Re: Installer: 32 vs. 64 bit
On Fri, Nov 09, 2018 at 05:31:00AM +, Chris Knadle wrote: > A logical place to check or the lack of BIOS virtualization features and show > an > error message for this would be within the .postinst script for the virtualbox > package in Debian. This way when Virtualbox is installed the user installing > it > can be warned that VT-x or AMD-V isn't active and give a hint as to how to fix It's pretty easy to miss messages in postinst scripts; sometimes a few hundred go flying past and some of them print information that is noise at best. Something this important probably ought to be a message in the GUI somewhere, since people expect to interact with VirtualBox via GUI. The cpu-checker package in Ubuntu might be worth stealing. The kvm-ok script runs a handful of small checks to try to diagnose if KVM-based virtualization acceleration will work or not: $ kvm-ok INFO: /dev/kvm exists KVM acceleration can be used https://launchpad.net/ubuntu/+source/cpu-checker https://launchpad.net/cpu-checker Thanks signature.asc Description: PGP signature
Re: Installer: 32 vs. 64 bit
> Or an user error. In either case, I don't get what a 32-bit _x86_ virtual > machine would be good for. Are you teaching some code archeology? Not at all. We're trying to make it compulsory for first year students to have a Unix installation on their personal machine. In practice, this means any of native Linux, native Mac OS, or virtualised Linux. (We've found Cygwin to be confusing, and we haven't looked at WSL.) Since we're trying to get this to scale across a few hundred eighteen-year olds (smart ones, thankfully), we're seeing all sorts of user errors. -- Juliusz
Re: Installer: 32 vs. 64 bit
> Filing a bug on src:virtualbox with severity 'wishlist' or 'normal' for this > issue to discuss it with the maintainer of the virtualbox package(s) seems a > logical thing to do. Unfortunately, we're speaking about running Debian under VirtualBox under Windows, so it would need to be something that happens in VirtualBox upstream.
Re: Installer: 32 vs. 64 bit
On Fri, Nov 09, 2018 at 01:40:59PM +0100, Adam Borowski wrote: If his students were doing code archaeology or deep embedded, such areas require enough base skills that getting spooked by 32 vs 64 bits would be beyond them. Everyone starts somewhere, even code archaeologists. At my former School a Lecturer taught Operating Systems by getting the students to write kernel modules for MINIX on a VM (networking iirc). I'm sure the majority of those students would get horribly tied up in 32 vs 64 issues like this if they experienced them; despite that, they broadly succeeded in writing the kernel modules and passing the course. -- ⢀⣴⠾⠻⢶⣦⠀ ⣾⠁⢠⠒⠀⣿⡁ Jonathan Dowland ⢿⡄⠘⠷⠚⠋⠀ https://jmtd.net ⠈⠳⣄ Please do not CC me, I am subscribed to the list.
Re: Installer: 32 vs. 64 bit
On Fri, Nov 09, 2018 at 11:36:36AM +, Jonathan Dowland wrote: > On Fri, Oct 26, 2018 at 10:41:32PM +0200, Adam Borowski wrote: > > Or an user error. In either case, I don't get what a 32-bit _x86_ virtual > > machine would be good for. Are you teaching some code archeology? Do you > > want to prepare 32-bit images for something deeply embedded? Neither sounds > > an activity fit for your students. > > I'm not sure we are necessarily the experts in what is a fit activity > for this teacher's students. If his students were doing code archaeology or deep embedded, such areas require enough base skills that getting spooked by 32 vs 64 bits would be beyond them. Less variants -> less confusion -> less pain for the teacher. There are architectures where running a 32-bit VM might be a worthy use of your time, but it's not the case here. > > For anything else, you want an amd64 kernel, possibly running i386 or x32 > > code. > > IMHO there are a remarkably small number of situations where x32 would be a > sensible suggestion. As a lapsed x32 porter, I agree. But it's still more sensible than i386. Meow! -- ⢀⣴⠾⠻⢶⣦⠀ Have you heard of the Amber Road? For thousands of years, the ⣾⠁⢰⠒⠀⣿⡁ Romans and co valued amber, hauled through the Europe over the ⢿⡄⠘⠷⠚⠋⠀ mountains and along the Vistula, from Gdańsk. To where it came ⠈⠳⣄ together with silk (judging by today's amber stalls).
Re: Installer: 32 vs. 64 bit
On Fri, Oct 26, 2018 at 10:41:32PM +0200, Adam Borowski wrote: Or an user error. In either case, I don't get what a 32-bit _x86_ virtual machine would be good for. Are you teaching some code archeology? Do you want to prepare 32-bit images for something deeply embedded? Neither sounds an activity fit for your students. I'm not sure we are necessarily the experts in what is a fit activity for this teacher's students. For anything else, you want an amd64 kernel, possibly running i386 or x32 code. IMHO there are a remarkably small number of situations where x32 would be a sensible suggestion. -- ⢀⣴⠾⠻⢶⣦⠀ ⣾⠁⢠⠒⠀⣿⡁ Jonathan Dowland ⢿⡄⠘⠷⠚⠋⠀ https://jmtd.net ⠈⠳⣄ Please do not CC me, I am subscribed to the list.
Re: Installer: 32 vs. 64 bit
Paul Wise: > On Fri, Nov 9, 2018 at 1:32 PM Chris Knadle wrote: > >> A logical place to check or the lack of BIOS virtualization features and >> show an >> error message for this would be within the .postinst script for the >> virtualbox >> package in Debian. This way when Virtualbox is installed the user >> installing it >> can be warned that VT-x or AMD-V isn't active and give a hint as to how to >> fix >> it. Alternatively a /usr/share/doc/virtualbox/README.Debian file could >> contain >> a warning about this for the user to read, which assumes the user knows to >> look >> for that. [I checked -- right now the virtualbox source package in Debian >> contains neither AFAICT.] > > I think printing errors like this in the postinst is unlikely to > attract the attention of users who such a warning might be targetted > at. Yes I'm also concerned about that; we (Debian Developers) have been discussing that issue for a long time. A Debconf prompt is probably better, but it's also more invasive and those are meant to be avoided unless necessary. And in this case the issue started with a Debian VM running in VirtualBox on a Windows host. *shrug* > Even for the extra-technical ones who do read their apt logs > judiciously (or have a script to do that for them), it also doesn't > fix the situation when folks are running Debian VMs in VirtualBox > running on other platforms. > > So probably the VirtualBox UI should be indicating the lack of > VT-x/AMD-V and inducing users to reboot and turn that on in their boot > firmware. Looking at the code, I think this is already the case. In the source package for VirtualBox in Sid I see text for warnings related to this in translation files such as src/VBox/Frontends/VirtualBox/nls/VirtualBox_eu.ts; however they don't appear in src/VBox/Frontends/VirtualBox/nls/VirtualBox_en.ts. The VirtualBox_en.ts and qt_en.ts files are significantly shorter than other translations, but I think that's due to the English translation being built-in. -- Chris -- Chris Knadle chris.kna...@coredump.us
Re: Installer: 32 vs. 64 bit
On Fri, Nov 9, 2018 at 1:32 PM Chris Knadle wrote: > A logical place to check or the lack of BIOS virtualization features and show > an > error message for this would be within the .postinst script for the virtualbox > package in Debian. This way when Virtualbox is installed the user installing > it > can be warned that VT-x or AMD-V isn't active and give a hint as to how to fix > it. Alternatively a /usr/share/doc/virtualbox/README.Debian file could > contain > a warning about this for the user to read, which assumes the user knows to > look > for that. [I checked -- right now the virtualbox source package in Debian > contains neither AFAICT.] I think printing errors like this in the postinst is unlikely to attract the attention of users who such a warning might be targetted at. Even for the extra-technical ones who do read their apt logs judiciously (or have a script to do that for them), it also doesn't fix the situation when folks are running Debian VMs in VirtualBox running on other platforms. So probably the VirtualBox UI should be indicating the lack of VT-x/AMD-V and inducing users to reboot and turn that on in their boot firmware. -- bye, pabs https://wiki.debian.org/PaulWise
Re: Installer: 32 vs. 64 bit
Juliusz Chroboczek: >> When discussing virtual machines it would be helpful to mention which virtual >> machine hypervisor is being used, because the resulting behavior can differ >> depending on hypervisor. > > It was VirtualBox under Windows. The underlying issue was that VT-x was > disabled in the BIOS, and hence VirtualBox didn't offer any 64-bit > machines. The student tried her best to make it work, I don't think she > can be blamed for failing. > > -- Juliusz Okay -- thanks for taking the time to find the root cause of the issue. Unfortunately VT-x being disabled in the BIOS is not an abnormal situation, so it's not any kind of "error" per se that normally needs to be reported. This is a feature specific to virtualization and many end-user machines aren't used that way -- and in my experience the virtualization features ship default to being turned off. A logical place to check or the lack of BIOS virtualization features and show an error message for this would be within the .postinst script for the virtualbox package in Debian. This way when Virtualbox is installed the user installing it can be warned that VT-x or AMD-V isn't active and give a hint as to how to fix it. Alternatively a /usr/share/doc/virtualbox/README.Debian file could contain a warning about this for the user to read, which assumes the user knows to look for that. [I checked -- right now the virtualbox source package in Debian contains neither AFAICT.] Filing a bug on src:virtualbox with severity 'wishlist' or 'normal' for this issue to discuss it with the maintainer of the virtualbox package(s) seems a logical thing to do. -- Chris -- Chris Knadle chris.kna...@coredump.us
Re: Installer: 32 vs. 64 bit
If VT-x is disabled, the virtual machine will be sluggish, so if it works, it'll be a bad experience. Don't do that. On Thu, Nov 8, 2018 at 11:04 AM Juliusz Chroboczek wrote: > > > When discussing virtual machines it would be helpful to mention which > > virtual > > machine hypervisor is being used, because the resulting behavior can differ > > depending on hypervisor. > > It was VirtualBox under Windows. The underlying issue was that VT-x was > disabled in the BIOS, and hence VirtualBox didn't offer any 64-bit > machines. The student tried her best to make it work, I don't think she > can be blamed for failing. > > -- Juliusz > -- -- Best Regards. This is unedited. This message came out of me via a suboptimal keyboard.
Re: Installer: 32 vs. 64 bit
> When discussing virtual machines it would be helpful to mention which virtual > machine hypervisor is being used, because the resulting behavior can differ > depending on hypervisor. It was VirtualBox under Windows. The underlying issue was that VT-x was disabled in the BIOS, and hence VirtualBox didn't offer any 64-bit machines. The student tried her best to make it work, I don't think she can be blamed for failing. -- Juliusz
Re: Installer: 32 vs. 64 bit
>> I've been encouraging my students to install Debian on their personal >> machines, and we've found out that a lot of them get the wrong Debian >> installer: >> >> - some of them attempt to install an AMD64 version of Debian in >> a 32-bit-only virtual machine; > Why are they creating 32-bit virtual machines? Perhaps this is a bad > default in the VM manaager? Yes, it was a bad default in the laptop's BIOS (VT-something was disabled). HP's bug, obviously, but it would have made for a better experience if the Debian installer had complained in a comprehensible manner straight away. >> - others attempt to install an i386 version on 64-bit hardware. > This should work, in general. It won't work on a 64-bit system that > only supports EFI boot - and the installer won't be able to report > that, unless it includes a dummy 64-bit EFI program just to do that. It was a recent laptop, and the user had manually enabled BIOS boot. The installer hung with a black screen just after the initial menu. > We should not do in this in the second case, since it is supposed to > work. (But a warning might be reasonable.) Please. -- Juliusz
Re: Installer: 32 vs. 64 bit
> This is not what I get. > - 32bit debian on 64bit machine: this should be working fine > - 64bit debian on 32bit machine: I get the attached message > If it's not what they get, there is some bug and more investigation is > needed. I no longer have access to their machines, so I'm unfortunately unable to check. -- Juliusz
Re: Installer: 32 vs. 64 bit
Why are they creating 32-bit virtual machines? At least with virtualbox 32-bit VMs can run on any host. 64-bit VMs require VT-x which is all too often disabled in the BIOS.
Re: Installer: 32 vs. 64 bit
Le vendredi 26 octobre 2018 à 14:41:31+0200, Juliusz Chroboczek a écrit : > […] > Could somebody please speak with the installer people so they make sure > that the installation fails with a friendly user message in both of the > cases outlined above? Regardless of the discussion, you can, and should! It's just about filing a bug on the bug tracking system. As you're a long date user of Debian, I assume you know that, but it'd be a great idea to submit such a bug, even if you wish to bring the discussion on -devel, instead of just sending this for a discussion in devel. -- Pierre-Elliott Bécue GPG: 9AE0 4D98 6400 E3B6 7528 F493 0D44 2664 1949 74E2 It's far easier to fight for one's principles than to live up to them. signature.asc Description: PGP signature
Re: Installer: 32 vs. 64 bit
On Fri, Oct 26, 2018 at 08:17:11PM +0100, Ben Hutchings wrote: > On Fri, 2018-10-26 at 14:41 +0200, Juliusz Chroboczek wrote: > > I've been encouraging my students to install Debian on their personal > > machines, and we've found out that a lot of them get the wrong Debian > > installer: > > > > - some of them attempt to install an AMD64 version of Debian in > > a 32-bit-only virtual machine; > > Why are they creating 32-bit virtual machines? Perhaps this is a bad > default in the VM manaager? Or an user error. In either case, I don't get what a 32-bit _x86_ virtual machine would be good for. Are you teaching some code archeology? Do you want to prepare 32-bit images for something deeply embedded? Neither sounds an activity fit for your students. For anything else, you want an amd64 kernel, possibly running i386 or x32 code. Heck, even an i386 kernel would run fine on a 64-bit capable VM. > > - others attempt to install an i386 version on 64-bit hardware. > > This should work, in general. It won't work on a 64-bit system that > only supports EFI boot - and the installer won't be able to report > that, unless it includes a dummy 64-bit EFI program just to do that. Installing i386 is also an user error unless you know you specifically need that. It's less egregious than a 32-bit VM, but still should be strongly discouraged. It kind of worked before, but then we got melted spectrum mitigations -- new CPUs get a lot more to mitigate, and that really sucks on 32-bit kernels on 64-bit hardware. Just don't. i386 kernels should be used only if you actually have 32-bit hardware or help someone who does. You may still run 32-bit userland if you so wish. Meow! -- ⢀⣴⠾⠻⢶⣦⠀ ⣾⠁⢰⠒⠀⣿⡁ 10 people enter a bar: 1 who understands binary, ⢿⡄⠘⠷⠚⠋⠀ 1 who doesn't, D who prefer to write it as hex, ⠈⠳⣄ and 1 who narrowly avoided an off-by-one error.
Re: Installer: 32 vs. 64 bit
On Fri, 2018-10-26 at 14:41 +0200, Juliusz Chroboczek wrote: > Hi, > > I've been encouraging my students to install Debian on their personal > machines, and we've found out that a lot of them get the wrong Debian > installer: > > - some of them attempt to install an AMD64 version of Debian in > a 32-bit-only virtual machine; Why are they creating 32-bit virtual machines? Perhaps this is a bad default in the VM manaager? > - others attempt to install an i386 version on 64-bit hardware. This should work, in general. It won't work on a 64-bit system that only supports EFI boot - and the installer won't be able to report that, unless it includes a dummy 64-bit EFI program just to do that. [...] > Could somebody please speak with the installer people so they make sure > that the installation fails with a friendly user message in both of the > cases outlined above? We should not do in this in the second case, since it is supposed to work. (But a warning might be reasonable.) Ben. -- Ben Hutchings The obvious mathematical breakthrough [to break modern encryption] would be development of an easy way to factor large prime numbers. - Bill Gates signature.asc Description: This is a digitally signed message part
Re: Installer: 32 vs. 64 bit
Hello, Juliusz Chroboczek, le ven. 26 oct. 2018 14:41:31 +0200, a ecrit: > In both cases, the installer crashes with no useful error message This is not what I get. - 32bit debian on 64bit machine: this should be working fine - 64bit debian on 32bit machine: I get the attached message If it's not what they get, there is some bug and more investigation is needed. Samuel
Re: Installer: 32 vs. 64 bit
On Fri, Oct 26, 2018 at 02:41:31PM +0200, Juliusz Chroboczek wrote: > I've been encouraging my students to install Debian on their personal > machines, and we've found out that a lot of them get the wrong Debian > installer: > > - some of them attempt to install an AMD64 version of Debian in > a 32-bit-only virtual machine; > - others attempt to install an i386 version on 64-bit hardware. > > In both cases, the installer crashes with no useful error message (in the > former case, it crashes just after installing grub, in the latter case, it > crashes straight away). I don't think an i386 version of anything can crash on 64-bit hardware for hardware reasons. Are you sure it's the actual reason? -- WBR, wRAR signature.asc Description: PGP signature
