On Thursday, March 24, 2016 at 9:07:41 AM UTC, Erik Bray wrote: > > On Thu, Mar 24, 2016 at 1:10 AM, Dima Pasechnik <dim...@gmail.com > <javascript:>> wrote: > > > > > > On Wednesday, March 23, 2016 at 4:50:10 PM UTC, Dima Pasechnik wrote: > >> > >> > >> > >> On Wednesday, March 23, 2016 at 3:33:11 PM UTC, Dima Pasechnik wrote: > >>> > >>> > >>> > >>> On Wednesday, March 23, 2016 at 3:06:46 PM UTC, Erik Bray wrote: > >>>> > >>>> On Wed, Mar 23, 2016 at 3:14 PM, Dima Pasechnik <dim...@gmail.com> > >>>> wrote: > >>>> > Hi Erik, > >>>> > > >>>> > would this work on Windows running in a VM? > >>>> > Ideally, for testing purposes, that's what one would prefer over > >>>> > having > >>>> > a dedicated box for testing. > >>>> > >>>> It should work. I really ought to be doing that myself but I've been > >>>> using a dedicated box (on the plus side that has revealed interesting > >>>> things that occur during *uninstallation*). I just don't have a > >>>> license handy by which I could easily make a Windows VM. > >>> > >>> > >>> let me try with Google Compute Engine if this will fly... > >> > >> > >> it does not work out of the box (I tried it on an GCE instance running > >> Windows Server 2012 R2 Datacenter) > >> the install bails out at the moment VirtualBox gets installed. > >> > >> It has left behind few things, one is Docker Quickstart Terminal. > >> It I try it I get: > >> > >> Running pre-create checks... > >> Error with pre-create check: "This computer doesn't have VT-X/AMD-v > >> enabled. Enabling it in the BIOS is mandatory" > >> Looks like something went wrong... Press any key to continue... > >> > >> I also got GIT GUI, Git Bash, and GIT CMD. (GIT GUI seems to work, I > did > >> not try others) > >> > >> Interestingly, I can install VirtualBox directly (just the default from > >> their website), although when I try installing a guest OS, it tells me > they > >> must be 32-bit. > >> (Does this mean that the host is 32-bit?) > > > > > > well, it probably won't work this way, at least according to a number of > > internet links, GCE cannot run nested VMs, at least not directly. There > is > > something by Oracle/Ravello that seems to be able to fill this gap. > > As well, it is meant to be in some versions of the upcoming Windows > update, > > beta of which is already released... > > Oh well, thanks for trying. In any case this isn't its intended use. > The point of this is for individual users to install sage on their > personal computers. > > (Incidentally if you wanted to run a docker container in a GCE machine > this can be set up easily using docker-machine: > https://docs.docker.com/machine/drivers/gce/, but that has nothing to > do with Windows). > > certainly it is not intended use, but we still need to think about reliable automated testing of this setup.
> Erik > > > >>>> But as long as you can run Virtualbox inside Virtualbox (which last I > >>>> tried works) then yes. > >>>> > >>>> > On Wednesday, March 23, 2016 at 11:15:58 AM UTC, Erik Bray wrote: > >>>> >> > >>>> >> Hi all, > >>>> >> > >>>> >> I've been working for a few weeks on an installer for Sage on > >>>> >> Windows, > >>>> >> which takes advantage of Docker to accomplish this.* The goal of > >>>> >> this > >>>> >> project is to make it possible to run Sage on Windows with as much > >>>> >> transparency as possible, such that the user isn't really aware > that > >>>> >> there is any virtualization involved. As you can read in my > report > >>>> >> for the OpenDreamKit project on Docker containers [1] there are > >>>> >> limits > >>>> >> to this. > >>>> >> > >>>> >> However, in the ideal case a user simply downloads and runs an > >>>> >> executable--clicks through a graphical install wizard, and then > gets > >>>> >> a > >>>> >> desktop icon which launches a Jupyter notebook (with sage and > >>>> >> terminal > >>>> >> support) in their default web browser. Although there are still a > >>>> >> few > >>>> >> rough edges [2] the alpha version of the Sage for Windows > installer > >>>> >> that I have for you today does just that: > >>>> >> > >>>> >> > >>>> >> > >>>> >> > https://github.com/embray/sage-windows/releases/download/v1a1/SageMath-7.0-1a1-fat.exe > > >>>> >> > >>>> >> My hope is for this to eventually be adopted into the SageMath > >>>> >> project > >>>> >> as the "official" distribution for Windows, replacing the existing > >>>> >> VM-based solution as I believe that this gives an overall > >>>> >> lighter-weight and more transparently "native" user experience. > In > >>>> >> the future the same approach could also be adopted--I think--to > >>>> >> provide a "local" installation of SMC. > >>>> >> > >>>> >> Now, if anyone with access to a Windows machine (Windows 7 or > newer), > >>>> >> it would be a big favor if I could get a few testers to bang this > >>>> >> around a bit and see what breaks and what works and what could be > >>>> >> improved. > >>>> >> > >>>> >> To be clear, right now it only supports running the notebook, > though > >>>> >> I'm also working on making it possible to run `sage` at a Windows > >>>> >> command prompt (almost working). Also be aware if you try to test > >>>> >> this: The biggest limitation for now (as described also in [1]) is > >>>> >> that for Docker on Windows hardware virtualization support is > >>>> >> required > >>>> >> to be enabled. If this is not enabled the most likely outcome is > >>>> >> that > >>>> >> the installer will fail with an error message like "Could not > start > >>>> >> Docker VM". In this case you will have to grub around in your BIOS > >>>> >> settings to find hardware assisted virtualization support--this of > >>>> >> course is going to be the most difficult aspect of making this > >>>> >> available to "average" users. A workaround may be possible but > I'm > >>>> >> not sure yet. > >>>> >> > >>>> >> Be aware also that the installer can take a few minutes to run (as > >>>> >> much as 5 minutes even on a reasonably fast machine) mostly due to > it > >>>> >> being highly compressed. > >>>> >> > >>>> >> Anyways, I look forward to your questions and feedback! > >>>> >> > >>>> >> Thanks, > >>>> >> Erik > >>>> >> > >>>> >> > >>>> >> * I'm aware of the irony that I only just recently chided someone > on > >>>> >> this list for appearing dismissive of working on native Windows > >>>> >> support for sage, while at the same time promoting a VM based > >>>> >> solution > >>>> >> for Windows :) Nonetheless I intend this only to be a temporary > >>>> >> solution, albeit a nicer solution than currently exists for > Windows. > >>>> >> > >>>> >> [1] > >>>> >> > >>>> >> > https://github.com/OpenDreamKit/OpenDreamKit/wiki/D3.1-Virtual-images-and-containers#sagemath > > >>>> >> [2] https://github.com/embray/sage-windows/issues > >>>> > > >>>> > -- > >>>> > You received this message because you are subscribed to the Google > >>>> > Groups > >>>> > "sage-devel" group. > >>>> > To unsubscribe from this group and stop receiving emails from it, > send > >>>> > an > >>>> > email to sage-devel+...@googlegroups.com. > >>>> > To post to this group, send email to sage-...@googlegroups.com. > >>>> > Visit this group at https://groups.google.com/group/sage-devel. > >>>> > For more options, visit https://groups.google.com/d/optout. > > > > -- > > You received this message because you are subscribed to the Google > Groups > > "sage-devel" group. > > To unsubscribe from this group and stop receiving emails from it, send > an > > email to sage-devel+...@googlegroups.com <javascript:>. > > To post to this group, send email to sage-...@googlegroups.com > <javascript:>. > > Visit this group at https://groups.google.com/group/sage-devel. > > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
