/srv/irclogs.ubuntu.com/2015/10/10/#ubuntu-installer.txt

cyphermoxslangasek: so I think it should still work even if it allocates the free space00:03
cyphermoxit certainly does run grub-installer properly on x86 :)00:04
slangasekwho here is familiar with how partman-md is /supposed/ to work?00:04
slangasekcyphermox: the fundamental issue here is that we are *not* wanting to create a RAID over two partitions, we're wanting to create a RAID over two disks and then create partitions on the RAID00:05
cyphermoxsure00:05
cyphermoxbut why is that an issue?00:05
cyphermoxif it seems to work properly...00:05
slangasekcyphermox: how are you figuring that it works properly?  we're having this conversation precisely because a bug has been reported00:06
cyphermoxslangasek: a bug specifically on ppc64el, which uses very different code paths than x8600:06
cyphermoxand I seem to be able to reduce the difference there too (pending testing)00:07
cyphermoxpart of the problem on ppc64el is that we also depend on PReP partitions which are not the typical grub /boot partitions and get detected via prep-bootdev, an "external" program grub-installer also ships00:07
slangasekcyphermox: the firmware needs to see a PReP partition.  You can accomplish that one of two ways.  1) Combine two disks into a RAID1 array, partition it, make one of those partitions a PReP partition; or 2) create partitions on the underlying disks, combine those individual pairs of partitions into RAID1 arrays, and mark one of those as a PReP partition00:08
cyphermoxso my analysis as to "it seems to work" is based on the fact that on amd64, it seems to work, I don't deny we still need to figure out how well that scales to ppc64el/opal booting00:08
slangasekcyphermox: from what you've told me, you've tried to do exactly what I also expected to work - that you could select two disks for RAIDing and create partitions on top of them and have it work00:08
slangasekbut that is *wrong* and doesn't work *anywhere* except MBR00:09
cyphermoxslangasek: it's also why it's explicitly marked as not supported in anything but x86 :/00:09
slangasekthat is wrong> the implementation is wrong and does not match the intent00:09
slangasekcyphermox: it's not even supported on x86 with UEFI, it only works for MBR00:09
slangasekdo we automatically add the PReP partition to fstab?00:10
cyphermoxno00:10
slangasekok, so it's a magical thing that's handled via grub directly00:10
slangasekdo you have the bug number for this handy? :)00:10
cyphermoxyeah00:10
cyphermoxso I'm already working on a fix, but there's multiple moving parts and I'm new at the whole thing, so trying to understand things as I change them :)00:12
cyphermoxhttps://bugs.launchpad.net/ubuntu/+source/debian-installer/+bug/148736500:12
cyphermoxit clearly currently doesn't work on ppc64el; we've established that00:13
slangasekcyphermox: how are you intending to solve this?  The install method presented in the bug, that of /dev/md0p1 being used as the PReP partition, will never work with the current state of partman-md00:13
cyphermoxthere's two pieces; teach grub-installer to correctly pick the members of the array when installing grub; and fix partman-md to use the underlying disk when we can somehow detect that it's otherwise unpartitioned00:15
slangasekok, so you are planning to change partman-md to use whole disk instead of creating a partition00:15
cyphermoxso presumably in the case where either it's just free space, or there is no partition table00:15
slangaseksorry, I didn't understand that was your aim00:15
cyphermoxit wasn't until we discussed this00:15
slangasekif you do that, then yes, this all works00:15
slangasekah ;)00:15
cyphermoxI don't pretend to know all the intricacies of RAID or booting, so we refine as we go ;)00:16
cyphermoxI'm starting to think I'll have to discuss this with KiBi too before I get too deep in the code00:16
cyphermoxmy question for you though would be, should this really be a priority for 15.10, or should we focus on more broken things since there is a workaround and people need to write their own partitioning recipe anyway?00:19
slangasekcyphermox: basically, it's like this.  software RAID is invisible to the firmware (whichever firmware it is).  The only RAID that's compatible with being booted from is RAID1, because it's a straight mirror, so the firmware just sees it as two disks with matching content.  Any other RAID scheme requires that the boot logic sit *outside* the software RAID array00:19
slangasekon MBR you dodge this because MBR doesn't boot from a partition00:20
cyphermoxyeah00:20
slangasekon UEFI or opal, you need a partition to boot from; which means either full-disk software RAID + partitions inside it, or selective RAID of individual partitions00:20
cyphermoxyup, got that00:22
slangasekcyphermox: so instead of committing to redesign partman-md, I think we should instead look at better enforcement of partitioning policies00:25
slangasekbecause you shouldn't have to wait for grub to try to install before you find out there's no prep partition00:25
slangasekhistorically there have been architecture-specific partman modules to enforce this00:25
cyphermoxthere alreday is a partman-prep IIRC, I could enforce things there00:26
slangasekIOW I think option 1) is what people generally *want* to have working, but until that works we can get by with directing people to option 2)00:26
cyphermoxenforcing PReP not to be on RAID for now should be simple00:29
slangasekit doesn't require that it not be on RAID, it just requires that it not be a /dev/mdNpM partition00:30
slangasekwell00:30
slangasekthe use case here is that you want both disks to be equally bootable00:31
slangasekif grub can handle that without having to be pointed at /dev/md0 == (/dev/sda1+/dev/sdb1), that's fine00:31
slangasekbut having a RAID1 over the individual PReP partitions may be preferable00:32
cyphermoxwell, that would get complicated to do without an early_script, but I'll give it a test too00:33
slangasekcyphermox: so I was mistaken, the partitioning enforcement is apparently usually done by the bootloader installer packages rather than partman modules per se.  e.g. silo-installer (available in Debian only)00:40
slangasekaboot-installer had one, but good luck finding that in any Debian release...00:41
cyphermoxthere is still some level of enforcement in partman- packages, for instance in the size/location for the ESP00:41
slangasekok00:41
cyphermoxand partman-prep already checks for location00:41
cyphermoxso it should be fine to check there that PReP isn't over /dev/md0p[0-9]+00:42
cyphermoxthat's the easy part though, I still need to get grub-installer right for when it's partitioned right :)00:42
* slangasek nods00:43
* cyphermox uploads a new grub-installer revision to his installer-dev PPA00:43
cyphermoxthis is going to be pretty much all for today I think, I'll start an install with it Monday so I get the results on Tuesday morning.00:45
slangasekalright00:45
slangasekI'm going to braindump to the bug00:46
cyphermoxcool.00:46
cyphermoxslangasek: you mentioned dvorak too?01:23
slangasekI did01:24
slangaseklet me double-check it01:24
slangasekcyphermox: possibly specific to UEFI boot01:26
cyphermoxwat01:31
cyphermoxI don't see how it would change anything01:31
slangasekcyphermox: well when you're using syslinux you get the option to select the keyboard layout from the bootloader01:31
slangasekso that's what most people use?01:31
cyphermoxsure01:31
slangasekif you're using booting under UEFI you don't have that01:31
cyphermoxwell, perhaps, and that will change bits in the command-line01:32
cyphermoxbut when you detect the keyboard and select dvorak it should still get you the right thing01:32
slangasekso it's possible the keyboard selector within d-i is broken for everything, but not usually noticed on !UEFI01:32
slangasekwell yes, it should ;)01:32
cyphermoxit's possible, but I thought I had fixed it when you last reported this months ago01:32
cyphermoxit was missing kbd.01:32
cyphermoxI'll try the server image with and without EFI in a bit, when I'm done with laundry01:33
slangasekwas my last report before or after 15.04 release?01:33
cyphermoxafter01:34
slangasekah01:34
cyphermoxwasn't it this summer?01:34
slangasekquite possibly01:34
slangasekbut as I said, I was booting a 15.04 image01:34
cyphermoxoh01:34
cyphermoxwell, then maybe I should SRU this fix ;)01:34
slangasekSRUing it won't get it into the already-produced image01:34
cyphermoxno01:34
cyphermoxugh, and console-setup and kbd need to be on the d-i image itself01:35
cyphermoxtry a 15.10 image, so at least I'm reassured it works?01:35
cyphermoxI could sprint with a dvorak keyboard, too :)01:36
slangasekyeah just finished downloading that01:37
slangasekcyphermox: bad news; same problem on 15.10 beta01:39
slangasekOct 10 01:38:13 main-menu[235]: (process:1141): /etc/console-setup is not writable. No files will be saved there.01:43
slangasekOct 10 01:38:13 main-menu[235]: (process:1141): gzip is not accessible. Will not save cached keyboard map.01:43
cyphermoxlooks like the same kbd issue :(02:02
cyphermoxso, first, bios boot and I pick dvorak at the keymap panel in gfxboot.02:04
cyphermoxI do get dvorak02:05
cyphermoxnow, uefi02:05
cyphermoxso, English (US), then English (US) - Dvorak02:06
cyphermoxstill got dvorak?02:06
cyphermoxthat was with ubuntu-server from 2015100902:08
cyphermoxwhat image did you use?02:08
cyphermoxslangasek: ^02:08
cyphermoxor are you using the keyboard-detection?02:09
cyphermoxhum, if you go back and re-do the keyboard selection it looks like maybe it won't update the keymap properly02:11
cyphermoxneh, the keyboard-detection is broken altogether02:12
cyphermoxyup, the detection algo is broken, but the selection from lists appears to work, at least as far as picking between us and us:dvorak and having the right setting applied following that is concerned02:14
cyphermoxI bet it's broken in Debian too02:16
slangasekcyphermox: ah; I didn't yet test a manual keyboard selection with 15.10, only the keyboard detection method, and that failed02:19
cyphermoxyeah I confirm it's really broken02:20
slangasekok, but manual selection works02:20
slangasekon 15.1002:20
cyphermoxah, no detection in debian?02:20
slangasekit didn't work for me in 15.0402:20
slangasekcorrect, the detection is an Ubuntuism02:20
cyphermoxslangasek: yeah, the manual selection works02:20
cyphermoxhow would you go about picking dvorak on Debian?02:22
cyphermoxdoesn't look like I have the choice at all after American English02:22
cyphermoxoh, so that's the 'plugin-detect-keyboard'02:24
cyphermoxshiny.02:24
cyphermoxslangasek: so where does this plugin come from?02:27
cyphermoxso after detection it looks like you got setupcon to run, as it should, but still not setup the keymap properly02:34
cyphermoxah, oops. looks like we mangle XKBVARIANT along the way02:38
* cyphermox adds that to the list for Tuesday02:41
slangasekcyphermox: ok; so did you file a bug to track this?03:45
infinityslangasek: FWIW, OPAL doesn't require a PReP partition, since it doesn't actually use grub.  OpenFirmware does, however, and we install the same way under OF/SLOF/OPAL to preserve sanity (and let disk images migrate).08:48
infinityslangasek: And, not, it's not added to fstab, because it's a raw partition with bootloader of choice (yaboot or grub, generally) blatted directly to it, is contains no filesystem.08:49
infinitys/is contains/it contains/08:50
=== psivaa_ is now known as psivaa

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!