I've spent the last 3+ years asking people what they would like from a guest operating system in the cloud. Sometimes framed as "Why did you choose Fedora?", sometimes as "Why didn't you choose Fedora?", and sometimes basically the generic question.
I'd say that overall, the reason people say that they chose what they did was either familiarity, or that they found documentation — or another person — doing a similar thing, and they just followed whatever OS that had.
When I ask what they *want*, though, there's a somewhat different story. It's pretty universal, though: a small, simple base without much risk, and a library of components to go on top of that.
Fedora Cloud Base is a decent job of being a small base, although we still have a lot of dependency bloat and updates churn. But the library of stuff — languages, services — is difficult. We've got a great set of packages, but they're largely irrelevant, because the versions are usually changing too quickly. Mostly, you've got to bring your own stacks.
I'd hoped that we could answer this by slimming down the base and then offering a wide selection of SCLs on top. But, I don't think that's really panning out. The base is way less minimal than I'd like, and I don't know a good way to manage the updates situation. And SCLs are both still somewhat stuck *and* unlikely to explode (in the good sense) if they get unstuck.
For people who chose Fedora Cloud already — familiarity, or they found someone else familiar — we're probably okay. No one has anything negative to say about the work we've done — in fact, people who have chosen it generally say good things. I think it's very useful to keep producing Fedora Cloud Base for that group. But... it's a small club.
So, enter Atomic Host plus containers. This is, basically, exactly what people have been asking for. The ostree tech brings some order to the base, making updates more reliable and testable. And containers bring us the library of components — at the very least making it easier to bring your own, and ideally providing a new, better way for us to offer different versions, possibly with a different lifecycle.
That's why I'd like to move the Cloud Base image to a dedicated cloud.fedoraproject.org page along the lines of http://arm.fedoraproject.org, and replace Cloud with Atomic Host as a top level on https://arm.fedoraproject.org/, and to rename Cloud WG to Atomic WG (but still keeping the Cloud SIG to work on the Base image).
This is all just my 2¢, but I hope you'll consider them 2¢ with a lot of prior listening. If you have a counter story which will help us significantly grow adoption of Cloud Base *instead*, I'd love to hear it.
Matt,
Although I just downloaded the Fedora Cloud, I want to test it and, if it is what I am looking for, let some of my customers who live on DeskTone, give this a test drive from a fast thumb drive on a laptop or even a modified Chromebook. These people are attorneys and real estate professionals that need the dependability of the cloud without Redmond controlling how they use the vehicle to get there - something lean and mean.
If the team can get it smaller and it can access DeskTone, the universe may expand more quickly.
Bruce
RUWACH GROUP Integrated Technology Professionals __________________________________________________ Bruce Harrison, MSIS bfharrison@ruwachgroup.com www.about.me/bfharrison 540.226.0729 - Direct 877.338.9264 ext 700 Toll Free Other Numbers:
- 877.338.9264 option 1: Sales
- 877.338.9264 option 2: Support
- 877.338.9264 option 3:
On Oct 27, 2015, at 6:37 PM, Matthew Miller mattdm@fedoraproject.org wrote:
I've spent the last 3+ years asking people what they would like from a guest operating system in the cloud. Sometimes framed as "Why did you choose Fedora?", sometimes as "Why didn't you choose Fedora?", and sometimes basically the generic question.
I'd say that overall, the reason people say that they chose what they did was either familiarity, or that they found documentation — or another person — doing a similar thing, and they just followed whatever OS that had.
When I ask what they *want*, though, there's a somewhat different story. It's pretty universal, though: a small, simple base without much risk, and a library of components to go on top of that.
Fedora Cloud Base is a decent job of being a small base, although we still have a lot of dependency bloat and updates churn. But the library of stuff — languages, services — is difficult. We've got a great set of packages, but they're largely irrelevant, because the versions are usually changing too quickly. Mostly, you've got to bring your own stacks.
I'd hoped that we could answer this by slimming down the base and then offering a wide selection of SCLs on top. But, I don't think that's really panning out. The base is way less minimal than I'd like, and I don't know a good way to manage the updates situation. And SCLs are both still somewhat stuck *and* unlikely to explode (in the good sense) if they get unstuck.
For people who chose Fedora Cloud already — familiarity, or they found someone else familiar — we're probably okay. No one has anything negative to say about the work we've done — in fact, people who have chosen it generally say good things. I think it's very useful to keep producing Fedora Cloud Base for that group. But... it's a small club.
So, enter Atomic Host plus containers. This is, basically, exactly what people have been asking for. The ostree tech brings some order to the base, making updates more reliable and testable. And containers bring us the library of components — at the very least making it easier to bring your own, and ideally providing a new, better way for us to offer different versions, possibly with a different lifecycle.
That's why I'd like to move the Cloud Base image to a dedicated cloud.fedoraproject.org page along the lines of http://arm.fedoraproject.org, and replace Cloud with Atomic Host as a top level on https://arm.fedoraproject.org/, and to rename Cloud WG to Atomic WG (but still keeping the Cloud SIG to work on the Base image).
This is all just my 2¢, but I hope you'll consider them 2¢ with a lot of prior listening. If you have a counter story which will help us significantly grow adoption of Cloud Base *instead*, I'd love to hear it.
-- Matthew Miller mattdm@fedoraproject.org Fedora Project Leader _______________________________________________ cloud mailing list cloud@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/cloud Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
On Oct 27, 2015 7:24 PM, "Bruce Harrison" bfharrison@ruwachgroup.com wrote:
Matt,
Although I just downloaded the Fedora Cloud, I want to test it and, if it
is what I am looking for, let some of my customers who live on DeskTone, give this a test drive from a fast thumb drive on a laptop or even a modified Chromebook. These people are attorneys and real estate professionals that need the dependability of the cloud without Redmond controlling how they use the vehicle to get there - something lean and mean.
If the team can get it smaller and it can access DeskTone, the universe
may expand more quickly.
Bruce
You have me a little confused, Bruce. Fedora Cloud is an image intended to run in the cloud, to *build* services like DeskTone. You don't run it on bare metal, at least not without a lot of work - at which point, you may as well start with something different. Fedora might do what you're looking for - but what is that?
--Pete
On Tue, Oct 27, 2015 at 08:23:59PM -0400, Bruce Harrison wrote:
Although I just downloaded the Fedora Cloud, I want to test it and, if it is what I am looking for, let some of my customers who live on DeskTone, give this a test drive from a fast thumb drive on a laptop or even a modified Chromebook. These people are attorneys and real estate professionals that need the dependability of the cloud without Redmond controlling how they use the vehicle to get there - something lean and mean.
Hi Bruce. If I understand you, Fedora Cloud Base is *not* what you're looking for, although we may have something in Fedora. First, I know of several people doing their own "desktop as a service" based on Fedora, including at least one law firm. That might interest you as a possible replacement for Desktone, if that's what you're looking for.
But it sounds also like what you're looking for is a hosted file / calendar / etc sync-and-share server — what people think of as cloud services like Google Drive and etc. We have several things in the greater Fedora universe that might fit that bill, and running ownCloud — possibly on top of the Fedora Cloud Base image, or on Fedora Atomic — might indeed fill that need.
The Fedora Cloud edition downloads, however, won't give you this out of the box. (It might be something we want as a Fedora Server role in the future, though.)
I think that overall, this might be another reason for demphasizing "Cloud" as a Fedora edition. When we use the term, we mean it in the sense of cloud computing — on-demand self service, broad network access, resource pooling, elasticity, and measured service. And the Fedora Cloud image is just one part of the picture. It's an operating system meant to run inside an IaaS — infrastructure as a service — cloud provider, like Amazon EC2 or an OpenStack or Eucalyptus instance you configure yourself (see https://www.rdoproject.org/ or http://www8.hp.com/us/en/cloud/helion-eucalyptus.html). You could then use that to build up a Platform as a Service (on which you could run Software as a Service), or any other "XaaS" — including file storage and sync. But we don't currently have a turnkey solution for that.
But, maybe I'm misunderstanding you. Can you explain your needs a little more fully?
Matthew,
Thanks for taking the time to respond. I did not quite make myself clear. I saw this post today and it is more succinct:
Josh Boyer jwboyer@fedoraproject.org
Oct 29 (4 days ago) to jzb, Fedora, server, tamertas On Thu, Oct 29, 2015 at 9:16 AM, Joe Brockmeier jzb@redhat.com wrote:
On 10/28/2015 08:21 PM, Josh Boyer wrote:
The *could* be the same thing, except cloud-init is terrible and I hate it and if that was the single offering we had for some kind of C&S WG I would cry. I hate it because it is ridiculous to use in a non-cloud environment, and Server very much has that as part of it's reach.
Forking this thread briefly because I think this deserves its own discussion.
I apologize if my rambling wasn't clear on this point. Hopefully this tangent is short-lived.
Is your objection primarily to the concept of cloud-init or the implementation? If it's the concept, not much we can help with there. If it's the implementation...
Well, neither really. Admittedly my use of the Cloud images, and therefore cloud-init, was in attempted to boot it in a VM and log in more like a traditional install for simple test purposes. That didn't work and getting it to the point where I could log in required running some virt-tool thing to modify the image offline. So in the context of "Server & Cloud", where people expect to be able to log in after an install in many cases, cloud-init makes it really hard and is ill-suited to that kind of environment.
Specific to cloud environments, I have no idea if the hassle of getting it setup is the norm or worthwhile. I've been told it is, and I can see where having the infrastructure setup to provide the credentials already in place might make the hassle much less problematic.
(It is also quite possible I hit a bug in the cloud image. I tried running the local setup to provide cloud-init with ssh keys and it didn't work, hence the virt-tool thing. It has been a while since I tried again.)
We've talked about replacing cloud-init a few times in the past, but there are two objections:
- cloud-init is "standard" and we have an uphill marketing battle to get
our image adopted with something else.
- lack of a great alternative.
I completely believe both of these.
Mike has talked about a "rich boot process" previously, and I wonder if we're ready to start working on that?
I'm not sure what "rich boot process" means. I'd immediately interpret that as "a real init process" which to me means using systemd. Somehow I don't think that's what you're thinking... :)
Also, one of the CentOS GSoC projects was "Flamingo" "a lightweight contextualization tool that aims to handle initialization of cloud instances." [1] Maybe this is something we could look at for F24? CC'ing Tamer Tas, the student who worked on that. (It's targeted at being a cloud-init replacement for Atomic, so...)
That might be nice for "get rid of python" reasons. If it had cloud-init compatibility that would be even better, since people wouldn't need to migrate their provisioning infrastructure.
josh
RUWACH GROUP Integrated Technology Professionals __________________________________________________ Bruce Harrison, MSIS bfharrison@ruwachgroup.com www.about.me/bfharrison 540.226.0729 - Direct 877.338.9264 ext 700 Toll Free Other Numbers:
- 877.338.9264 option 1: Sales
- 877.338.9264 option 2: Support
- 877.338.9264 option 3:
On Oct 28, 2015, at 10:54 AM, Matthew Miller mattdm@fedoraproject.org wrote:
On Tue, Oct 27, 2015 at 08:23:59PM -0400, Bruce Harrison wrote: Although I just downloaded the Fedora Cloud, I want to test it and, if it is what I am looking for, let some of my customers who live on DeskTone, give this a test drive from a fast thumb drive on a laptop or even a modified Chromebook. These people are attorneys and real estate professionals that need the dependability of the cloud without Redmond controlling how they use the vehicle to get there - something lean and mean.
Hi Bruce. If I understand you, Fedora Cloud Base is *not* what you're looking for, although we may have something in Fedora. First, I know of several people doing their own "desktop as a service" based on Fedora, including at least one law firm. That might interest you as a possible replacement for Desktone, if that's what you're looking for.
But it sounds also like what you're looking for is a hosted file / calendar / etc sync-and-share server — what people think of as cloud services like Google Drive and etc. We have several things in the greater Fedora universe that might fit that bill, and running ownCloud — possibly on top of the Fedora Cloud Base image, or on Fedora Atomic — might indeed fill that need.
The Fedora Cloud edition downloads, however, won't give you this out of the box. (It might be something we want as a Fedora Server role in the future, though.)
I think that overall, this might be another reason for demphasizing "Cloud" as a Fedora edition. When we use the term, we mean it in the sense of cloud computing — on-demand self service, broad network access, resource pooling, elasticity, and measured service. And the Fedora Cloud image is just one part of the picture. It's an operating system meant to run inside an IaaS — infrastructure as a service — cloud provider, like Amazon EC2 or an OpenStack or Eucalyptus instance you configure yourself (see https://www.rdoproject.org/ or http://www8.hp.com/us/en/cloud/helion-eucalyptus.html). You could then use that to build up a Platform as a Service (on which you could run Software as a Service), or any other "XaaS" — including file storage and sync. But we don't currently have a turnkey solution for that.
But, maybe I'm misunderstanding you. Can you explain your needs a little more fully?
-- Matthew Miller mattdm@fedoraproject.org Fedora Project Leader _______________________________________________ cloud mailing list cloud@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/cloud Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Hello,
I am a systems engineer, managing a mid scale infrastructure in AWS and data centers (few thousand servers). While I am not entirely sure if this is the right thread to bring this point up or not, I would still like to share my concerns. Please forgive me, if this looks too far from the topic. For me, I often have to move away from Fedora Cloud or Server editions because of limited support cycle. I don't want to run a distribution in my production environment where security updates won't be available after a short amount of time. I am not sure if atomic updates or ostree would help here. I would love to run more of Fedora Cloud or Server instances if, at least, security updates are supported for a longer time. That said, I still have a fair number of instances running in production because Fedora often packages newer releases of tools and software which is required to run some of the applications. Aditya Patawari http://blog.adityapatawari.com/ https://fedoraproject.org/wiki/User:Adimania India
On Wed, Oct 28, 2015 at 4:07 AM, Matthew Miller mattdm@fedoraproject.org wrote:
I've spent the last 3+ years asking people what they would like from a guest operating system in the cloud. Sometimes framed as "Why did you choose Fedora?", sometimes as "Why didn't you choose Fedora?", and sometimes basically the generic question.
I'd say that overall, the reason people say that they chose what they did was either familiarity, or that they found documentation — or another person — doing a similar thing, and they just followed whatever OS that had.
When I ask what they *want*, though, there's a somewhat different story. It's pretty universal, though: a small, simple base without much risk, and a library of components to go on top of that.
Fedora Cloud Base is a decent job of being a small base, although we still have a lot of dependency bloat and updates churn. But the library of stuff — languages, services — is difficult. We've got a great set of packages, but they're largely irrelevant, because the versions are usually changing too quickly. Mostly, you've got to bring your own stacks.
I'd hoped that we could answer this by slimming down the base and then offering a wide selection of SCLs on top. But, I don't think that's really panning out. The base is way less minimal than I'd like, and I don't know a good way to manage the updates situation. And SCLs are both still somewhat stuck *and* unlikely to explode (in the good sense) if they get unstuck.
For people who chose Fedora Cloud already — familiarity, or they found someone else familiar — we're probably okay. No one has anything negative to say about the work we've done — in fact, people who have chosen it generally say good things. I think it's very useful to keep producing Fedora Cloud Base for that group. But... it's a small club.
So, enter Atomic Host plus containers. This is, basically, exactly what people have been asking for. The ostree tech brings some order to the base, making updates more reliable and testable. And containers bring us the library of components — at the very least making it easier to bring your own, and ideally providing a new, better way for us to offer different versions, possibly with a different lifecycle.
That's why I'd like to move the Cloud Base image to a dedicated cloud.fedoraproject.org page along the lines of http://arm.fedoraproject.org, and replace Cloud with Atomic Host as a top level on https://arm.fedoraproject.org/, and to rename Cloud WG to Atomic WG (but still keeping the Cloud SIG to work on the Base image).
This is all just my 2¢, but I hope you'll consider them 2¢ with a lot of prior listening. If you have a counter story which will help us significantly grow adoption of Cloud Base *instead*, I'd love to hear it.
-- Matthew Miller mattdm@fedoraproject.org Fedora Project Leader _______________________________________________ cloud mailing list cloud@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/cloud Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
On Wed, Oct 28, 2015 at 10:59:01PM +0530, Aditya Patawari wrote:
For me, I often have to move away from Fedora Cloud or Server editions because of limited support cycle. I don't want to run a distribution in my production environment where security updates won't be available after a short amount of time. I am not sure if atomic updates or ostree would help here. I would love to run more of Fedora Cloud or Server instances if, at least, security updates are supported for a longer time.
With Atomic, the idea would be that you'd basically just move to the newer host release seamlessly, because your containers will run in the same way.
That said, I still have a fair number of instances running in production because Fedora often packages newer releases of tools and software which is required to run some of the applications.
And, on the other side of the above, you can update your containers as fits your schedule, and mix and match e.g. CentOS and Fedora containers as appropriate.
On 10/28/2015 02:08 PM, Matthew Miller wrote:
On Wed, Oct 28, 2015 at 10:59:01PM +0530, Aditya Patawari wrote:
For me, I often have to move away from Fedora Cloud or Server editions because of limited support cycle. I don't want to run a distribution in my production environment where security updates won't be available after a short amount of time. I am not sure if atomic updates or ostree would help here. I would love to run more of Fedora Cloud or Server instances if, at least, security updates are supported for a longer time.
With Atomic, the idea would be that you'd basically just move to the newer host release seamlessly, because your containers will run in the same way.
That said, I still have a fair number of instances running in production because Fedora often packages newer releases of tools and software which is required to run some of the applications.
And, on the other side of the above, you can update your containers as fits your schedule, and mix and match e.g. CentOS and Fedora containers as appropriate.
Theoretically you should be able to move from f23 to f24 with an atomic update when f24 happens. And would still be able to reboot back into f23 if you had problems with f24.
On Wed, Oct 28, 2015 at 11:38 PM, Matthew Miller mattdm@fedoraproject.org wrote:
With Atomic, the idea would be that you'd basically just move to the newer host release seamlessly, because your containers will run in the same way.
This would be a _major_ feature for me and several others who want leading edge tools and software to run their code. Sounds great to me.
On 10/29/2015 12:43 PM, Aditya Patawari wrote:
On Wed, Oct 28, 2015 at 11:38 PM, Matthew Miller mattdm@fedoraproject.org wrote:
With Atomic, the idea would be that you'd basically just move to the newer host release seamlessly, because your containers will run in the same way.
This would be a _major_ feature for me and several others who want leading edge tools and software to run their code. Sounds great to me.
I am assuming with F23 atomic image release we are ready to tell users the same i.e. we should be able to move from F23 to F24 through rpm-ostree upgrade seamlessly.
However I don't see any official post about it. We should write a fedoramagazine post for this as this is a very important thing for Fedora atomic users.
Thanks, Lala
On 10/27/2015 06:37 PM, Matthew Miller wrote:
That's why I'd like to move the Cloud Base image to a dedicated cloud.fedoraproject.org page along the lines of http://arm.fedoraproject.org, and replace Cloud with Atomic Host as a top level on https://arm.fedoraproject.org/, and to rename Cloud WG to Atomic WG (but still keeping the Cloud SIG to work on the Base image).
This seems in line with what we agreed to a bit ago, just moving the needle slightly [1], right? I mean, we've already discussed moving focus to Atomic, so ... this seems like a further implementation detail?
[1] https://fedorahosted.org/cloud/ticket/117
cloud@lists.stg.fedoraproject.org
-
Aditya Patawari -
Daniel J Walsh -
Harrison, BF -
Joe Brockmeier -
Lalatendu Mohanty -
Matthew Miller -
Pete Travis