CIQ's live chat with the ProTech Show! Part 7

always kind of my goal in terms of where i was going and what i was doing and and how to get there and so i always had a passion to this we created before doing centos i created chaos linux which is a community rpm based distribution of linux and that kind of kicked out or spawned uh centos but as i said there were some things we did right and some things i think we could have done better and i had a lot of experience that i felt would be really um useful for this but but most importantly i really wanted to get back into the

distribution uh i really wanted to to build this again uh and and take all of the things that i did right do that again and take all the things that i did wrong and try to fix them and uh so you could say you could say there was a little bit of a chip on my shoulder as well uh with the announcement that something that i created was then um end of life and i'm saying it was end of life even though you know that's arguable because it changed it changed life it it um metamorphosized in a matter of speaking it's it's it's no longer

what we set out to do and what we set out to do was to solve some critical problems and critical needs that a lot of people in the world and in the community had and uh and it's no longer doing that and so i gotta say between between us and the 50 billion people that are watching this um that uh yeah there was a little bit of a tip on my shoulder i i wanted to come back and and bring bring it back to what it was always meant to be and and fix that someone messed with your baby didn't they essentially you could you

could put it that way well like i said you attracted a lot of people and you've got a whole team built so i'm going to invite a few more of them into this chat here to talk to us and i've met a few of you but not all of you so i want to do a few introductions let us know who you are and what you do should we start with mustafa if you want to go first just because you're the first one almost hello everyone i am a software engineer over at ciq i am also a release engineering lead over at resf and i

mainly work on import build and packaging tools girl todd you're up next you're on mute at the minute todd let me see if i can only mute it there we go i got you there we go oh good morning andrew thank you for hosting uh where are you located i'm just curious i'm over in northeast england so it's um starting to get dark now it's uh just gone five o'clock excellent well good morning uh todd braley i'm vice president sales here at ciq cool robert you're next hey thank you andrew um my name is robert adolph and uh i'm a chief product officer and uh

yeah had some crazy ideas with greg when we uh started all this i'd love to hear them and amy last but not least hi my name is amy i've been with the support team for about a month now but i'm a long time uh linux user going back to the 90s and part of the reason why i came here is because of the impact greg had on my life with zen os and so i'm just really excited to be here supporting our customers and helping change other people's lives wow we should've got you to do the intro for greg that's a good and so there's

um the the letter c iq have been mentioned so i think it's it's useful for us to understand it a bit better the first i heard of ciq or control iq was i did a video and it mentioned rocky linux in it and it said that oh they don't have any support yet but it'll be interesting if that comes along and brim immediately sent me a message going two weeks just wait two weeks and he told me to google ciq which i did and my initial reaction was oh this must be some random managed service provider that specializes in linux into some kind of sponsorship

with rocky and then i looked at the people involved ceo greg kurtzer suddenly realized this is not just a random third party there's a lot more going on here so could you tell us about the relationship between rocky linux and caq and how does that work i'm happy to take it but i also don't want to hog the mic uh robert do you want to uh jump in yeah absolutely so um you know one of the things greg learned through all this was you know we did not want you know rocky linux to be part of ciq um so it was definitely something that it

is outside of and separate completely um completely for individuals as well as uh enterprises etc and we want to build that community that way ciq is uh actually a software development team that and support team that is building not just os's containers orchestration solutions built around performance workflows so we are separate completely but very dedicated to the project okay guys in the chat there i can see a few of you saying hello if you have any questions for the team feel free to drop them in the chat there can't promise we'll answer all of them but if you don't ask we definitely won't answer it

so go ahead and drop anything in there and i've got a few that were sent through ahead of time that we can chat through as well so and going back to you greg for a minute i'm going to force you to hog the mic for a little bit um you mentioned that there was a few things you would like to do differently this time there's there's stuff you've learned when you built centos and you know the journey you went through that and you've learned things or some things went well some things that didn't go well if you were to condense that all down to what

is the one biggest change you've made when doing rocky versus centos what would that one biggest thing be so the one biggest thing is really to create it uh to be more of a community um so sentos uh sento started off really as i said kind of as a as an offspring of of chaos linux but what what ended up happening was you know we didn't have ci cd we didn't have you know cloud wasn't huge at that point um or at least as big as it is now and and there was a lot of technology that was kind of missing from how do you

build an operating system so what we did is well each one of the developers basically grabbed their laptop and we started building packages and we started assembling the packages and as we did that and as we built these packages and assembled these packages you know we wanted to push these to mirrors and push them live and we did it from from our workstations we did it from our laptops which now means that the security umbrella um that we have to manage is basically confined within the circle of trust of the individuals doing that work and that meant that there's a the way you manage now

security is you keep those circles very small so the culture of centos was always very small as a development team there was probably about five plus or minus maybe two or three people um pretty much along the entire centos journey in terms of core developers people actually maintaining the operating system and this is why you saw things like you know and probably if you've been using centos for a while you probably recall there's a number of times in which you know centos centos releases were delayed or you know it just took a while to get a new new point release out or a new major

version out and it's because this small team of of individuals have been just i mean they've been floored with the amount of work that they had coming at them and they were they were doing this not as their primary job they were doing this as their as their alternative jobs right their weekend and in evening jobs which meant that um this poor team was completely overburdened and overloaded for about a decade or so until red hat actually um hired them but what would i do different i would want i wanted to first build up the infrastructure necessary so we can actually build a proper community

so we can actually expand beyond just a few developers and and that infrastructure is really key because it allows us to do things like allow other people to come into play to become developers become contributors without compromising the security or the integrity of the resulting project and that was what we spent the entire beginning of the project doing not building an operating system but building an infrastructure and it took us about four months to build the infrastructure what we did initially is we replicated what fedora is using and what we believe red hat is using so we can get the closest uh result to what

they have and then and then basically leverage that and build off of that for our own needs and again it took about four months to do that took about two months then from there to throw the operating system at that infrastructure and actually get the operating system built and then one month of testing and seven months later i think that's seven if i did my math right in my head seven months ish later uh we basically had a ga release cool so just to clarify you're not building it on your laptop anymore [Laughter] and another little known fact is um on a laptop i'm predominantly

a mac user so early versions of centos were actually done on a mac wow do we have to kick them off the stream now guys um they they already give me enough crap [Laughter] i'm sure they do i think we've had a question come through and it's it's very much crystal ball time so mystic knight wants to know if anyone's going to make a prediction on how many servers are going to migrate from centos to rocky over the next year so who's got a crystal ball to try and answer that one well i've got some data on this but it's just a it's it's a

little bit of actual data a little bit of a swag and then we can take a bigger swag and i i'm not sure if i can say this so uh um censor me if if i but a swag to me is a scientific wild ass guess so i'm going to take a swag at this um okay i i i can't work without so uh we've had about um i think it's about 600 700 000 downloads from our tier zero mirror if i remember the numbers right um at this point and we've been released now for what what has it been mustafa robert by three months

or so i think it's been somewhere around there um and uh so that's our tier zero mirror that's our kind of like top prime you know uh you know cdn mirror now we have about a hundred tier one mirrors we don't have visibility into how many downloads that those tier one mirrors are doing but based on some conversations we've had uh with with some of the hosts uh of these mirrors we think that our tier zero is basically satisfying roughly 50 of the downloads a lot then the tier one mirrors are basically doing the other 50 so if we've got let's just say three quarters

of a million ish downloads from our tier zero means we've got about 1.5 million downloads in about what three months for each download and in terms of conversations i've had personally with enterprises and individuals each download is obviously doing at least one install but most of them are doing multiple installs per per download so i think it's a pretty safe assumption to say we're probably over 1.5 million installs we're probably over 2 million installs we're probably over three million installs at this point but it's hard this is where the swag comes in it's a guess can't tell you exactly but that's in three months now

if you look at the graph of of downloads of course there was a gigantic spike on release and every time there's a press release or an event like this we get another spike but basically what ends up happening is it kind of is gonna it's it's tapering off an asymptote asymptoting to something that's going to be very um a much more sustainable kind of longer term um rate can't tell you exactly what that rate is because we're still super high we're still getting like we're still coming down from the from the major spike but it's still it's it's massive in terms of the number of

of users we have uh and now i'm going to approach it from the other direction 50 at some point centos was serving and installed on approximately 50 percent of uh the the enterprise servers that were reachable via netcraft and i remember a net craft survey being done a long time ago where centos was one of the primary dominant operating systems i went back tried to find these things couldn't find them i even went back through the archive trying to find these things and couldn't find them i don't know because i couldn't find it i i know it was there i know it was there but

here's here's what's happened over the past you know i'd say five-ish years uh since then is uh red hat took over took over centos and i don't mean to imply that that had a negative effect on it but in combination with that canonical was shifting their their business focus significantly and they did a fantastic job pivoting from laptops and workstations to cloud and enterprise and i got to hand it to them they did a really good job on this and as a result canonical and ubuntu usage started coming up and went up quite a bit the last survey that i saw was about 20 to

25 of um of enterprise and cloud resources is running centos today about 20 to 25 i think it's fair to say um at this point that they are the rocky's going to take over for a large percentage of that um can't say exactly what percent but i think that's fair to say i also think it's fair to say um that the and i'm gonna roughly call this something i know i talk a lot i just realized how long i've been talking i'm sorry um but uh i i think it's fair to say that if we consider enterprise servers um or or rather the enterprise linux

standard which is what previously was was red hat enterprise linux we look at this as now all of a sudden we have an enterprise linux standard right red hat is with with centos stream red hat pushed their development behind closed doors into the community which is a good thing now we kind of have this like enterprise linux standard up there and we've got many derivatives of that enterprise linux standard we've got uh obviously rocky we got red hat enterprise linux we got oracle we got almo there's a bunch of them out there there's navy there's um i know i'm forgetting at least a couple of

them at this point but there's a bunch of these is what i'm saying but if we count all of these as is together because they all should be compatible with each other we count all of these together as the enterprise linux standard i think together all of our all the open source projects behind this red hat and everybody if we work together i think we can actually start moving uh back towards 50 percent but we have to work together that would be uh very interesting to see yeah i mean 20 is still a good old chunk considering the size of the internet but yeah i'd

be interested to see if that climbs up higher i'm going to get my crystal ball out and make another prediction i think you're going to see a big spike in about two and a half months because i think there's going to be a lot of admins when that december deadline comes up are going to suddenly panic and start installing things so be interesting to see how that plays out if the graph suddenly shoots up again i think there'll be a few panicked installs on um new year's eve all seems to be the way um we have another question from again mr knight which i think

will lead into another interesting little chat so he's asking about the relation between rocky and other tech supported by ciq um so if can ask you to answer that then also tell us a little bit more about the support ciq offers for rocky as well because it's interesting to know because i think you're doing something a little bit different than most people you're not just aiming for large enterprises or you're looking to help the individual as well who's gonna take this one then andrew you you picked someone [Laughter] i picked someone i'm gonna pick on todd i told you i'm mute again okay maybe we'll

pick on someone else well todd's what's up there's audio oh there we go no andrew i can jump in if you want there we go uh no worries um so uh you know in the in the beginning and our theory on this was we wanted to support people right so whether that was a startup with you know one linux admin that just needed help or a hobbyist at home to you know some of the largest institutions in the world right so and then the other piece to that is um if you go to some of these large institutions there's all the way from complete maintainers

of the colonel themself all the way down to somebody that just came out of college themselves right so um we wanted to have a mechanism to do that obviously we also uh have a traditional mechanism for folks as well but we really wanted to focus on people and a lot of our model is built around how many people actually need support not necessarily the number of servers and sockets that are out there and amy's actually on our support team and maybe could give you a little bit of her perspective as well it's something i haven't seen done before when it was explained to me i

i had a moment where i was like well this is actually really cool and i think people are going to be interested in this essentially you get support through a person and not how many machines are responsible for i mean in the in the age of cloud i mean how many machines are you responsible for they come up they come down the existing model in some cases doesn't make sense for people for some people it totally does and we support that as well but this is a new model and we're seeing what people think about it so far the response has been really good and

and i'm excited about it personally i wish that this was something that had been offered to me in the past you know that my mic's working i'll add one more thing sorry about that guys so you know i mean i just started september first but you know we we were on a call with a uh a very large meat entertainment company doing a huge streaming service one of the big ones and and you know they they hadn't had a a huge centos shop thousands of nodes in different locations and uh it was problematic you know there were there were times to where you know their

production slas weren't met that sort of thing and uh you know just having that comfort level you know for an enterprise like that that's that's relying on you know a two-hour response time uh to get things done and open a ticket get it fixed immediately it just uh it brings a whole new level of support you know into what centos did not have before so um i i see that as a a major advantage of uh of what ciq is doing for support on any linux platform yeah and andrew just a as i'm looking at the question uh greg actually left off one part in

the beginning we were actually looking to build the next generation of high performance computing when we started off on ciq um and centos was going to be a main piece of that so it was something that mattered to us we were asked to support centos as part of that stack um so when this happened it actually affected our entire company as well because greg created a container solution for high performance computing and we've been working on an orchestration solution so we weren't just uh somebody that wanted to come out and support this you know we're a group of software developers that are trying to innovate

um this was something that affected us personally as well right and the customers that we were trying to serve so you know there was more to it than you know just you know coming out with an os if that makes sense so it pulled the rogue out from under you guys as well then so that's that's interesting then because i i've got it's actually running the centos 7 still i've got a server in my house so i need to upgrade that but you can imagine why i held off on it if i upgrade that to rocky can i as an individual if i've got a

problem i can just come and you'll help me without charging like full enterprise rates that's that's an interesting way to do things how do you set yourself up for that because back in the previous life i did you know dual public computer support and i work for a company that does enterprise computer well server support it's a very different ball game and you know the skills required are different the the pricing models are very different how do you do that in a way that is affordable for your users and all sustainable for you guys yeah big great question so you know we think about this

a lot so you know and amy kind of alluded to this every time we talked to somebody to come on our team we kind of asked them hey as an user of linux you know what would you have liked what would you want right so you know what we heard a lot was hey i want a really good capability to click and find my answer or interact with the chat and get my answer immediately or i call somebody once and they answer me right i get to an engineer i get to a software developer i get to somebody that understands and has been in my

spot before so it comes down to automation maybe some ai um some self-help that's maybe a little bit more advanced all the way up the stack of just getting better and repeating what comes up more and more um as something that people can access extremely fast but you know mustafa you know and amy might give you some you know perspectives from their angle as they're building this so i spend a lot of time on support forums just in general to see what type of problems people have and spoiler alert a lot of people have the same problems right so if you stay ahead of that

and you have the data and the information required to provide good solid answers and follow-ups you're going to help a very large number i'm not a statistician but it will be very helpful to capture that information and make it available all right yeah that's an interesting challenge i'll i'll be very interested to see how that goes and um how it develops in the future because it does feel like you're doing something that it's maybe not completely unique but it's certainly different like i don't know if anyone else is trying to address that same scale of you know massive large media company and me messing around

with the server in my house for storing my files on yeah i'll add one more thing andrew um you know what greg and rob have done really well with the sport team is that you know they've done a really good job of tying a lot of uh you know great rocky and centos engineers right into our support team so you know that's uh they're working in parallel together to to get these uh you know tickets open problems solved very very quickly so it's not just support people that's escalating from person to person it's you know when someone calls in they're they're typically getting an engineer

to talk to that's good you don't have like a level one where you know someone else you have you tried turning off and on again several times so um i'd like to jump in real quick with with another perspective too which is um the messaging for this is really powerful right because we're not i mean we as amy said we can count sockets we can count servers but um we don't like to it's not how we prefer to work we prefer to just support the people right we don't support servers we support people and people are what make up the companies right right the cut

the success of the company is 100 tied to the success of the the individuals and this brings everything right back down to the people and when you look at uh how we are approaching this to your point andrew we have a wide variety of support and sla options one of which for example starts exactly out of the enterprise with an individual and uh we were talking actually with several uh consultants um a little while ago and you know some people that are doing very specific things you know whether it be you know copy or whether it be um uh yeah you know website development or

or you know whatever that may be and they're they're they're they're single person shops right there they're their own consultants and they said you know it would be really great if if we if when we had a problem i know that we can contact you right i don't have an escalation point right it's me at this company if i can't solve a problem that i run into on a customer's you know server uh their web server or whatever i can't i i don't have an escalation path right i'm on google i'm on stack overflow trying to figure this this sort of stuff out it'd be

really great if we you know if i had some form of you know some manner to escalate and we're getting the exact same message now from very large it teams at very large companies they're like we also need some sort of escalation path if we can't solve something and when we're looking at this and again i don't want to count cloud instances i don't want to count um how many you know how many containers somebody is running or how many servers or virtual servers somebody is running i want to just support that person whatever that person is responsible for we are responsible for his that

person's success is our success and that's how we're looking at this it is very very people focused and people oriented and it also scales very nicely again you know not only do you have different slas for different levels of support but you also if we're supporting the you know the individuals a small company who may only have a few it people versus somebody like target or walmart who has hundreds of it people the model scales model scales actually very well so that's how we're approaching this we're looking at this from a completely different perspective and um and so far the response has been tremendous it's

it it almost feels so obvious i cannot believe this has not been done widely before it's an interesting way you put it about it's about the people one of the first things i got told in my current job which is about 15 years ago i've been there a while um i started off on a service desk doing it support for businesses and my boss took me aside when i did my first on-site visit and said the thing to remember is that it's about the people because i can be as good technically as you know doesn't matter how good technically you are if they don't like

you you're right that's just how it works like people don't buy you don't get repeat business based on just technical skills is about the feeling you give your customers it's how they feel about you that's what really matters at the end of the day and yeah it's something that stuck with me so that what you're saying resonates um i i'm starting to get worried about this mystic night guy because our gals don't actually know but um i think they've got some access to the notes i've written down for questions to ask you because they're asking a lot of the same things so this one's come

up about support resulting in patches so i had a question to ask you about what you do with things like bugs because as you support you know all these installations of rocky there's going to be a time where you come across an issue and you realize well this is a bug but it's not a rocky bug it's a red hat bug and if we fix it then we break that bug for bug compatibility with red hat and i was wondering if centos stream gave you guys some kind of opportunity to shorten that feedback back loop because that's all about you know getting stuff quicker upstream

so if you find a problem you go i can see how to fix it but i can't just put it into rocky source code or or i'm breaking stuff and i'm breaking that compatibility does stream give you an opportunity to get that further upstream so that you can then get it back down again into rocky and in the hands of your customers i know i've been talking a lot but i just have to jump in yes yes yes 100 thousand percent uh centos used to be behind red hat red hat was was developed behind locked doors right we did not have insight into what is

going into red hat with the exception of fedora but fedora was so far ahead that we didn't actually have direct insight as to what's going into red hat and um centos stream so centos was end of life centos stream was pushed in front of red hat which means red hat enterprise linux is now a child of centos stream it's a very weird kind of shift um from that perspective because it used to be a child of rel now all of a sudden its child is its parent a different name for it would have caused a lot less confusion i completely agree as a matter of

fact um i i talked to several of the red hatters uh for quite a while and we've debated over this uh and and they you know out of out of the debate i still i still agree with you that they could have used a different name and it would have been more valuable to use a different name but centos stream is an incredibly valuable positive uh direction for what what just shifted because it does now because it's in front of red hat enterprise linux we are now a downstream child of centos stream as is red hat enterprise linux now keep in mind though right now

red hat is the dominant player putting putting patches and printing putting fixes and whatnot into uh centos stream but over time we're hoping this is going to change even further to the point where this becomes this enterprise linux standard that i was talking about that repository that location needs to become more community focused and community oriented such that when we find bugs we now go and we can commit them um or or do a merge request into uh the centos stream get just like red hat does just like other projects are going to be able to this enterprise linux standard is so tremendous in terms

of what it is uh what it has just pivoted what has just happened is incredibly valuable if you were to have taken like you know the cards and and and just you know this is this is the hand we had last year and for the previous you know 16 years what we just did is we just got re-dealt an entirely new hand and it is um it is so much better than than where everybody was last year it is really it is a better move everything is landing in my mind in a better spot uh the the dominant uh compatible free alternatives to red hat

enterprise linux are now no longer being maintained by red hat they're being maintained outside of red hat which means red hat does not have to manage a conflict of interest situation where it was basically had they were giving away something that competed with their their their commercial product i want them to go make money on their commercial product but not at the expense of the free product and when you have that all in one company you do have to manage a conflict of interest uh so um going back just to the questions that that particular question from from mystic knight uh we would be pushing

patches and suggestions back up into that git and my goal my hope in this is that that would trigger conversations that would trigger discussions from not only you know not only from us and the community but from red hat from oracle from alma from everybody we can now all go into those discussions we can actually now make decisions collaboratively and this enterprise linux standard will then again start to um increase in terms of terms of prevalence in terms of domination even right i don't mean domination in a negative way just it just it it's that will help other people leverage it and i think it will increase the amount of utilization um well beyond 20 25 of all enterprise resources i think we can actually start moving back up towards 50.

that'll be great all those different distros essentially working together via this one point of communication if nothing else it's a bit weird to think of oracle and red hat working together but yeah so that yeah that's that's that's very positive i'm glad to hear somebody saying positive stuff about uh centos stream as well because a lot of the the comments i get about it on my channel are usually negative and that's quite often because people are positioning it as a drop-in replacement for normal centos what they now refer to as centos linux to differentiate it and it it doesn't do that it has a different

purpose that doesn't mean it's bad it's really good it's just different so it's good to hear that you know someone's saying you know this is why it's good that's that's really positive let me have a quick check through our quest and see what else we got coming in oh guess who mystic knight's going to the one about edge computing let me pop that on the screen what is the smallest rocky configuration planned at this stage would that be the minimal installers or something else i'm going to call on mustafa for this one we don't exactly have anything specific for edge computing right now but we

definitely think that it's important to start thinking about especially since all the big distros have started to advertise for edge computing specific resources but i think we need to work together with the other distros and centos stream maybe just look into another seg within the resf but it's something we definitely think about you did i just also add cigs the special interest groups um we have an embedded sig excuse me as well as a cloud sig we've got a container sig um i know that there are people currently doing raspberry pi work with with rocky and again all of this will hopefully come back into

special interest groups which will then become available uh to the uh to the community okay i'm glad you jumped in there because i forgot to unmute myself typical um that leads on to a question i got sent in by one of the viewers previously um i can't really pronounce neon without screaming because it was just ah that's all other name said um and i'm not just taking the piss out of it it literally just was lots of a's um but they had asked about whether there be any alternative spins of rocky like things like a kde version obviously this wouldn't be a bug for bug

with red hat type thing but that's an interesting idea are you looking at doing something like that with the special interest groups maybe yep exactly so special interest groups will offer a variety of different types of functionality we have a hyperscale one a cloud one as i mentioned an embedded one um uh there's a high performance computing one which as you can imagine i'm very interested in and uh so what we will be doing in these special interest groups is really oh there's a laptop sig there's a there's a workstation sig a legacy sig media entertainment seg rendering sig so all of these things are

being discussed right now now we decided that we were going to approach things a little bit more serially as as specifically as as it pertains to special interest groups and we decided we really wanted to focus on getting everything stable and solid before we start going down the direction of special interest groups and we have a couple more things that we're still um on our on our roadmap that we'd like to get done um the big one is secure boot so we're still working through secure boot trying to make trying to get that done um and i'll i know this wasn't one of your questions

um so far at least but it's it's really important for us obviously to get secure boot done we've got a lot of advocates requesting secure boot at this point we're in a queue like everybody else right now we're in a queue to get this done and it is just it is just a hold up on other other organizations um and we're just waiting for this for our turn in the queue basically um so once that gets done uh we're really going to start engaging very heavily in special interest groups mustafa has done a ton of work on build creating a build system or or updating

and modifying the build system such that we can very easily bring in special interest groups uh lewis on our team as well has uh the rocky team has also been doing a tremendous amount of work on this um and the infrastructure team so everybody we're all we're in preparation right now for the special interest groups but you are going to see fairly soon and we may decide just because we've been in a holding pattern for so long on secure boot uh we may end up deciding just to start moving forward with with special interest groups um you know and not wait for the secure boot

but you know give it another couple weeks and we're gonna kind of see how things play out but um uh i forget where i was going with this now um special interest groups oh builds there we go sorry sometimes i talk myself so long that i forget what my point was um so uh we are doing additional build so kde desktops and whatnot there has been a lot of interest in that but i'm gonna actually bring up something proactively which is there's also work that we're doing on the ciq side so all everything that i just mentioned is rocky side it's community now on the

ciq side uh one thing about our business model is we don't hold anything back the comm we want ciq to basically just help and facilitate rocky but we want rocky and the rocky enterprise software foundation to go in whatever direction that the community needs it to go in we're here to support the community we're not here to control or or influence the community in a negative way we just want to help it that's it so from that perspective um you know we don't want to hold it hostage now there is something that's coming down the pipe that we're working on right now which is a

commercial variant of it and it's commercial variant uh predominantly because we're working with a commercial partner that created some amazing security uh benefits and those those pieces are not uh open source right there it is a commercial company um the name of the company is run safe and they've been doing a phenomenal amount of work in terms of securing the applications that run on an operating system and what we're doing is we're actually going to be bringing that together and this is super important because as we're talking about additional builds and additional capabilities excuse me this is something now that we have engaged in a

in a direct partnership with them to bring that application level security to the entire operating system so imagine that and real quick what they are doing is they have found out a way to secure buffer overflows when i say secure buffer overflows i mean that uh most software uh hacks most not all but i'd say most are actually coming in through something called a buffer overflow which is a bug in software uh what they and what most people do then is they overflow this buffer they put in some malicious code in the overflowed section of the buffer and then they have a method to execute

that uh through the application what what runsafe has done is they basically they have mitigated against that such that buffer overflows via malicious or escalations and hacks via malicious code using a buffer overflow is no no longer possible they can actually protect against every buffer overflow so the code still has a bug don't get me wrong the code may crash but it will not result in exploited code it will not require or or malicious code being executed so um we're going to take that capability and roll that directly into the operating system so the whole operating system now is secured against buffer overflows and and

that's one of the major things that that we've got coming in and again that's going to be an entirely new build for rocky through ciq uh amy i saw you go off mute for a second you know i was just wondering i hate the blind side you hear but i'm just gonna do it um so how does that work with containers is that can you run run safe in a container yes extend the life cycle of a container yes it is because there's no it's not a kernel module even uh it exists entirely in user space so it doesn't protect against kernel exploits but it

does protect against application exploits and in many cases it's an application exploit that leads to a kernel exploit so this this another tool in the arsenal sounds like to me that's awesome exactly exactly and it's some really cool capabilities and yes it will work inside of a container so if you have a container that you have secured with rocky plus plus which is what we're currently calling this if you have a container that you've secured with that that you built your base container out of that it doesn't matter where you bring that container it's protected so you can run that on something that isn't rocking

it it it keeps the inherited protection exactly also also could i just jump in regarding spins because we actually have an official expense with uh with some workstation and kde images available in the repos but it's not well advertised we do talk about in in the matter most chat but but we're working on getting that official and advertised as well so we do have some spins already cool so ah if you're watching it's out there so that's that was a question i was going i was going to ask you about the run safe stuff because i saw the press release on it just after brim

had told me you were working on something that would make rocky unhackable so my question to brim was like have you ever heard of murphy's law you're calling things unhackable you didn't say that so you said it gets rid of a lot of um exploits but buffer overflows to be fair are a huge number of them like that is a lot um my understanding of it based on the a quick google of run safe and having looked through the material was it's kind of like address space randomization on steroids where if a dress fits randomization is taking a book ripping all the pages about it

and putting in a different order this is more like putting the entire thing through a shredder and say try reassembling that is that a fair way to think about it exactly it's a really good way of looking at it and every time you run the app you're in a different randomized address space so it doesn't follow like the application every single time it randomizes everything so it's always different so if someone was trying to like learn it and then the next time it loads it's it's different again yep cool yeah but the big thing too andrew that kind of draws to it it doesn't change

the app at all right so um at the same time it's it you don't have to write code to adjust what you're doing with your application oh are there any compatibility things that you need to plan for or just it just runs no it just runs api abi completely compatible um and no impact on performance now from again from a high performance computing perspective there's no so okay impact on performance it doesn't take much cpu time to quickly randomize the amount of stuff that it's randomizing it's incredibly it's incredibly quick um but uh but yeah it's uh it's fast once you once you've invoked the

application and you've had the bring up it runs at pure native speed right because once it's in memory then it's effectively done its job that makes sense brim's watching no bother i had to ruby on that one i mean you saw i'm sure you've seen what happened in video when they called something unhackable um about a week later they accidentally released um released uh unsecured version of it that wasn't locked down so it's it's a yeah you do that on the internet you suffer the consequences let's have a look see what else we've got in terms of questions from viewers then um there is one

from mystic knight i'm going to skip on that for now i'll come back to that i'm going to ask another one which was i'm gonna reword it a little bit so this one came in ahead of the stream and originally it was asking whether you guys would be building like a beta version of rocky from centaur stream and personally i can't see what the point of doing that would be and if you want to why would you rebuild a free rebuild of centaur stream when it's free anyway and the whole point is it's buying up to date so by the time you've rebuilt it it

won't be as up to date anyway but on the the idea of beaters how would you be doing that will it be red hat release the beta you rebuild it as a rocky beater or it'll be red hat release the final version and then you build a beta based on that for testing mustafa do you want to take that one i'm sorry i was a bit distracted uh could you just repeat the last bit sure so the the questions about rocky beaters will the the beta first a rocky nine will that be based on a red hat nine beta or will it be once red

hat to the final release of red hat nine you'll then build in your first build and that will be a beta that comes out okay so currently uh even though uh centos linux will be end of live red hat still uh publishes uh eight uh version of of all packages into a specific branch that we pull from and we have talked to some red hat engineers that uh even though they will be moving to get lab for for a roll nine and that they will continue or at least try to continue to maintain that that they will push to that branch so we will mostly

follow uh red hat or rail releases and when we get access to new versions within that branch which usually happens anytime red hat releases a beta or or a a new stable version uh we will use that if i understood the question correctly all right i think i think answers it yeah i should point out actually i just realized it looks like the stream is sponsored by guinness it's not i don't even like guinness but it's a good-sized mug so uh coming back to mystic night then the other questions well it's about the governance model for rocky and someone uh give us a quick rundown

of that sure i can i can jump into that so uh rocky started off basically being a as you said at the beginning uh andrew uh it was it was it was a comment to the blog post and and in addition to um to hey i'm thinking of starting a new you know a new version of this a new version of centos it also said come and join this slack and the slack was hbcng and hbcng is the is what robert mentioned earlier is the next generation of high performance computing that we've been really you know building out and within probably an hour all of

a sudden we started noticing more and more people joining like you know i first started with like you know 10 and 20 then 100 and then it was like hundreds of people joining at a time and uh in a very short amount of time you know we topped 10 000 people um all joining this slack now i promise this gets into governance almost everybody was asking me directly and messaging me in slack how can we help how can i help what can i do now normally having people offer to help is fantastic and it's it's helpful but having 10 000 people asking to help is

is and there's no way to properly address this because every person is you know being an individual who's offering to help and wanting to be part of this they don't want to be like they don't want to hear that they're just one of you know 999 other people all you know asked they you know so i'm trying to literally i was going through slack it would take me hours to go from one side of slack up at the top and just scroll down responding to everybody or trying to respond to everybody but we actually got to the point where uh the free version of slack

only allows 10 000 messages well when you have 10 000 people 10 000 messages does not last long so i actually knew in many cases i would by the time i got to the bottom names that i saw were no longer there because the messages i already dropped out of slack so i know that i missed i i lost a lot of stuff there was linkedin messages there was emails i mean everything was blown off the so what happened was and here's how the governance was the initial governance was created is people just started working people just said okay i want to do this part

and i'm going to start this some of those people just kind of kicked off certain initiatives other people started doing it other people started to help organize some of the things that were happening and um what what kind of naturally kind of create was created was a was a meritocracy an ideal meritocracy right we don't know who any of these people are in many cases they have their real name but in a lot of cases there was some sort of weird you know nickname that they were using there was most people didn't have time to put a picture or anything in there so it was

completely uh everything was based on a meritocracy and whoever just kind of naturally rose to doing different jobs and different things um all of a sudden we started to filter up with with a with a governance model or i should i should say not governance a management model now that management model identified a number of people who were kind of naturally leading certain initiatives within rocky and those people that were naturally leading those initiatives uh pretty much came out being kind of the the the head of the governance or the really what we called team leads but the team leads all filtered into um this

this this kind of board um initially we called it the board of advisors uh to to rocky but what we what we're doing right now and we're actually in the process of doing is really finishing uh solidifying that and formalizing that so we we have um resf board we're going to have rocky board and um and other sub boards uh in association with us and we're actually right now in the process of developing all of that and building all of that out so um long story to basically say we're working on it but hopefully it gives some additional context in terms of where we are

now how we got here and and hopefully where we're going with the fact that we're going to have boards um sub boards and and whatnot um an interesting uh uh perspective is um i'm actually going to at least at this point this may change but at this point i've been really pushing for i don't i don't even want to vote um i would possibly be a tiebreaker but um i want the i want the community to be uh to be voting on this i want the community to be running this so uh that's kind of how we're we're currently organizing this so if there is

something that ends up with a tie if we end up having a even number of of people i'll probably end up just being a a tiebreaker but that's the governance model that we're kind of moving into so hopefully all that makes sense yeah so a lot to work on there and i'm i'm just terrified of the thought of 10 000 messages coming through at once like my inbox is bad enough and i've been away on holiday for a couple of weeks i don't want to even imagine what that's like you must must be a good speed reader to try and even get fifty percent of

that um luckily it wasn't really at all ten thousand people it was probably more like just half of that just okay well perfectly manageable then all right if there's anyone wants any more questions pop them in the chat otherwise i'm going to start asking silly questions i mean they say there's no such thing as a silly question but whoever said that hadn't met me um let's see what we got okay current version of rocky linux is version 8.

that's a coincidence i've noticed is that there are also there's also another rocky series that has eight installments in a minute not so rocky balboa films so the ninth of them creed three is apparently in the works red hat are working in rel nine so the question is who's gonna get to rocky nine first oh that's a that's a good one um mustafa you want to take that i have no idea um we we are we are doing things um to prepare uh for that and um some of the beta code um and stream code is is already is already out there so and we're

gonna be playing with that um very soon and hopefully starting to get ahead of it i believe many many of the contributors have already actually started working on the started building packages and whatnot coming up with everything from build orders to um the tendencies that are needed and what not the goal hopefully is that you know the build system is going to do the dominant amount of the work um we've built up a lot of that infrastructure to be able to pull directly out of the get um the central stream get and be able to then just start building packages uh so and automating a

lot of that so that's our goal so the goal is to just have all of this start to automatically get kicked off and just chug through yeah and also also i'm just i'm just kidding we have we have more tools now than we than we did when we first started out when we first started out we we just had to it was just trial and error now we have specific tools for specific tasks regarding build orders regarding release orders uh all regarding even up to abi compatibility uh a level uh build order detection and repo categorization we have so much more now and we're still

working on making everything uh wholly automated regarding even though security errada we have that now too so uh we will have a better planning tools for for for a real nine and rocky nine uh when when real nine drops uh so i would think that it would be a much shorter pro process than what we did when we first started out all right good one we've got a question more serious question here from vrfxco um asking about what else you guys working on in ciq's you've talked about the the run save security thing which is really interesting but um any other little gems you can

let us in on robert do you want to take that yeah definitely andrew um so uh yes the answer to the question directly um there is definitely things we're thinking about inside of to fill customers needs when it comes to rocky itself however regarding security itself we we have a full cryptographic signing uh methodology that applies all the way from the os to deployments to the container to even a helm chart all the way to cloud deployments all the way to on-prem deployments that we have actually built into our entire stack um so you'll be seeing more and more about that here shortly um we're

working on some uh proof of concepts of some you know other uh partners that you could imagine would be a part of that ci process and also uh some big customers that have asked us for that supply chain security of software and so yes from the os perspective it matters to us but all the way up the stack to running apps running workflows orchestrating data with the workflows and getting people what they need so they can perform what they want to do is securing that entire supply chain is what we're trying to do i hope that makes sense i'm actually i'm actually gonna gonna drop

a drop a little bit of a bomb here um we haven't not publicly stated what we what we've been doing because we've been kind of running this in stealth um there's there's not even any hints on our website really about what we've been working on and what we were founded to do so i have a fully buzzword compliant description of what it is that we are we are doing and i'm going to give the buzzwords and everybody can kind of take it and take what they want out of it but what we have what we were founded to do is to modernize high performance computing

and in doing so we learned that we're not only creating hbc 2.0 we're creating something much much bigger and i would describe it as it is a cloud native in cloud hybrid federated meta orchestration platform for for doing all performance intensive workflows services and management of data there's my buzzword compliance string it is um it basically allows organizations to go on-prem multi-prem cloud multi-cloud hybrid cloud merge all of those together and edge and run jobs workflows services in a secure way as robert was describing complete supply chain security and integrity across all of those different environments and we can even encrypt it while it runs

yeah but andrew to bring it back to the topic of of your of the podcast um that doesn't happen unless you do a great job at the os level um and you know that the libraries don't go to the container you know it doesn't get cloud native without a great underpinning and a great base um and the resf team mustafa greg amy the rest of them have done just an incredible job and all the maintainers of linux period right so this is you know we're grateful to all of them so you know we're just trying to honor their work and give it back to the

community and something they can use that's productive for them i hope that i hope that makes sense yeah that's that makes a lot of sense all right i know you guys um originally had a calendar invitation for half an hour and we've just gone past an hour now so i'll i'll bring it to close with one last question and this one's for greg and it is possibly the most important question of your life so no pressure your the previous distribution you were known for sent to us and people have debated how to pronounce it over the years is it santos sent os or santos oh

my goodness is it pronounced gif or jif oh it's well i'm american i think it's gif uh i don't know if that is how to kick you off if you said jeff because that would clearly be wrong okay awesome fantastic no i i i would argue with the um the inventor of the gif uh format and say it's gif yeah and we thought his staff has to send a message he agrees as well all he said it's about the gif so we don't actually know what he means i thought you're gonna ask me what it how to pronounce sentos i know i i wanted to

just lead you up the garden path a little bit because i believe you you you rhyme it with mentos whereas i say centos so thanks for not shouting at me but i'm i'm stuck in my ways i'm not going to change well there's a really funny story about it um which is you know when the name was first proposed to me uh by by a contributor i said i love the name but whatever you do don't let's not make it like a union of two words i forget what that's called right now um i'm blanking on what that is but let's not make it a

union of two words let's not make it sent and then os because it is an os but the scent part makes no sense it's not a cent it doesn't cost anything even though cent is cheap it's an infinite increase from nothing so scent makes no sense sent os doesn't make sense to me but if you merge it centos sentos to me makes no difference but centos is the one that just to this day still gives me the the shivers but um but i and i used to be i used to actually correct people that shows you how immature i used to be i used to

actually correct people like it mattered in life right like no it's that's not how you do it this is how you do it and um but no now however anybody wants to wants to um and just to be clear andrew i was money corrected just want to point that out but i did it really nicely locked him in a dungeon until he agreed the last quick one before we go and i said that was the last one that was the last before the last one um vrfx just follow up to his previous question as asked when can they see that so your supply chain thing

it is um it is currently an active proof of concepts at customer sites now if people are interested in it um let them know where you heard about it because again we've we have been a very very limited release of this so far just because we wanted to really iron out any sort of bugs make sure the uis and usability features are all great but do reach out to us you can reach out to me directly robert you can reach out to info at uh control iq.com or ciq.co um either one of those goes to the same place but what we're doing is um so

far we are in proof of concepts we have people that are that are actively running this um and they are testing this we do expect this to go out to market in phase one of ga which is going to be um uh not fully federated it's going to be a cloud and multi-cloud but that'll be out by the end of this year uh probably november decemberish time period you're going to see some announcements around that we're hoping to align it with super computing but uh we may end up um either going a little bit before or after that but that's our goal okay also pretty

soon actually all things considered yeah and the last thing i would mention is uh first support i mean if uh if you're just interested in support you know you can contact myself or robert and we're happy to walk through um you know the different different support packages everything from from nodes to people so if there's listeners out there that are interested in in the in rocky support we're happy to uh jump on the phone and give you those different options cool just get in touch any questions give you guys a shout all right well i'm going to wrap it up there thanks very much i'll

let you get back to your your morning i suspect you all need another cup of coffee by now do i have a closing thought andrew you sure can greg not like not like i haven't talked enough throughout this whole thing um but uh i do have a closing thought which is um you know just to do what todd said if you do need any help if you need any support obviously absolutely contact us we we're here for you but i also want to say if anybody's interested in being part of what we're doing with rocky we are better organized now message me jump into the

matter most message me message mustafa jump into the different channels um and just join us uh we we it literally it is like a constant party going on in mattermost there's always stuff happening there's always discussions going on and um and don't even get me going when we actually have our online parties um they're actually really really fun and mustafa we it's time for us to i think have another one um here in just a little bit so um yeah it's a lot of fun um come and join with what we're doing uh be part of the community uh you're welcome to to be a

user we're you know and to leverage everything we're doing but come and join in it's it's fun yeah and robert has been working since i started five you know five five days a week covering us during the day and uh and a me at night so i get about two hours sleep so you're welcome to contact me during your working hours wait you're only working five days a week robert you're going way too easy on todd we're here for you guys yeah and also i'm almost in the same time zone as andrew so you can contact me as well cool all right thanks very much

thanks to all of you for coming on it's been great to chat to you um thank you hopefully other people have as well um i'll wrap that up and hopefully speak to you again soon so thank you you're a great host thank you good morning good night whatever is in your time zone thank you andrew bye