| blut | How can I let a preseed/early_command run in the background? My command is currently http://paste.ubuntu.com/18779153/ and the 'sh /kill-partman.sh &' doesn't go to the background, eventhough I can run the exact command in the console during the installation and it works. Any ideas? | 11:33 |
|---|---|---|
| slashd | hi, I have someone that can't sucessfully install Xenial due to a scsi transfer length issue. In order to troubleshoot I want him to test the same Xenial install using another kernel version to see if this will succeed. How can I skip the ISO default kernel and force debian installer to pick the desired kernel ? I've been told that "base-installer/kernel/override-image string " might do the job, but no quite sure where t | 15:14 |
| slashd | he kernel needs to be store for d-i to grab it and install it.... | 15:14 |
| slashd | I sent an email to the mailing list ^^ | 15:49 |
| dmj_s76 | cyphermox: I've been trying to write a fix for the ubuntu installer for systems with nvme drives. Currently (in 16.04) ubiquity fails to install grub. | 16:29 |
| dmj_s76 | Actually found a small 1 line change that correctly installs grub where the nvme drive is the only disk in the system... | 16:31 |
| dmj_s76 | but for some reason in bios mode when there is a non-nvme disk too, ubiquity/misc.py grub_default() targets /dev/sda1 instead of /dev/nvme0n1 even though we're installing to the nvme drive (default). The installer then fails to install grub to the sda drive. | 16:35 |
| dmj_s76 | Is there a reason for why ubiquity naively picks the sda drive? | 16:37 |
| cyphermox | must be something broken in ubiquity itself; I expect grub is otherwise doing the right thing (we've done a few fixes for NVMe before) | 16:56 |
| cyphermox | please file a bug against ubiquity | 16:56 |
| dmj_s76 | cyphermox: https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/1600308 | 17:19 |
| cyphermox | dmj_s76: thanks. I'll get back to it soon; but for now I need to focus on other pressing stuff for today; and then I'm on vacation for two weeks | 17:19 |
| cyphermox | so I'd get back to it after vacation, I think | 17:20 |
| dmj_s76 | who else maintains ubiquity in case I have questions? | 17:20 |
| dmj_s76 | hoping to have a patch ready early next week | 17:21 |
| cyphermox | dmj_s76: pretty much only me, more or less | 18:20 |
| cyphermox | dmj_s76: send me an email and I'll see what I can do | 18:20 |
| cyphermox | or file a bzr merge proposal, and if someone feels like reviewing ubiquity they can merge it while I'm away :) | 18:20 |
| dmj_s76 | cyphermox: I'll have a merge proposal for the single drive case...I think it's really two different issues in the same area. | 18:22 |
| dmj_s76 | will have that today | 18:23 |
| cyphermox | ok | 18:23 |
| cyphermox | I won't have the time to look at it today I'm afraid | 18:23 |
| dmj_s76 | that's fine, I'm just hoping to have it in time for 16.04.1, to make the next several months easier for our customers. | 18:24 |
| cyphermox | ok | 18:25 |
| cyphermox | it's getting pretty short for 16.04.1; but please propose your patches and we'll do what's needed to land it if possible | 18:25 |
| dmj_s76 | it's 1 line that only applies to nvme disks | 18:26 |
| dmj_s76 | but definitely getting short | 18:26 |
| dmj_s76 | cyphermox: I have a merge proposal ready. | 19:15 |
| dmj_s76 | It should make installs painless for at least half of our users with nvme drives. | 19:16 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!