text
stringlengths
0
1.62k
[809.38 --> 814.66] the system's working containers are one of the things that that really enables that um and and that's
[814.66 --> 820.44] how we got started with docker so early on so help me out with this infrastructure so core os is a
[820.44 --> 827.06] linux distribution which is a kernel and some supporting software that only runs containers
[827.06 --> 833.40] and as long as i have a container that you can run you're going to keep that that underlying
[833.40 --> 838.96] infrastructure updated for me right my goal for core os is for you to not have to think about your
[838.96 --> 845.76] os anymore um i want i just think that when you know a patch for shell shock comes out and we have
[845.76 --> 850.08] all these individual ops teams around the world scrambling patching their servers that's just like
[850.08 --> 856.14] a redundant effort that the world doesn't need to do that my team of of os guys can centrally patch
[856.14 --> 862.36] the kernel and patch your open ssl vulnerability and deploy it to every server um that is sort of opted
[862.36 --> 867.78] into our platform um and and we can take care of it for you centrally and and the tools that we've
[867.78 --> 873.50] designed allow us to do it that way and do it you know kind of ultimately scalably as well uh we can you
[873.50 --> 879.92] know you we can run millions of servers this way um you know and i can have my my small team of
[879.92 --> 885.08] engineers developing the patches to do that so so what if i'm a user of the core os distribution but
[885.08 --> 891.84] not necessarily a customer of your guys's ongoing update service how would i go about uh managing the
[891.84 --> 899.32] the upgrades so just like you know the way ubuntu and red hat or fedora work um so our gift to the
[899.32 --> 904.40] world is that steady stream of updates so we don't commercialize the updates themselves in fact all of
[904.40 --> 909.10] our open source projects core os rocket fleet we don't directly commercialize it all um so like our
[909.10 --> 914.72] our gift is that we give you that steady stream of updates um you know free of charge it's almost a
[914.72 --> 919.10] community service um now again that's why we care a lot about our sustainability and why we sell
[919.10 --> 922.86] commercial products is to you know keep keep those efforts going nice
[922.86 --> 930.40] so of course you needed docker because you did not have the container environment when you guys
[930.40 --> 936.40] wanted to launch um is that fair to say yeah i mean if you think about docker sort of originally
[936.40 --> 943.30] it's it's a tool to download and run a container it's it's very much like apt-get or yum but instead of
[943.30 --> 950.12] an rpm or a .deb file you're you're downloading uh you know a container image and executing it uh and
[950.12 --> 954.54] that's it was our package manager effectively and still is our package manager today um but that
[954.54 --> 959.36] that's how we were using it another another component towards how you build a system that
[959.36 --> 964.52] you can automatically update another property that you want is you want any individual server to not
[964.52 --> 969.44] matter so you want to be able to pull the plug on any server and and your environment keeps running
[969.44 --> 974.26] now if you talk to your ops buddies they'll probably all agree that that's what they want too
[974.26 --> 978.94] um but if if you ask them if they can do it they'll probably say no and that's because it's too
[978.94 --> 984.90] difficult and and so that's why we started building kind of again at the lowest level a tool called etcd
[984.90 --> 990.14] which is a distributed key value store primarily intended for shared configuration among servers
[990.14 --> 995.90] as soon as you have more than one server you need to start sharing configuration um across those machines
[995.90 --> 1000.64] and we built that because we want to make it easier for people to build these distributed platforms
[1000.64 --> 1007.14] so you could run run things in this way um such that we can pull the plug on any machine and update it
[1007.14 --> 1012.72] any time without you taking a down time um and and uh really filling in the white space and etcd was
[1012.72 --> 1018.08] one of the first sort of areas of white space that that we saw uh needing to exist and etcd itself has
[1018.08 --> 1025.52] been adopted by i mean kubernetes mesos cloud foundry kind of like every cloud platform ish thing
[1025.52 --> 1032.44] that's emerging right now is chosen etcd is there an under underlying key value store nice yeah i'm looking
[1032.44 --> 1039.16] at the etcd github page now looks like you have 134 contributors over 4 000 commits uh this looks
[1039.16 --> 1045.22] like a really mature uh product so that's it's really awesome that's all open source and available
[1045.22 --> 1053.98] um and looks like it's pretty popular as well yep so you started off with docker and you're still you
[1053.98 --> 1061.06] still use docker to this day in your guys's core product um but there came a point in time where
[1061.06 --> 1067.08] the docker philosophy and perhaps the core os philosophy about how to do containerization
[1067.08 --> 1075.36] apparently those diverged can you tell us about that sure so docker originally came out it stopped it
[1075.36 --> 1081.22] talked a lot about this this standard container this idea that we could have a unit that's an
[1081.22 --> 1087.08] application that could be ran in many systems um and that you know what was decoupled from you know
[1087.08 --> 1091.46] particular implementation it was about like package in a container and you know stick the container on
[1091.46 --> 1095.56] a boat or stick it on a train and it all kind of works you know like using the shipping container
[1095.56 --> 1101.84] analogy um what's happened over time is docker is clearly on a path to be its own platform now
[1101.84 --> 1108.24] so while while docker started as a great tool for building a platform with and that's why we saw it
[1108.24 --> 1113.20] inside of kubernetes inside of amazon's cloud product inside of vmware's products because these are
[1113.20 --> 1118.72] existing platforms that wanted to add a container to it it's becoming a platform like those things i
[1118.72 --> 1123.42] just listed off now by adding its own clustering and everything and i get the product decision
[1123.42 --> 1127.84] there that's fine and i have no objections to it they should go and build their their platform that's
[1127.84 --> 1132.36] great i mean we'll probably build a platform at some time too you know um so it's a it's a fine
[1132.36 --> 1138.76] business idea um the the issue is we still want that standalone component that is not directly
[1138.76 --> 1143.38] commercialized at all that allows you to download and run a container essentially like a package
[1143.38 --> 1150.08] manager for containers to exist and so uh when it was clear that that you know docker was not investing
[1150.08 --> 1156.20] in things like standards around what a container is for interoperability or you know just getting some
[1156.20 --> 1162.28] basic security and composability issues right in the architecture of a unix tool um you know we said
[1162.28 --> 1167.76] hey it's easier for us to go build uh build a new thing that sort of serves the needs of what we want
[1167.76 --> 1172.82] uh you know versus you know send some pull request to docker that rewrites the project
[1172.82 --> 1183.68] and this uh culminated in a launch of a new tool along with um some specifications that you're trying
[1183.68 --> 1192.22] to get i guess formalized and community uh driven around uh containerization uh the tool we've
[1192.22 --> 1197.74] mentioned called rocket um the announcement for that came back early december and it seems to be
[1197.74 --> 1203.52] it caused a bit of a stir is that fair to say yeah you know that one got away from us a little bit
[1203.52 --> 1209.26] um if you look back on the whole thing the only messaging we've put out on rocket at all so far
[1209.26 --> 1215.58] is a blog post stating a couple technical reasons of why we built rocket all the press and all the
[1215.58 --> 1219.70] excitement and all the hacker news threads and everything was just fallout you know that the blog
[1219.70 --> 1224.92] post is is what stands there and if you read it from it with a technical lens you'll you'll see some
[1224.92 --> 1229.92] you know some very specific technical issues being addressed you know this is not politics nothing
[1229.92 --> 1235.58] we're just fixing some some technical issues and uh and all the fallout and docker's response and all
[1235.58 --> 1241.48] that stuff was just kind of extra extra uh stuff that got away from us uh and kind of our lesson learned
[1241.48 --> 1247.62] in all this is is uh you know people are watching people people people do are paying attention
[1247.62 --> 1252.18] putting words in your mouth too yeah i mean nobody can put words in my mouth because it's just written
[1252.18 --> 1257.18] on the blog post and if somebody is saying something that is not what's in the blog post then they're
[1257.18 --> 1262.68] just i guess you know making things up so it's probably an important point to say too that we're
[1262.68 --> 1267.58] not here to throw stones at anybody we're you know like i said before we got on this call like
[1267.58 --> 1273.58] and everyone who listens to the changelog knows that you know open source is hard right open source is
[1273.58 --> 1279.06] hard enough as it is without trying to call your buddy or your competitor or whomever you know to
[1279.06 --> 1284.18] a degree opposes whatever you're building bad or not right or whatever we're not here to do anything
[1284.18 --> 1289.24] like that whatsoever right um and i and i don't think you you definitely aren't because that's not
[1289.24 --> 1294.78] what you wrote and that's not the point um maybe to put a timeline on what's happened like this is
[1294.78 --> 1301.88] transposed over the last let's say i'd say 45 days december 1st was that original blog post and
[1301.88 --> 1307.40] uh the tech crunch article that put words in your mouth like fundamentally flawed was posted later
[1307.40 --> 1314.72] that same day so um what exactly happened inside of core west inside of the team when you released
[1314.72 --> 1321.16] rocket put out this blog post and what was sort of the i guess the press frenzy how did that impact
[1321.16 --> 1326.84] internally and did did anybody get how did the team react to it i guess is more or less what i'm trying
[1326.84 --> 1331.72] to ask i'm really proud of the team and how they reacted i mean there wasn't a lot for us to do
[1331.72 --> 1336.60] um you know we did get a lot of calls from press and such and we essentially just read the blog post
[1336.60 --> 1345.68] back to them um but did you read the post by the way yeah exactly so so we um that that the team didn't
[1345.68 --> 1351.66] really react or do anything there was a lot of sort of you know drama in air quotes on you know hacker
[1351.66 --> 1357.54] news and tech crunch and things like that but the team um you know even internally us deciding to ship
[1357.54 --> 1363.06] it um something that we had been working on for three weeks by the way um uh you know was a go
[1363.06 --> 1368.68] no-go based in technical merit you know my the engineers are not at all they're you know hardcore
[1368.68 --> 1374.76] open source developers and and we we don't want to create redundant efforts or or anything so we saw
[1374.76 --> 1381.14] a clear kind of white space and that white space was you know a a tool that was secure and composable and
[1381.14 --> 1387.60] and had an open standard um for running a container and and so we built the thing we wanted to exist
[1387.60 --> 1392.14] um for that and the team i think is stuck by it and we're continuing to invest in it and it's coming