On 12/15/2016 09:48 AM, Reg Tiangha wrote: > On 12/15/2016 09:43 AM, Foppe de Haan wrote: >> On Thursday, December 15, 2016 at 2:15:20 PM UTC+1, Reg Tiangha wrote: >>> On 12/15/2016 06:01 AM, Reg Tiangha wrote: >>>> On 12/15/2016 05:00 AM, Foppe de Haan wrote: >>>>> Okay, stuck again. Downloaded a fresh d8 template, updated it and built >>>>> kernel again to have a clean slate to work from. Everything builds, no >>>>> error beside the expected ones afterwards. But when i switch to pvgrub2 >>>>> and boot, it shuts down seconds after boot attempt. Full log attached. >>>>> >>>>> ..[30m.[47mWelcome to GRUB! >>>>> >>>>> >>>>> ..[37m.[40m.[37m.[40m.[37m.[40merror: no such device: >>>>> /boot/xen/pvboot-x86_64.elf. >>>>> >>>>> Reading (xen/xvda/boot/grub/grub.cfg >>>>> >>>>> ..[H.[J.[1;1Herror: file `/boot/grub/fonts/unicode.pf2' not found. >>>>> >>>>> error: no suitable video mode found. >>>>> >>>> I can't replicate, but none of my systems have UEFI or if they do, I'm >>>> using legacy boot because Qubes wouldn't install if UEFI mode was enabled. >>>> >>>> That said, if it worked for you before, it should still work now. >>>> >>>> Are you sure you set up grub correctly after the compile? >>>> >>>> sudo mkdir /boot/grub >>>> sudo update-grub2 >>>> >>>> The only other thing I can think of is to switch to a regular kernel, >>>> try the two grub commands again and see if any errors pop up, and if >>>> not, switch back to pvgrub2 and see if it boots again. >>>> >>>> >>> Oh, wait. When you said you downloaded a fresh Debian 8 template and >>> updated, do you mean that you updated it to Debian 9 rather than just >>> install the regular Debian 8 updates? I'm running another test system >>> based on Stretch on bare metal, and I think there's something buggy in >>> one of the packages that was released recently (maybe X?) as it's been >>> locking up a lot for reason I can't figure out. I'm hoping I can just >>> wait it out and it'll fix itself later in another update. It could be >>> related to your problem if that's the case? >>> >>> And I just realized there was more to the log that you attached. Still >>> can't figure out what caused the kernel panic from looking at it, sorry. >> No, I mean running 8/stable, no other repos active. Would dom0 running uefi >> matter to the VM >> >> As for the rest, I've rerun every step a few times, so I can't imagine I >> missed anything. update-grub2 gave me the message below, but according to >> the qubes howto page, those errors don't matter. >> >> Anyway, thanks for your replies / time, I guess I'll poke around the log etc >> a bit more. >> >> >> ----- >> user@debian-8:/boot$ sudo update-grub2 >> Generating grub configuration file ... >> Found linux image: /boot/vmlinuz-4.8.13-coldkernel-grsec-1 >> Found initrd image: /boot/initrd.img-4.8.13-coldkernel-grsec-1 >> /usr/sbin/grub-probe: error: cannot find a GRUB drive for >> /dev/mapper/dmroot. Check your device.map. >> /usr/sbin/grub-probe: error: cannot find a GRUB drive for >> /dev/mapper/dmroot. Check your device.map. >> /usr/sbin/grub-probe: error: cannot find a GRUB drive for >> /dev/mapper/dmroot. Check your device.map. >> No volume groups found >> done >> > Yeah, those particular errors don't matter. > > I guess later today, I'll try reinstalling the debian-8-template from > the yum repository and try to build the kernel again from scratch, just > to make sure. What I've been doing with every Qubes machine of mine is > installing the templates from the DVD or the repository, updating it, > and then immediately cloning it and working off of the clone. That way, > all of my original template rpms are still pristine. > > However, I don't think there's any difference between what's on the 3.2 > iso and what's the latest debian 8 template from the repository so I > don't know if that'll do much in my testing; I think they're the same > version? I'm sure a dev will chime in if that's not the case. > > Your problem is very strange, especially if coldkernel had already > worked before. > >
The last thing I can think of off the top of my head at the moment is that every Qubes VM by default outside of sys-net have "nopat" set as a kernel option. If you use pvgrub2 though, it ignores any of the kernelopts options that are set by qvm-prefs. To make your coldkernel template vm boot with the nopat option, you have to make an /etc/default/grub file that contains "GRUB_CMDLINE_LINUX="nopat" (along with any other kernel options you would normally pass through) before running update-grub2. So if you haven't done that yet, maybe try making that file and updating grub and try again? That's the only thing I can think of at this point. -- You received this message because you are subscribed to the Google Groups "qubes-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to qubes-users+unsubscr...@googlegroups.com. To post to this group, send email to qubes-users@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/qubes-users/o2uhv9%24r2r%241%40blaine.gmane.org. For more options, visit https://groups.google.com/d/optout.
