pilots”. RSS is a foundational system, an organizational or departmental computing platform. Adding hardware, virtual machines of custom local design that do special useful local things well integrated into the file and admin structure is a very good idea and well provided for. Changing RSS code itself? Well, by agreement, yes it is possible. But you probably do not want to do that.
- 10K lines of code is sort of ‘a lot to some, not so much to others’. It’s the quality, not the quantity. But what does it do? It crafts and monitors the configurations and installations and good operating order of everything involved. It does backups, keeps track of what’s working and what’s broken, keeps people who care posted about how things are going, moves loads to working systems and, well, read the rest of the website for details.
- A network and multi-system hardware/software architecture designed to optimize performance and interoperability for even a very large local environment, collected in echelons of squads composed of 1 to 16 far-flung locations. It is ready for as much ‘scale’ as a few hundred thousand people at the same time might throw at it. If your one department grows beyond that many people at the same time accessing your systems– well, I expect such already have a few “Ph.D’s in everything” already sleeping by server cabinets ready to solve that one for you.
- The fourth ‘from almost the ground up’ version of a design that started over 10 years ago. Now the maturity of the subsystems and economics and reliability of the hardware is ‘right’. The number of competitors/vendors for each subsystem that have been tried and replaced far exceeds the number that found its way into the final design. Been there, done that, so you don’t have to. That’s what you’re paying for.
- Lives by the core value originally credited to the U.S. Navy Seal Teams: When time matters, two is one, and one is none.
- Compatible with today’s internet, and ready to be your organization’s Web 3.0 basis.
- Is built and tested under very tight and limited hardware constraints, much tighter than any client installation. Why? Because when it is made to perform, to make the most of and be reliable under conditions of load with limited resources — when those resources expand it’s faster and more reliable than anything else built under conditions of excess storage capacity and luxury speed.
Though RSS is feature laden in itself, it’s meant to be extendable by people who are really good at what they are good at — without forcing folks to learn all the details of everything else as well. In other words: Teams are able to use what RSS offers, and can grow it, without breaking what’s there or having to learn everything we are good at in order to add to it. It’s not an ‘everything depends on everything’ design– you can add capacity where you have demand without having to ‘upgrade the whole thing’. No need for additions disrupt what isn’t broken. At a click, add ‘processing cores and memory’ to boost speed of functions that get heavy use, or add whole dedicated computers that ‘just do that thing you need a lot more of’ — How? Install the OS, Install RSS, set the hostname to indicate the function required, run RSS. Done. Can it be a pair of used computers that together do more yet cost less than one new one? Yes.
What’s the marble red carpet staircase logo about? Because it’s at your locations, RSS is an end to dependency: RSS owners can offer their resources to others, can be their own cloud providers both to their own teams and to clients, along with benefiting from all the customary local capabilities.
RSS is the fourth version of a design in the works for a decade. It is our ‘magnum opus’, designed to be reliable foundational product, capable in itself, comprehensible in all important result dimensions, freeing from dependency, maintainable by ‘technically able but not genius-level local folks’, extensible — without requiring the owner to ‘know everything about everything’ in order be a platform for locally generated further capability. A feeling of maximum stability but generated through the latest design possibilities. That’s what I was aiming for in the look of the ‘Chic skyscraper foundation’ with the accent of red-carpet marble climbing stairway upward.
Some important trends have come together to make it “time” for RSS, such as:
That fact was recognized years ago and written up as ‘Moore’s Law‘. Most devices with computers in them are so fast these days, most people aren’t surprised when a spreadsheet recalculates before their finger gets off the button, or a web-page displays faster than they can read it. Systems that displayed “Please Wait” or “Please Reboot” were once accepted as expected, but that software now is considered ‘legacy’ or ‘badly written”. One computer on a chip is obsolete before it is released, now a chip has 4 or 10 ‘cores’ — several whole computers on a chip the size of a cookie. A 10-core server chip now costs less than a good steak dinner, and a whole server with two of them and more memory than it takes to hold several perfect resolution full length movies costs less than $1,000.
Every year news articles assure us today’s encrypted information crossing the internet is ‘safe and private’– until a year or two later news revealed even the largest most technologically well funded and prepared firms, even government agencies and the latest technologies were ‘breached’. But, never fear, cloud advertising supported headline writers assure us there’s something new that’s keeps your information crossing the net ‘safe and private’.
- Recall internet encryption technology that showed the ‘safety padlock’ on your browser, “SSL”, for many years was touted as safe and secure”. Now that’s gone and it’s “TLS, safe, secure”.
- Information leaving your location without need has the same feeling as walking after midnight in an alley behind a bar in a tough district with a purse or wallet bulging in your pocket— but you have this can of pepper spray. Is it illegal for someone to take what you have? Why yes it is. Does government have good success arresting the bad guys? Not so much. Is there something maybe you could do differently to avoid that risk? Well, why yes there is. Are the late night bar owners going to tell you how to protect yourself better or leave earlier? Which is better for you? Which is better for them? Bingo.
- The core contribution to your productivity of most major software products hasn’t changed much over time — a spreadsheet from 1986 looks pretty much like 1996, 2006 and now. Someone who started using computers a few years ago would have no major problems running an early Mac or a “Windows 97” box. Double entry accounting systems have been around for decades, the version that did your payroll 20 years ago could still do your payroll today if the tax tables were updated. A few years ago, most software products were sold in a store– pay once and you’re good for as long as your computer can read the install disk. Computers have only ever gotten faster… but is ‘cloud’ a good reason in itself to not only send data across the country every time you change something, nevermind pay monthly for access to software that could run locally? Software’s major capabilities change very slowly, if at all. Which makes some sense after all, as software has more in common with a book or blueprints or a math equation that’s the same 20 years later than it is to a bottle of soda pop that goes flat an hour after it’s opened.
- The best security is not to be a target in the first place, that’s always better than risking whether your security precautions are better that day than the bad guy’s capabilities. Avoidable risk is a good idea, as is sensible and necessary risk. RSS balances that.
- With RSS: Nothing leaves your building unless it has to, and only when it has to– not every time a staffer in the same building accesses it “because cloud”. With RSS company websites and calendars and documents and databases and email stores are available in the building whether or not “the internet is up and not being weird”. Even better, cloud providers, if they prefer, can locate their services in your building. RSS is extensible.
- For instance, Email from one desk to another in RSS says in the building, it doesn’t first go over your internet service provider’s wires, then over whoever connects to your mail providers, then into your mail provider’s systems where it is a sweet, ripe, million-eggs-in-one-basket hack target target, then back again. And not just when you send it, but again every time they check it as well.
- It’s not just the data itself, but logs of your business activity times and work patterns and addresses of your business clients and partners being collected– then packaged for sale.
- The logs and ‘meta-data’ being collected and sold regarding your business practices, information about with whom you are conversing and when being sold as a product— because buried among the two dozen paragraphs of legalese you gave permission when you clicked ‘I agree’. You read and understood all that legal language, right? You know paragraph 15, using a word defined ‘just so’ in section 2, along with paragraph 35 means you said ok to everything. Right? Half of nobody knows, small comfort that is.
- While you might be assured your firm’s staff isn’t ‘personally identified’– how many businesses of your type are on your block that buy and sell from your suppliers during your published operating hours? Search and replace that ‘anonymous ID number’ in the sold data with your firm’s name — and there it all is. The dark side of “Data Science”.
- With RSS, not even a ‘map’ of your technology layout ‘leaks out’ through innocuous seeming ‘requests for the current date and time’. No third party without your consent sells information about which websites you visit and when, nor who online is visiting your websites– because RSS within each location looks up the internet address of the websites your team seeks to visit.
Notice how even your mobile device in no-internet, no-wifi ‘airplane mode’ plays movies or edits documents at full speed? Any computer that operates faster than you can type is ‘fast enough’. With RSS, everything that can possibly done locally is done locally. With a design that keeps performing when parts break. One benefit is: the network capacity you pay to receive from your internet service provider is used just and only to fetch and put what can’t be provided any other way. Not each time you update a spreadsheet cell or check your email. (But if you choose to collaborate with someone who only has remote access — RSS supports them, but even then without burdening the internet with locally entered work).
As ‘everything in an RSS system is under client company control’, from public to private presence, federation and resource sharing with others: there is no risk that client efforts can be lumped ‘for’ or ‘against’ political or other enthusiasms arising from the leadership choices of ‘cloud providers’. Or, choices that may result in some share of your business partners or customers being ‘cancelled’ or insulted or driven offline by ‘blunt instrument cancel-technology error’.
So long as there is at least one internet service provider connected to your location, your clients can reach out and if they so desire those from the public can contribute. Does that mean your client operations via RSS are somehow ‘above the law’? Quite the opposite, it means our partner and the client’s partner is the government of laws and due process — not subject to the variable views of the leadership of cloud tech companies.
What then does RSS offer? Integrating several computers so ‘the whole system that does what most need keeps working even if’– in a client setting with people who, because of RSS, do not need become experts in the complex configuration of everything, and in an affordable fashion– that’s a mouthful but that’s the hard problem RSS solves.
Teams with deep expertise in one area don’t have to ‘learn everything about everything’ in order to make it available in an organization based on RSS.
Are we saying an RSS location will never break down? RSS keeps running as we expect occasional broken parts. It means we have the maturity and experience to have an affordable design in place so everything keeps working– because we do not ignore that it is ‘when’ and not ‘if’ some aspect will break down. Usually each location will not suffer actual performance loss unless they delay to fix a broken part until the second or third instance fails. The storage subsystem doesn’t even wait that long — it makes always keeps three copies of data on different systems — using pre-planned capacity to make new copies immediately when it notices one failed.
What a nice change might it be if, during scheduled normal working hours, a repair staffer appears without having been called, instead of called because your work is disrupted and idle and customers are upset. What if a repair staffer just shows up unexpectedly, without having been called, does ‘something in the back cabinet’ then leaves and ‘nobody seems to know what was broken’. Why? Because in one trip, with the correct parts in hand, the RSS design permits remote diagnosis and the ability to fix something your team never noticed was broken. Compare that to ‘we can’t take credit cards or check appointments now because the internet is down’.
RSS’s ability to avoid loss often far exceeds its cost.
What a nice change it might be for a technology support / IT firm, if you could satisfy your customers 24×7 but only need to hire more capable technology folks and offer them 9-5 Monday – Friday hours? Because when something breaks, you have until the second one fails before your customers notice? How nice a change to your bottom line and staff retention might come from near non-existant second or third shift emergency repair calls, rare ‘weekend shutdowns for repairs and upgrades’?
The main problem RSS solves is using these hardware and software system cost and capability technologies together in an fault-tolerant, affordable way. RSS does add some technology on its own, but in the main orchestrates the great flexibility and configuration complexities of very time tested and proven subsystems so they work with one another as if written together.
The built in S3 capable, block device and internal / external network design (v4 and v6 whether or not the local ISPs offer v6) makes that realistic.
Fact.
Yes. That’s standard.
Rock Stable Systems is a project of Quiet Fountain LLC.
Email Us at frontdesk@rockstablesystems.com or call 563-650-7800
- A native, automatically configured, client-site hosted Git source code control system / sub-module (already in use internally).
- A native, automatically pre-configured reliable telephony or ‘pbx’ system. We’ve chosen the the Freeswitch package, as managed by the GUI and call-center functions added by Mark Crane in the FusionPbx product he leads (already in use internally in the RSS version prior to this one). RSS already provides video calling and group video conferencing via Nextcloud (documented elsewher on this website), the FusionPBX package links in voice and integrates traditional telephony with SIP operations. (We like Flowroute).
Contact Harry G. Coin. Kindly understand RSS builds on the work of dozens of hundreds of the best in the world today.
Those with experience who read my history in the link above might infer what “I’m about”. The publication of the first internal version of this website in April 2022 wasn’t a destination, but a milestone along a creator’s road I’ve loved traveling. Getting to know the work of so many other talented people, integrating it into a cohesive whole has been a joy. Now it’s time to change emphasis: less time creating, more time in the world of business relationships. I so hope to supply RSS to many, as I designed it with the aim to enable others to generate value. The architecture of the entry level system and pricing, and the first public version of the website came in early 2023. Onward!
Well, it’s a trick question. RSS has to ‘work with everything’, from Android to Windows to Apple and all the popular Linux distributions (such as Ubuntu, RedHat, Debian, Mint, Suse, etc. etc). As such, each subsystem RSS provides uses the ‘Linux Distro’ known to best support the contribution that subsystem offers. Currently RSS uses Debian-heritage subsystems (Ubuntu by Canonical) and RedHat heritage subsystems. Could they all be ‘switched’ to use just one? Yes. Is that a good idea? A question that comes up often.
Every sub-system documented here has from dozens to hundreds of companies and thousands of people in countries across the world who specialize in the details of that subsystem. They’d love to help. For details, see the links provided in the documentation sections for those features on this website.
Yes!
“Too long, didn’t read”: No single point of failure, the whole cloud located within affordable department / office locations, multi-ISP support, attention to scale, 10K+ lines of python, Ceph, Alma/RedHat, Ubuntu/Debian, Freeipa, Nextcloud, WordPress, Nginx, HAProxy, Android, Apple, Windows, KEA, Bind9, Email, web-hosting, databases, video conferencing, mobile file/photo sync + redundant storage, calendars, KVM, Libvirt, Dovecot, Postfix, MariaDB, Galera, SIP, DNSSEC, WebRTC, STUN, TURN, Kerberos, LDAP, SSL, Certificates, SSH, NTPSec, Chrony, Gerbera, GIT, Eclipse, Autofs, Nftables, WireGuard, VPN, Sync, some security stuff only documented to clients, Wireshark, Selinux, Apparmor, Spamaassasin, Spamhaus, Samba, PHP, Redis, Git, Freeswitch, — and the dozens of sub-packages required by them, 5 rack-mounted servers in dishwasher sized cabinet or less, up to racks and racks, and, etc.
Hopefully, that should help you justify more reading time! For more, see the rest of this website.