new fog appliance
-
Was thinking of making an ‘appliance’, basically just a ready built fog server vm… might just help people who want to get started quickly with fog? I’m just not sure if there’s any demand for it though.
Not sure what it would looks like… maybe some lightweight ubuntu build. Would people want a ‘ready to use’ version which you just turn on and then change things as you need like IP address etc? Or would it be better to maybe automate it all? Like when it logs on first time, run a script that downloads the latest trunk and starts the installer?
-
Considering I don’t use the actual Fog server for image storage I can keep the ‘appliance’ very small.
If I stored my images in my Fog VM it would mean that VM would need to expand to huge levels.Making NAS NFS storage more normal would make a VM Appliance easier.
If we keep storage inside the VM you will get massive arguments about how big the VHD should be.Installing Debian (My preference) and running the install script isn’t exactly hard.
Someone planning to image computers would need to know more than that to get running in the first place. -
@VincentJ said in new fog appliance:
If we keep storage inside the VM you will get massive arguments about how big the VHD should be.
Don’t think so… At least not when using non-pre-allocated image type.
-
@VincentJ said in new fog appliance:
@andyroo54
Installing Debian (My preference) and running the install script isn’t exactly hard.
Someone planning to image computers would need to know more than that to get running in the first place.I know what you are saying about it not being hard, to you, , but think about fogs target demographic. I mean I don’t have hard stats but most surely would be mainly windows sysadmins, that’s where I come from. To me, one of the things holding fog back from being more widely used is it’s ease of use in terms of setting it up, in terms of guides and say, an appliance.
I honestly think clonezilla shouldn’t really exist… why use that when you could use fog?
I think that honestly it is beneficial to be able to install a distro and configure from scratch, but I think as a kind of ‘gateway drug’ an appliance could be really useful. Most windows admins don’t have the time to get into the kind of depth you need to get fog really working well as it is at the moment. I don’t know… is that too harsh? That’s just how I feel. Not so much fog itself, but the guides could be better. I’m happy to help make them better.
I don’t want to make it seem like I’m bashing fog, I love it, and the devs who work on the software are amazing. I thought for a long time Fog was a lost cause but I think Tom Elliott kind of helped revive this, I can’t say I’ve followed it THAT closely but that’s what I’ve noticed anyway.
Maybe I went a bit off topic… ! I do agree I think about keeping the storage separate from the appliance, but like Sebastian said if you use non pre allocated type that kinda gets around it, right?
-
To have even less initial configuration, I’ve got a script that allows for a mobile FOG server, basically allows it to get an IP lease (with whatever IP) and then it sets all the relevant settings on the server to use that new IP.
I think that coupled with a simple DHCP reservation - which I would think most sys admins know how to do, we could create a ready-to-go VM.
-
@Jbob I would suggest CentOS 7 or Fedora 23 or whenever it comes out, 24.
CentOS 7 is starting to get a little old.
-
@Jbob What would you suggest for handling the CA and certificates on a ready-to-go VM?
Clearly we don’t want everyone out there with the same CA and certs. Perhaps a script that runs once on the next boot and also removes itself from running on boot as well?
-
Another thing to ask ourselves is how much HDD space would we want the VM to have?
Sure, it can be adjusted but the whole point of a ready-to-go VM is for people who haven’t dived into Linux to be able to setup FOG without issue.
I’d maybe suggest 250GB. But of course the actual file would be much smaller because of all the free space.
-
@Jbob You know what might be better is just running the installer (first time run to install fog) with the -y argument (to just go go go) on first boot… and of course have all the needed components pre-installed to make it faster.
-
@Wayne-Workman one thing to keep in mind about the installer is that if there is a schema update or first install it pauses. @Tom-Elliott is there an argument to prevent the schema update pause or auto update the schema?
Please help us build the FOG community with everyone involved. It's not just about coding - way more we need people to test things, update documentation and most importantly work on uniting the community of people enjoying and working on FOG! Get in contact with me (chat bubble in the top right corner) if you want to join in.
-
@Jbob it’s the -y argument. makes the installer do it for you, there are no pauses with the -y argument.
-
@andyroo54 said:
I honestly think clonezilla shouldn’t really exist… why use that when you could use fog?
Please think twice before saying this. I am not involved in clonezilla at all but we owe them A LOT (understatement, really!). FOG makes heavy use of things like partclone and would never been as good without it!!
Not so much fog itself, but the guides could be better. I’m happy to help make them better.
I totally agree!! Thanks for pointing that out. I started to work on the wiki but I get distracted by so many daily posts in the forums all the time and I wonder if we will see even more posts when we have that ready to got VM for dummy users ready - no offence, I am just curious! I am looking forward to the VM really!
Would be great if you get involved and help improving the documentation because this is very important I find. You are more than welcome. Please talk to Tom Elliott in a chat session to get a wiki account!! Please come back to me then and we shall talk about where to start. Thanks in advance.
-
@Jbob
./installfog.sh -y -
@andyroo54 said in new fog appliance:
I honestly think clonezilla shouldn’t really exist… why use that when you could use fog?
Yeah, I cringed at that statement too. Clonezilla definitely has it’s place in the cloning world. For instance, it’s the most simple way to take an image of a fog server! And also, Clonezilla is for people that don’t have a super-fast network and don’t have an extra computer laying around to be used as a server… or for a network where PXE booting doesn’t work for some reason or another. There are many reasons why Clonezilla is absolutely valid and why it’s needed.
You can’t compare FOG to Clonezilla. They both image, but it’s like comparing a 1/2 inch short-well socket to a 1/2 inch deep-well socket. Both sockets will turn a 1/2 inch bolt but each one is superior in different circumstances and spots.
-
Instead of a VM, could you create a netinstall iso that downloads and installs the OS for you and then launches the FOG installer.
Smaller download, runs on any VM that has internet access, lets the user set their own RAM, vCPUs, HDD and Networking.
You will NEVER please everyone… for me a 250GB VHD would be way too big… for someone else it would be way too small…
2vCPU - reasonable for most, but my home hypervisor wouldn’t run it properly as it’s only on a celeron. 1vCPU, could we run into issues where there isn’t enough CPU? (Neither would a HP Microserver)
1GB RAM - should run OK… but someone will moan that their low end PC ran out of RAM running it… Someone else will moan that it should have more…
In order to run FOG, they need to change their DHCP, You shouldn’t force this onto the FOG server because it could lead to them effectively killing their network when their DNS doesn’t work. They need to know what they are doing to some degree or we will just end up with angry people who don’t have a clue moaning on the forums, giving FOG a bad name.
There is no replacement for knowing your own environment and how it works. We should emphasize that this is not ‘easy peasy’ and will require some level of knowlegde.
-
@VincentJ said in new fog appliance:
There is no replacement for knowing your own environment and how it works.
+1,000,000
-
@VincentJ said:
… could you create a netinstall iso …
Although I kind of like the idea I refuse to go this way. I have looked into building bootable ISOs lately and it’s a very dark place!!! Some people have 32 bit CPU some 64 bit (yes even in VM!) and then some use BIOS and others UEFI which is even more hell then the 32/64 bit issue!! I don’t wanna answer those hundreds of “ISO does not boot on my XYZ” questions.
Sure we won’t be able to create a VM that suits everyone. But except the disk size (which we can make 512 GB - non pre allocated) all the other things you mentioned can easily be adjusted by really any person being able to download and start a VM! So I don’t see the point of digging into the ISO mud…
-
Another thing to point out here is as soon as 1.3.0 is released, I’m going to make full-length installation videos for all the popular distros of Linux. I will explain the things I do in the video and link to the written Wiki article that it follows (which I will also create).
We might not need pre-made VMs if the documentation was more noob-friendly.
But that said, I still think making a VM would be great fun and… why not? lol.
-
So now we’ve gone from 250gb to 512gb…
Next we’ll get the angry person who says their storage broke all their other VMs because they overallocated and filled things up.
Saying that the other things can be adjusted… The people who don’t know what they’re doing wont know to change them and will manage to break things somehow.
Happy users might tell a couple of others… Unhappy users are generally much more vocal.
-
@VincentJ Then we will just make the VM 8GB and make it mandatory for everyone to expand the /images partition.
