IRC channel logs


back to list of logs

<catonano>ah wait
<catonano>I have this line in
<catonano>packages=['distutils', 'distutils.command'],
<catonano>so yes of course I need distutils in the native-inputs
<ng0___>afaik this changed at some time
<catonano>yes, they moved from setuptools to distutils or the other way around. I have distutils in the dependencies becauuse I have copied stuff around
<ng0___>I mean, this should be part of the buildsystem
<ng0___>the guix one
<ng0___>ah, ok
<catonano>build succedeeed !
<ng0___>oh, your own software?
<catonano>a small project. i was asked to help, it's an emergency ;-)
<catonano>for now it's empty, I have thhe file only ;-)
<catonano>I ust wanted to see i I could operate thhe python tooling through Guix
<catonano>it's about scraping a web page
<catonano>I remember i saw a python framework for scraping
<catonano>it's important that it's python so other people could contribute some code too
<catonano>also, Cloure on Guix is still a bit rusty. I saw someone packaged lein, togh !
<catonano>ok, I think I'll call it a night, now
<ng0___>alright, good night
<catonano>ouch ! Scrapy is not packaged in Guix :-/
<catonano>paroneayea: could you manage to get a cloure repl, finally ?
<quiliro>guix package -i hello
<quiliro>pun intended!
<quiliro>i want toconnect remotely via ssh and start guix system ini...while it is installing i want to disconnect without the command suspended
<paroneayea>catonano: ha, yes-ish, i took a shortcut
<quiliro>then i would like to reconnect and check if the command was successful
<paroneayea>I used C-u M-x inferior-lisp
<paroneayea>C-u M-x inferior-lisp <RET> java -cp /home/cwebber/.guix-profile/share/java/clojure-1.8.0.jar clojure.main <RET>
<quiliro>what is the root password of the guixsd install usb?
<catonano>paroneayea: I see. Thanks
<quiliro>i did guix system init
<quiliro>rebooted and grub worked but then it coud not find my-root
<quiliro>and gives a guile prompt
<quiliro>Error: in proceedure scm-error
<quiliro>failed to resolve partition "my-root"
<buenouanq>quiliro: and it was labeled properly when you made the partition?
<quiliro>let me check
<quiliro>buenouanq: i rebooted with usb and executed the following command
<quiliro>mount LABEL=my-root /mnt
<quiliro>it was successfully mounted
<quiliro>perhaps because (bootloader (grub-configuration (device "/dev/sdb")))
<quiliro>and and now that i removed the usb it is sda?
<quiliro>perhaps i should use 0.12 which recognized usb as the last device
<quiliro>but this has no sense
<quiliro>would someone please give me a tip?..... or at least moral support
***Acou_Bass is now known as eddie
***eddie is now known as Acou_Bass
<quiliro>bms_: hello
<bms_>How are you?
<quiliro>bms_: with problems installing guixsd
<quiliro>bms_: and you?
<bms_>I'm well. What are your problems. I've installed it (with varying degrees of success) multiple times.
<gazon>quiliro: don't pin your hopes on my help (n00b here!) but can we see the system config.scm?
<quiliro>Error: in proceedure scm-error
<quiliro>failed to resolve partition "my-root"
<quiliro>and gives a guile prompt
<quiliro>it is at boot
<quiliro>grub is ok
<quiliro>and it gave no error on installation
<quiliro>it is my forst boot with that installation
<quiliro>first boot
<gazon>it seems quite possible that you are correct that the device name has changed since you removed the usb
<quiliro>gazon: but it has trouble finding my-root not sdb
<quiliro>gazon: so, what do you suggest
<quiliro>bms_: what is your take?
<quiliro>reboot with 0.12 and reinstall?
<quiliro>with definition as sda
<bms_>Sounds like you labeled your root filesystem something other than my-root, which is the default in config.scm
<quiliro>in bare-bones.scm
<bms_>Oh, yes, I just always name my own config. I'd reinstall with 0.13, but if you want, 0.12 works too. Just make sure to take notice of the filesystem labels.
<gazon>i don't think (bootloader (grub-configuration (device "/dev/sdb"))) is going to work because the device needs to be correct at install time for grub to be written.
<quiliro>bms_: 0.12 used to recognize hard disks first then usb
<gazon>s/sdb/sda/ i mean
<quiliro>sdb is the hard disk when booting with the usb
<quiliro>so... i don't have any ideas
<bms_>Yeah, "/dev/sda" is what you want. You should set it to the target disk, not the USB used for installation.
<quiliro>bms_: are you sure?
<gazon>but how about initially setting to sdb and then *before* you reboot, change it to sda
<bms_>Give me a second. I'll check mine.
<bms_>Initially setting? In bare-bones.scm, you should set to the target disk. I'm assuming that there is no trouble loading the USB for installation.
<bms_>Mine looks like this: (bootloader (grub-configuration (grub grub-efi) (device "/dev/sda"))).
<bms_>Don't worry about the (grub grub-efi) unless you have a UEFI/EFI system.
<gazon>the problem seems to be that, after booting the usb, the target is sdb, but after install and reboot, the sdb device becomes sda
<quiliro>gazon: i don't think the config file is used after installation
<quiliro>unless you reconfigure
<quiliro>gazon: exactly...your last coment
<bms_>Oh... yeah, I'm pretty sure those names aren't constant. You should set a label when you use mkfs, like it does in the manual. Set the config file pointing to the hard drive's name as it is when you install.
<gazon>quiliro: no you're right, but it might still work if you change the config and system reconfigure --no-bootloader (or whatever)
<bms_>--no-bootloader would work, if you already have Grub installed and would like to write a manual menu entry.
<bms_>At least, that's how I've used it in the past.
<quiliro>so you suggest i boot with the usb and define (bootloader (grub-configuration (device "/dev/sda"))) and run 'guix system reconfigure --no-bootloader /mnt/etc/bare-bones.scm /mnt'?
<quiliro>(bootloader (grub-configuration (device "/dev/sda"))) in /mnt/etc/bare-bones.scm
<bms_>Yes, but with reconfigure, you don't need to specify /mnt, just /mnt/etc/bare-bones.scm
<gazon>no no. do the install as before, setting bootloader to sdb...
<quiliro>i used sdb before
<quiliro>and it does not work
<gazon>then when install has finished, change the config to sda and reconfigure
<quiliro>oh...before rebooting!
<gazon>the bootloader is already written to the device so --no-bootloader
<quiliro>bms_: do you agree?
<gazon>yes, that's the suggestion, but who knows if it will work...
<gazon>it's worth a try no?
<bms_>I agree.
<quiliro>it is worth a try
<methalo>quiliro: i use (bootloader (grub-configuration (device "/dev/sda")))
<methalo>and works!
<gazon>what would be better is if there was a way to refer to devices by some other (permanent name), but i don't think the hurd has a devfs like linux
<quiliro>methalo: do you have the usb recognized as sda and the hard disk recognized as sdb when you boot with usb?
<methalo>quiliro: no, usb is recognized as /dev/sdb
<quiliro>something weird is happenning...i think my usb ¡is getting reconfigured
<quiliro>methalo: so you don't have the same problem as i do
<gazon>oh no
<quiliro>i will leave it
<quiliro>i think that it is good because i will have a personalized usb...and if it is non modifiable, then nothing will happen
<quiliro>the problem will be if it does not boot
<quiliro>but if it does not boot, i can always dd it again
<quiliro>thank you for making me make that test
<quiliro>i am learning!
<gazon>me too.
<gazon>something bms_ said though
<gazon>no, ignore that. what you need is a small team of experts.
<quiliro>experts are people that have a lot of experience
<quiliro>so we are experts!
<gazon>heh heh, just not with guix.
<gazon>but going back to that thing that bms_ said: how about, instead of (file-system
<gazon> (device "my-root")
<gazon>no no no it won't work.
<gazon>ignore that.
<gazon>i was thinking you could try "/dev/disk/by-uuid/<some uuid>", but I don't think it will work.
<quiliro>but "/dev/disk/by-uuid/<some uuid>" will work!
<gazon>will it? why?
<quiliro>it broke my usb....i will reboot that machine
<gazon>ACTION goes afk for a few minutes
<quiliro>let me check...the problem is either the label or the device
<quiliro>so if i change to the uuid, it might work
<quiliro>i will read the manual to make sure
<quiliro>ok...rebooted now...using sda reconfigured the usb...but since it is not returned to normal
<gazon>yes, uuid should work: (file-systems (cons (file-system (device "<some uuid>") (title 'uuid) (mount-point "/") etc.
<gazon>the title should be 'uuid instead of 'label
<quiliro>gazon: thanks
<DoublePlusGood23>Is LVM still not working on guixsd?
<gazon>DoublePlusGood23: i don't know the answer to your question, but there is a package named lvm2:
<quiliro>DoublePlusGood23: is there a bug report?
<DoublePlusGood23>bms_: heyo
<DoublePlusGood23>gazon: Yeah it seems like it still lacks LVM not sure how I'll do my cryptswap now
<wsddn>Quick question: using guix packages with the -m for manifest option: why would in one generation guix uninstall and install a package whose version has not changed (but the hash has changed) with no substitutes?
<quiliro>if the software has not changed and the hash changed, that means the way it is compiled has changed....does that clarify a little, wsddn
<quiliro>so it is a different version by way of that detail
<wsddn>quiliro: so the package definiton changed?
<quiliro>not necesarily
<quiliro>the definition says which packages it uses
<quiliro> are right
<quiliro>wait...let methink well
<quiliro>the definition changes when the hash changes
<quiliro>the hash changes when the dependencies change
<wsddn>I ask because in the space of 40 minutes weechat and icecat reinstalled twice
<quiliro>so yes....the definition changes when i change the configuration of the package or its dependencies even if the package or its dependencies use the same versions as before...even more if those packages change versions
<quiliro>well i have never changed, it would be the developers
<quiliro>how did you notice?
<quiliro>can you list-generations?
<wsddn>Noticing that weechat was compiling twice
<quiliro>or guix package -I package_name
<quiliro>then you will see if there are changes
<quiliro>or both commands
<quiliro>do you know how to use them?
<quiliro>guix package -I weechat
<quiliro>guix system --list-generations
<wsddn>Generation 5 May 25 2017 01:41:13
<wsddn> + xorg-server 1.19.3 out /gnu/store/wvgkdp3k2dh76vcrrkapg66b9ifq8f7j-xorg-server-1.19.3
<wsddn> + xinit 1.3.4 out /gnu/store/bnx71vgr4blvdbc3yjx6p1ryyb6yq8a3-xinit-1.3.4
<quiliro>rather the -I will work if you have not yet installed
<wsddn> + mpd 0.20.6 out /gnu/store/9k8wzbrwq7z0f7qsi4xc4x9fdyp5b886-mpd-0.20.6
<wsddn> + icecat 52.1.0-gnu1 out /gnu/store/70djgkdw52xz2ip37c7hjx9vz9c9avih-icecat-52.1.0-gnu1
<wsddn> + weechat 1.8 out /gnu/store/4hxpy3fipc7sja2vm0vbzn5x26cc2smq-weechat-1.8
<wsddn> - htop 2.0.2 out /gnu/store/awycv8g2jcgnlpnaxlryk46zn6n8znir-htop-2.0.2
<wsddn> - mpd 0.20.6 out /gnu/store/c4jg4ch4jd3hw7wbbb7xrlxphmvjjdsb-mpd-0.20.6
<wsddn> - icecat 52.1.0-gnu1 out /gnu/store/rcva5fyv76ip4996v356dvxglrcghl6c-icecat-52.1.0-gnu1
<wsddn> - weechat 1.8 out /gnu/store/0bbna8pd63bry3db971znvcbyjxw7dg9-weechat-1.8
<wsddn>Generation 6 May 25 2017 02:20:03 (current)
<wsddn> + i3status 2.11 out /gnu/store/36khdy1m6w1m2hw2dl13fpld1269dw3l-i3status-2.11
<wsddn> + thunar 1.6.6 out /gnu/store/3zl4x6i5n7l4z7yq75cq85fs5rkvaz3v-thunar-1.6.6
<wsddn> + feh 2.18.3 out /gnu/store/0m8l6y84lglm3m9vd59nzg7swmhc2wx7-feh-2.18.3
<quiliro>and compare with guix package -s weechat
<wsddn> + mpd 0.20.6 out /gnu/store/vpj9vgxjr1fvjm3fm4m33vnx6ls73z6q-mpd-0.20.6
<wsddn> + htop 2.0.2 out /gnu/store/awycv8g2jcgnlpnaxlryk46zn6n8znir-htop-2.0.2
<wsddn> + emacs 25.2 out /gnu/store/qhrgmhsl6by2awaklqpwhlilc7p1kc8s-emacs-25.2
<wsddn> + vim 8.0.0600 out /gnu/store/9066jaf2w2vxfd31wzg4ciq6bwbc7mmn-vim-8.0.0600
<wsddn> + git 2.13.0 out /gnu/store/a09p3svmf6lx7nsinzl1sa52q7c64jnk-git-2.13.0
<quiliro>wsddn: please do not flood
<wsddn> + breeze-icons 5.34.0 out /gnu/store/4spl14nzdf1dnb2fx7m9ikw3znabzj0p-breeze-icons-5.34.0
<wsddn> + gnupg 2.1.21 out /gnu/store/hnv8z8cl5c6q5ldxn06wx98zr78iwc73-gnupg-2.1.21
<wsddn> + password-store 1.7.1 out /gnu/store/ma5ji2sb5wkdc02dlqd50l3j42y4n145-password-store-1.7.1
<wsddn> + icecat 52.1.0-gnu1 out /gnu/store/nb64j11jys7ijzlb4a5swgqmw228gjs9-icecat-52.1.0-gnu1
<wsddn> + weechat 1.8 out /gnu/store/8p0dkk2l4zh3gq185hjpbn1sf4z1nrbr-weechat-1.8
<wsddn> - i3status 2.11 out /gnu/store/lxvlw79nv5ldk524jyi0l72hynlv8sdj-i3status-2.11
<wsddn> - thunar 1.6.6 out /gnu/store/rlixlzywd51c3fci1xb8w2p8rh95z164-thunar-1.6.6
<wsddn> - feh 2.18.3 out /gnu/store/f4436w7fjmmc536g48rzymcpha577vh1-feh-2.18.3
<quiliro>oh my !
<wsddn> - mpd 0.20.6 out /gnu/store/9k8wzbrwq7z0f7qsi4xc4x9fdyp5b886-mpd-0.20.6
<wsddn> - emacs 25.2 out /gnu/store/74q3x9brl22x58qdsgqvl8pc1bwg3gmg-emacs-25.2
<wsddn> - vim 8.0.0566 out /gnu/store/ax5cm9gr1741pcq17w7bhgss5nvq5470-vim-8.0.0566
<wsddn> - git 2.13.0 out /gnu/store/pp4gs05h5nz5iy00xphd1y8n8sa3qsy6-git-2.13.0
<wsddn> - breeze-icons 5.34.0 out /gnu/store/y6ddamf2rvb88rm3m61bxgqx9hfvqx9a-breeze-icons-5.34.0
<wsddn> - gnupg 2.1.20 out /gnu/store/pp6xwp2rv41m6wvk2v61vmnnq6vbdn1m-gnupg-2.1.20
<wsddn> - password-store 1.7.1 out /gnu/store/h7g56ph7195n427k72a9w1f0wnnl34zi-password-store-1.7.1
<wsddn> - icecat 52.1.0-gnu1 out /gnu/store/70djgkdw52xz2ip37c7hjx9vz9c9avih-icecat-52.1.0-gnu1
<wsddn> - weechat 1.8 out /gnu/store/4hxpy3fipc7sja2vm0vbzn5x26cc2smq-weechat-1.8
<wsddn>Sorry about the spam
<wsddn>Sorry was fighting with tmux
<rain1>its ok
<quiliro>i can see you have installed and removed weechat several times
<quiliro>wsddn: do you know how to paste?
<wsddn>Normally I would but xorg doesn't work in the right way so I'm stuck in the terminal atm
<quiliro>wsddn: easy:
<rain1>there is a site that lets you paste a text file from terminal
<rain1> and and some other
<rain1>using 'curl'
<quiliro><command> | curl -F 'sprunge=<-'
<quiliro>if you do not have curl installed use
<quiliro>guix package -i curl
<wsddn>I didn't add or remove "weechat", "emacs", or "icecat" from my user-packages.scm file, and I used "guix package -m user-packages.scm"
<quiliro>ws i have never used -m
<wsddn>Thanks both quiliro rain1 for the paste adivce :)
<wsddn>quiliro: My user-packages.scm file derives heavily from the second example given in the manual
<quiliro>(packages->manifest (map (compose list specification->package+output) '("emacs" "guile@2.0" "guile@2.0:debug")))
<quiliro>that one?
<quiliro>cool...want to use emacs
<quiliro>i suggest you run the command and check the list-genrations
<quiliro>so you can see what is done based on your changes
<quiliro>to the manifest files
<wsddn>The Generation 5 and Generation 6 (from guix packages --list-generations) I posted above are the result of running "guix package -m".
<wsddn>Could this all just be due to one of the dependencies being updated?
<quiliro>i would agree
<quiliro>test it without making changes
<quiliro>will it remove or add packages after the second guix package command without changing the manifest file?
<quiliro>that is the question my dear shakespear
<quiliro>i bet not
<quiliro>i cannot boot to my newly installed guixsd
<quiliro>it will complain about not finding my-root partition
<quiliro>even though grub works
<wsddn>quiliro: I think you were right about it being a dependency being updated. Thanks for your help, I am slowly but steadily learning more about guix (maybe someone should make a wiki)
<wsddn>Good luck trying to fix grub!
<DoublePlusGood23>quiliro: what's your config.scm look like?
<quiliro>This is my bare-bones.scm
<quiliro>with uuid
<quiliro>gave me the same error at boot with label
<mbakke>quiliro: try commenting out the swap partition
<mbakke>I mean the "swap-devices" line
<quiliro>mbakke: is that why it does not find "my-root" partition?
<quiliro>commented swap line with ;;
<quiliro>and running
<quiliro>guix system init /mnt/etc/bare-bones.scm /mnt --fallback
<mbakke>quiliro: just want to rule out that the problem is due to the swap partition (you mentioned device paths are not stable)
<mbakke>ACTION crosses fingers
<quiliro>you like to experiment a lot
<quiliro>it causes a lot of problems.....but a lot of learning too
<Apteryx>Anyone else having refresh problems with Geiser + Guile ?
<Apteryx>sometimes the errors don't get print
<quiliro>this is my new bare-bones.scm
<mbakke>ooh paroneayea on top of hacker news
<quiliro>oh! that is why download of guix package is 60 KB/s
<quiliro>mbakke: link please
<quiliro>by the way, what is the difference between guix pack and guix package -i ?
<paroneayea>mbakke: I am??
<paroneayea>oh mediagoblin
<quiliro>where is media goblin
<quiliro>this is the result of guix_system_init
<fusion809>Hey does guix write its output to a log file? I ask because as I'm running it from the command-line without a GUI I can't scroll back on the output all that far and it would be helpful if I could view the full output. I know redirecting the output to a file (with COMMAND > output.log 2>&1) would manage this but I'd like to see the output in real-time so I can see in real-time what guix is doing and this method wouldn't allow for
<janneke>fusion809: there is: guix build --log-file
<fusion809>OK, will that --log-file option be available for guix system init?
<janneke>fusion809: no, only for individual packages
<janneke>you'll have to run in an emacs shell or use tee or something
<fusion809>Ah. Thanks.
<janneke>i don't think we have log files for `guix system'
<fusion809>Could I suggest that feature be added like at a bug tracker? I don't think it'd take a lot of work to do and would be of benefit to both users and developers as it means that if an error occurs it should be easier to give developers the full output log.
<janneke>fusion809: sure
<janneke>always good to communicate, esp. if you hit some problem. i'm not sure what the priority will be, as you have several options to script it outside of guix already
<fusion809>I'm not used to the bugtracker ( and I can't seem to find an option to login (which I assume would be necessary to file a bug) or file a bug without one.
<efraim>I use screen and I set scrollback to 50000 to work around that. I thought you only got log files on successful builds
<quiliro>fusion809: why don't you do a remote install and you will be able to scroll back on the workstation
<fusion809>efraim: is that in a graphical terminal? quiliro: because I haven't the foggiest how to for a VM. I'm installing in a QEMU VM.
<quiliro>there are still problems with guix 0.13
<efraim>on my laptop i run screen inside terminology, on my aarch64 boards i run screen from tty1
<fusion809>Guessing screen has to be installed with guix? Asking as I'd like to give it a try.
<quiliro>you can scrollback as far as you config file determines....i think in bash it is .bash_profile
<fusion809>I'm at tty1. Does that apply to tty1?
<quiliro>scrren is installed with guix?
<fusion809>As the screen command doesn't seem to exist by default.
<efraim>i installed screen from guix, but it should also be available on all foreign distros
<fusion809>How do I set scrollback in bash? Guessing involves setting some environment variable. Googled it but seemed closest to apply to this situation and it doesn't seem to say how to do so with bash.
<fusion809>Oh well I ran screen and hopefully it is sufficient for this situation.
<fusion809>Seems to be as running Shift+PageUp scrolls back well.
<quiliro>fusion809: cool
<quiliro>that is the configuration file for bash
<quiliro>for each user
<quiliro>on their home
<fusion809>Yep I know that, but how to set scrollback using it.
<fusion809>Well ~/.bashrc is also a config file for bash. I usually use bashrc but bash_profile I'm willing to use too.
<fusion809>Confused as ya said, "you can scrollback as far as you config file determines....i think in bash it is .bash_profile" I assume that means you know how to set it up with .bash_profile. Not trying to insult ya just trying to understand what is going on here.
<fusion809>brb gonna switch to i3 on my host OS, KDE is freezing up on me and it's getting frustrating.
<quiliro>fusion809: i do not remember how i did it.....i think it is HISTFILE variable
<fusion809> it's command history not output history.
<fusion809>command history isn't what interest me, it's the output.
<fusion809>But thanks anyway I screen seems to suffice for me so I probably won't need to use bash
<quiliro>fusion809: that's right
<quiliro>i remember now
<quiliro>how can i see a log for boot for a system that connot detect the root partition but grub does?
<rekado>ACTION updates vlc
<quiliro>rekado: i love vlc!
<davidl>quiliro: when you installed with 0.12 image, do you run guix pull first?
<quiliro>davidl: i did not install with 0.12
<quiliro>this time
<davidl>ok. Have you had a working install with 0.13?
<quiliro>davidl: no
<quiliro>with 0.12 yes
<davidl>quiliro: ok. I had a working install with 0.12 too. But very unregularly.
<quiliro>i did guix pull with 0.12 but i guess there were package problems lately
<quiliro>about a month ago i had a working install
<quiliro>with 0.12
<davidl>it's that unregular still lol
<quiliro>but the last 7 days have been hell
<davidl>for me too.
<davidl>Only time it actually "worked" for me was when my installation ended up in /subvol/ and I couldn't boot.
<quiliro>perhaps we should develop an lts version
<davidl>There is
<quiliro>oh1 thats right!
<rekado>fusion809: you don’t need an account to use the bug tracker. Just send email to
<quiliro>davidl: it is not is Rolling release model
<quiliro>fusion809: that is the bug report
<davidl>quiliro: yeah, but maybe the rolling release will become quite stable later, like Arch/Parabola.
<quiliro>arch and parabola are not stable...check
<brendyn>stability is a very difficult thing to strive for
<davidl>quiliro: that's cool. I run Parabola on my work-pc and haven't really noticed any issues except maybe once and for individual packages which are beta.
<quiliro>brendyn: absolutely
<janneke>do i need to add a (c) line for a package version bump?
<quiliro>davidl: i have had a lot of problems over the years with parabola
<janneke>ACTION updates exwm 
<davidl>quiliro: apparently a lot more common than I thought.
<brendyn>I think at the moment we should take advantage of not having any backwards compatibility or LTS version to maintain. That way people like Ludovic can keep improving Guix unhindered until it's ready for prime time
<fusion809>quiliro: can I email without subscribing to any mailing lists?
<quiliro>fusion809: yes
<fusion809>Oh good. I dislike mailing lists as they fill up my inbox with emails about stuff I don't care about, very rarely are there replies to topics I care abut.
<quiliro>brendyn: it is not a problem to have an lts. rather it helps devs from having to support temporary issues
<quiliro>and newbies do not take as much time from the devs
<quiliro>because they are assured a working version
<rekado>quiliro: an lts is not going to happen any time soon
<sneek>Got it.
<rekado>sneek: forget an lts
<quiliro>rekado: i think it is not difficult...and i offer to learn to make it
<rekado>quiliro: it would require backporting countless fixes to old software, and that takes a lot of work.
<quiliro>i am not asking for others to do it
<quiliro>rekado: it is easier in guixsd than in debian
<rekado>quiliro: if you want to learn how to do this I suggest first to learn how to keep the software we offer through Guix up-to-date.
<quiliro>rekado: exactly
<rekado>lfam does a great job of applying security fixes in a timely manner, and it would help to have more people do this kind of work.
<rekado>for a start, you could subscribe to relevant security mailing lists and monitor them for reports of vulnerabilities affecting our software.
<quiliro>rekado: no...for starters, i must get my guixsd installation to boot
<fusion809>Hey guix system init fails due to grub. It gives this output
<fusion809>I created a partition table similar to that mentioned in "Preparing for installation". Except I had /dev/sda2 as a SWAP partition, likewise /dev/sda1 is the root partition. Never mentioned a partition specifically for GRUB.
<fusion809>I didn't have /dev/sda2 as SWAP
<fusion809>I have 8GB available for my GuixSD VM so no need for swap IMO.
<brendyn>fusion809: I think it is because you are using GPT
<fusion809>Ah so MBR is best?
<brendyn>Well GPT is supposed to be "better" but using MBR tends to just work without issues
<fusion809>Thanks I'll give it a go.
<brendyn>I believe you can Convert GPT to MBR easily
<brendyn>Otherwise, it looks like you need a BIOS partition
<fusion809>To late, rofl, just deleted my GPT partitions.
<fusion809>If BIOS partition is needed I'd suggest that the docs get fixed. Do I need to send another bug report to bug-guix
<rekado>GPT is supported, no?
<fusion809>Using GPT didn't work for me so its support isn't as good as it should be. Granted GuixSD is still in early development so bugs can be expected.
<brendyn>It works, you just have to set it up right
<brendyn>I don't think there was any bug
<rekado>brendyn: could be a bug. We just need more information on what went wrong.
<janneke>how do i change the file-name when doing a manual guix download?
<janneke>i.e, mimicking (file-name ...) field in a package
<rekado>janneke: I don’t think you can
<fredmanglis_>Hi. I'm trying to lint a package, but I think the lint command might be broken. How does one handle such situations?
<rekado>I usually just download with wget
<janneke>rekado: hmm, i keep getting build failures like: guix package: error: build failed: directory `/homeless-shelter' exists; please remove it
<rekado>and then maybe guix download file://$PWD/the-renamed-thing.tar.gz
<janneke>rekado: thanks, that works (i only tried download file://./the-thing)
<janneke>the workaround is to guix dowload by hand ... not sure why it sometimes fails
<rekado>that’s a weird error. At what point do you get it?
<rekado>ACTION converts packages to use modular Qt.
<janneke>when building a package, guix build or guix package -i
<quiliro>why is 'my-root' not detected on a new install?
<quiliro>with 0.13
<fusion809>Rofl tried to install with MBR and it gave the same error
<janneke>ACTION is building emacs-exwm-x
<rekado>ACTION goes to a conference, will be afk
<efraim>sneek: later tell rekado when qt-5.9 comes out its supposed to be a LTS release so I plan on upgrading our monolithic qt to 5.9 also
<sneek>Will do.
<efraim>sneek: botsnack
<efraim>trying to flash the firefly board, huge PITA
<efraim>'upgradetool' is binary-only, i386, rkflashkit doesn't want to recognize partitions
<efraim>and i forgot to check rkflashtool
<efraim>rkflashtool also doesn't want to recognize it
<amz3>héllo the-most-active-channel aka. #guix
<fusion809>Hey I'm gonna try creating a BIOS boot partition in case it makes a diff. The prob is how to do so with fdisk. cfdisk and gdisk can't be used for MBR so fdisk is my only option to my knowledge.
<fusion809>Saw but I can't make out from it how I'm meant to create a BIOS boot partition.
<fusion809>Seems to indicate I can't
<fusion809>Guessing I can as otherwise I'll have to use the potentially problematic GPT table
<amz3>fusion809: you are trying to install guixsd on a UEFI device?
<fusion809>Na QEMU VM. BIOS boot partition I thought wouldn't be used for UEFI devices. I know on my host OS I have UEFI firmware and I have no BIOS boot partition just a EFI and root partition.
<fusion809>In QEMU VMs, to my knowledge, BIOS boot is the only such firmware available.
<wigust>Hello. I have a trouble with UEFI installation too, but on hardware machine. I cannot configure grub in /etc/config.scm: (bootloader (grub-configuration (device "/dev/sda") (grub grub-efi))).
<wigust>unbound variable grub-efi
<amz3>wigust: can you paste your config.scm please?
<amz3>wigust: unbound variable means you did not import the thing correctly
<amz3>wigust: forget about pasting, I won't be able to troubleshoot anyway
<amz3>ACTION take the exit lane
<efraim>finally managed to flash the firefly
<efraim>rekado: i'm still showing gcompris-qt@0.70 vlc@2.2.6 avidemux@2.6.12 for qt@5, gcompris-qt was a pain, not sure what vlc needs, graph is actually too demanding for my machine
<wigust>without (grub grub-efi) i got grub-install: error: gnu/store/hash-grub-2.02/lib/x86_64-efi/ doesn't exist. Please specify --target or directory.
<wigust>There is no directory gnu/store/hash-grub-2.02/lib/x86_64-efi
<wigust>just grub/i386-pc/
<fusion809>I'll be, with BIOS boot partition the install goes fine.
<fusion809>The partition was labelled a Linux partition with fdisk and I formatted with as ext2.
<fusion809>I suggest the docs get updated. Should I reported this to bug-guix?
<quiliro>fusion809:i do not understand what is a bios boot mean a boot partition or a bios partition?
<fusion809>The one labelled "BIOS Boot" in cfdisk. mentions what a BIOS boot partition is used for.
<fusion809>in cfdisk and gdisk when dealing with a GPT partition
<wigust>I think i found missing import thanks to amz3 (use-package-modules bootloaders ...)
<amz3>wigust: :)
<wigust> is a good example
<fusion809>I labelled my root partition as my-root using mkfs.ext4 but for some reason it's having difficulty recognizing this partition and booting it
<fusion809>GRUB starts fine though, it's just when booting the system with GRUB where things go pear-shaped
<wigust>fusion809: did you check that labels are present? lsblk -f
<fusion809>Nope haven't I'll give it a go.
<fusion809>Doesn't seem to be present. Please don't tell me I'm gonna have to go through the slow process of running guix system init again. Please tell me I can label the partition without destroying the data on the partition
<fusion809> seems to suggest I can without loosing data, so I'll give it a go
<fusion809>Oh goodie it's booting fine now. Thanks
<wigust>fusion809: np
<catonano_>scrapy is a really cute framework. It' s too bad that it' s not packaged in Guix :-/
<wigust>So, i can boot into grub, but it want kernel to be in /store/... not in /gnu/store as i have. But it's not that important as i can change it while booting for now. There is an error: no suitable video mode found. Booting in blind mode.
<amz3>what's the purpose of the guix-patches mailling list? Otherwise said, if I have a patch I must send it to that mailling list?
<amz3>a patch for adding a package?
<janneke>amz3: yes
<janneke>the purpose is communication with debbugs
<janneke>and it helps keeping guix-devel clean, i guess
<amz3>so where should I send my package patch?
<amz3>there is a lot of activity!
<wigust>is there a faster way to `guix system init /mnt/etc/config.scm /mnt`?
<janneke>amz3: :-)
<janneke>wigust: are you using substitutes?
<wigust>What? I'm trying to install it via usb
<janneke>wigust: okay, so what is being slow?
<wigust>janneke: I failed with grub configuration. So I boot to liveusb guixsd. As I see `guix system init` reinstall all the things.
<wigust>So, I think I use wrong `guix flag`
<janneke>wigust: if you only change a configuration, you can `guix system reconfigure'
<wigust>janneke: but i'm in liveusb. So `guix system reconfigure /mnt/etc/config.scm /mnt`?
<wigust>or chroot inside?
<wigust>`chroot /mnt`?
<wigust>I cannot boot because root is on btrfs subvolume, and I set grub to search root on label (btrfs subvolume doesn't have label).
<janneke>wigust: ah, yes. not sure if you can already use guix reconfigure, then
<davidl>wigust: I had the same issue before when installing to btrfs subvol. There's a mount options setting for the file system part of the system declaration. See here:
<davidl>not sure if that helps though. I am trying to install to the btrfs root instead now.
<davidl>wigust: you can also try "btrfs set-default subvol=257" or similar but that didn't when I did it.
<davidl>didn't help*
<quiliro>i still have the problem of my-root not beinfg
<quiliro>i have mounted with
<quiliro>mount LABEL=my-root /mnt
<quiliro>so the label is correct
<quiliro>i have mounted on reboot
<wigust>i changed grub config manually on /boot and i have something like tty but not tty
<wigust>logo on top gnu/linux-libre
<wigust>is it ok?
<wigust>i also see kernel messages
<wigust>but cannot do anything except `,help`
<wigust>no init was called i guess
<davidl>wigust: I also tried grub command line, didn't work here either.
<davidl>wigust: I would try and use the mount options string if I were you in the system declaration and reinstall.
<wigust>davidl: tryed (options "subvol=guix-root"), but nothing changed in grub.conf
<wigust>only manually specified a could boot in kernel, but no init
<davidl>wigust: in "man mount" it says to use subvolid=<ID> or subvol=<PATH> where path is relative to the btrfs root. So I would check the subvolid with some btrfs command and then use that.
<davidl>wigust: but Im actually not 100% sure that these mount options are actually options for the mount commant and not something grub-specific.
<quiliro>i have a similar problem but not with btrfs
<quiliro>it is just ext4
<quiliro>has anyone had a successful guixsd installation on the past week?
<buenouanq>I successfully installed GuixSD on a Macbook 2 and a half weeks ago.
<davidl>qiuliro: it has finished saying it is successful a couple times, but then the luks-partition have become corrupted so it won't decrypt after reboot.
<brendyyn>I'm getting this error installing GuixSD: guix system: error: build fialed: getting attributes of path `/gnu/store/73c9...utils.scm': No such file or directory
<efraim>48 minutes for make on the firefly board
<jackhill>Would it be possible for guix pack to use relative paths?
<quiliro>davidl: test without encryption in order to see if it is that
<davexunit>jackhill: maybe, but it's a really difficult problem.
<quiliro>buenouanq: the problems with installation have been this week
<jackhill>davexunit: Is that because of runtime linking problems?
<buenouanq>have we been using --fallback
<quiliro>buenouanq: it is good news to exorsize mac!
<davexunit>jackhill: there's a variety of problems, but linking shared libraries is one of them. yeah.
<buenouanq>yes, I was very pleased to get this working
<quiliro>buenouanq: who are you asking? me?
<jackhill>davexunit: thanks! I really which I had guix available on this system…
<quiliro>buenouanq: about fallback
<davexunit>jackhill: someone has done work on relocatable binaries, but I haven't used it.
<ng0____>I procastrinated a bit more today.. is someone interested in a newsreader? I can finish the package and make a proper patch
<davexunit>I have made relocatable binaries for very specialized programs, but have no generic system.
<rekado>oh, I broke cross-compilation!
<sneek>rekado, you have 1 message.
<sneek>rekado, efraim says: when qt-5.9 comes out its supposed to be a LTS release so I plan on upgrading our monolithic qt to 5.9 also
<rekado>I have a fix but will only be able to test and push it tonight
<rekado>I only noticed because I’m testing my changes to add cross-compilation support to cmake-build-system.
<rekado>coming soon!
<ng0____>rekado: in did you mean that I could keep the first version I had of neo2-kbd and just add the native-search-path to kbd so that it can pick up neo2-kbd when installed?
<rekado>ng0____: not sure about what “the first version” refers to (can’t look it up right now), but I think the answer is a tentative “yes”: the idea is to make kbd look for other layouts in directories given with an environment variable.
<ng0____>the first version refered to neo2 not being an input-only-source but an extra package
<rekado>fixed the cross-compilation bug.
<rekado>so, neo2 ought to be a separate package that only installs a keymap into its output.
<ng0____>that was my initial idea
<ng0____>and I just add the search path to kbd.
<ng0____>thanks, that should be much more pleasant than this hack.. it's a hack whic hworks, but we have better tools ;)
<ng0____>which emacs functionality did I use when I was able to do M-e in the git commit view / editor and have it changed to 'edit' when doing rebase? or was that just maggit?
<paroneayea>ng0____: not sure about M-e but magit's interactive rebase is pretty incredible
<ng0____>I know git by hand better than tools wrapped around it.. I thought there was something I forgot to install at some point.
<ng0____>I think it was magit.
***Acou_Bass is now known as eddie
***eddie is now known as Guest74342
***Guest74342 is now known as Acou_Bass
<quiliro>i still have the error
<quiliro>i have been trying to install guixsd for a week
<quiliro>then i will make it a mirror of hydra
<quiliro>then i can go home to offline install
<quiliro>would someone take a look at it?:
<bavier`>quiliro: did you properly label your partition as 'my-root' when you created the fs?
<quiliro>bavier`: yes, and the proof is that executing the following command mounts it on /mnt:
<quiliro>mount LABEL=my-root /mnt
<quiliro>bavier`: am i wrong?
<bavier`>quiliro: yeah, seems right
<quiliro>what should i investigate
<quiliro>i have to do something
<quiliro>waiting is making me desperate
<quiliro>on /mnt i have all the directories usually existing in a working distro and i had no errors on installation (guix system init /mnt/etc/bare-bones.scm /mnt)
<quiliro>i could erase everything and start from 0
<quiliro>is it necesary?
<bavier`>quiliro: could you paste your config file?
<quiliro>the reported one is on: ... it's
<quiliro>bavier`: ^
<bavier`>quiliro: nothing looks immediately off to me, but I don't know where to go from here
<quiliro>could it be the lack of (lsh-service #:port-number 2222) . being it replaced with (service openssh-service-type (openssh-configuration (port-number 2222)))
<quiliro>bavier`: what would you do? start from 0?
<bavier`>that might help, but probably not
<quiliro>if there il nothing else to do.....
<quiliro>ok, since there are no suggestions, will erase partitions and format
<reepca`>quiliro: from which device does mount LABEL=my-root /mnt work? The device with the GuixSD installer on it?
<quiliro>reepca`: yes
<reepca`>out of curiosity, which file under /dev does the device with the GuixSD installer on it report itself as?
<quiliro>reepca`: that is what lsblk reports
<quiliro>reepca`: is that what you mean?
<quiliro>but 0.12 detected hard disks before usb
<quiliro>so the config file could be using /dev/sda
<reepca`>and is that device still plugged in when you attempt to boot from the newly-installed GuixSD?
<quiliro>reepca`: could that be the reason it does not work? do you think it would be good to try 0.12?
<quiliro>reepca`: no and yes
<quiliro>tested both ways
<quiliro>but cannot test if it is recognized as sda or sdb when booting with the hard disk because it will not give me bash shell....only guile shell
<quiliro>you can see the result of the installation done right before rebooting on:
<reepca`>hm... well, the installer and the installed system should both be detecting devices in the same way.
<quiliro>as i reported on
<reepca`>when it gave you that "Error: in proceedure scm-error", did it give a stack trace?
<quiliro>reepca`: what is that?
<quiliro>it would be great if someone would try to install to see if it is reproducible
<reepca`>A bunch of messages above the "error: in procedure scm-error" part
<quiliro>not that i remember
<quiliro>do you want me to reboot with the disk to see?
<reepca`>worth a shot
<wigust>liveusb and installed kernel are the same thing? Cannot boot to installed system with kernel panic
<reepca`>apparently someone had the same problem some time ago
<reepca`>wigust: the configuration used for the installer can be found in gnu/system/install.scm (at the end) in the guix source.
<quiliro>reepca`: read the solution to that irc chat
<reepca`>That said, I don't see any mention of a different kernel, so unless you changed something about that in your config.scm, they should be the same
<quiliro>reepca`: it was a problem with (device "/dev/sdx")
<quiliro>but my config does not have that error
<quiliro>mine is defined as /dev/sdb
<quiliro>what is wrong with installing with 0.12? would that be a problem?
<wigust>reepca`: thanks, will check
<quiliro>would i need to guix pull before guix system init?
<quiliro>reepca`: how did you find that chat particularly that date?
<reepca`>searching for "failed to resolve partition". Apparently that error message is particular to whatever we're using.
<quiliro> i found "waiting" too
<reepca`>have you tried using device names instead of labels, just in case that works?
<ng0__>neo-kbd works.
<ng0__>anyone with some minutes time looking for review?
<quiliro>reepca`: yes...same result but with the device name instead of the label
<quiliro>same error i mean
<quiliro>it is actually: failed to resolve partition label "my-root"
<quiliro>in the case of label
<quiliro>i do not remember the exact message in the case of uuid or device
<quiliro>but i did try those
<Apteryx>Could it be desirable to be able to tell a search path specification: please append a separator at the end? I'm finding such use case for Emacs
<Apteryx>I think there are other cases too, such as INFOPATH
<Apteryx>It usually means: whatever specified + the builtin lookup locations
<Apteryx>so if I'd set EMACSLOADPATH="some/elisp/library:" it would also look under emacs/share/25.2/lisp/... for example, to find its builtin lisp.
<wsddn>Has anyone tried or had any success running a rootless X in their user account?
<reepca`>quiliro: just to be clear, when I said "device names", I meant what mark_weaver said in that log you posted - (title 'device) and (device "/dev/sdX").
<quiliro>(bootloader (grub-configuration (device "/dev/sdb")))
<quiliro>(device "my-root")
<quiliro>(title 'label)
<quiliro>would that be correct for label?
<reepca`>for label, yes, if "my-root" is the label of the device in question.
<quiliro>and if my hard disk is sdb
<quiliro>lsblk reports sdb1 8:17 0 16G 0 part /mnt
<quiliro>being mounted with LABEL=my-root
<quiliro>did guix system init /mnt/etc/bare-bones.scm /mnt --fallback
<quiliro>after changing
<quiliro>(lsh-service #:port-number 2222)
<quiliro>(service openssh-service-type
<quiliro> (openssh-configuration
<quiliro> (port-number 2222)))
<quiliro>as /etc/configuration/bare-bones.scm has in the new 0.13
<quiliro>hope it works now
<quiliro>else will wait until a more stable guixsd
<quiliro>do not want to take more of your time
<quiliro>i can use more of mine but devs work is precious
<ng0__>don't be like that. if there's a problem, people help you :)
<ng0__>*will help
<quiliro>ng0__: are you sure i am not using time that would be more valuable in other tasks?
<ng0__>those who help you do so because they want to, not because they are pressured into it
<quiliro>of course that if i learn, i can help....have to weigh benefits
<reepca`>I'm going to attempt to init onto my other hard drive that I got specifically for setting up GuixSD now, I've been putting it off for awhile. Will use the device label and see how it works for me.
<quiliro>sure...i know it is good will...but i do not want the project to loose time
<quiliro>reepca`: thanks
<quiliro>populating '/mnt'...
<quiliro>Installing for i386-pc platform.
<quiliro>Installation finished. No error reported.
<quiliro>root@gnu ~#
<quiliro>shall i reboot?
<reepca`>I guess
<quiliro>same error...
<fusion809>Ah folks I've noticed something odd about GuixSD and I'm wondering if it's an oversight or deliberate. For some reason on my recently installed GuixSD system my PATH variable for the root account is shown in this screenshot What's relevant is not what's in it, it's what's not in it: /root/.guix-profile/sbin. Without that useradd, usermod and other important commands are not
<fusion809>available from the command-line. Most distros have these commands in the root PATH so I'm guessing this is an oversight, but I'm here to ask am I right or wrong in this belief?
<fusion809>If it's an oversight I'm happy to file another bug report, but if not I'd rather not irritate the devs by filing a false bug report.
<fusion809>I have not edited ~/.bash_profile, ~/.bashrc, /etc/profile or any other shell-relevant files. The only thing I've done it is set my root password and installed a few packages (vim, screen, git, wget and curl) with guix. So this seems like this is the default root $PATH.
<efraim>fusion809: go ahead and file the bug, it looks like it should be there
<fusion809>Thanks, I thought everyone was either confused what the problem was or not sure whether it was an oversight.
<ng0__>which commands? useradd and usermod are not really persistent on GuixSD
<efraim>root's profile's sbin was missing in either case
<lfam>fusion809: Did you install (as root) any packages that include an sbin directory?
<efraim>our webkitgtk doesn't target wayland
<lfam>fusion809: The value of PATH is generated based on which packages the user has installed. So, if you install something with an sbin directory, then your PATH will include that directory next time you open a login shell
<fusion809>lfam: not so far. That's not the problem it's that /root/.guix-profile/sbin contains several binaries that I, for one, think should be in the $PATH by default like groupadd, groupdel, cfdisk, fdisk, fsck, etc.
<lfam>fusion809: groupadd and groupdel won't work on GuixSD. The disk-related programs, on the other hand, are most likely found in /run/current-system/profile/sbin, which is in your $PATH
<lfam>You're free to install groupadd etc, but they won't do the right thing
<reepca`>fusion809: See section 6.2.2 of the manual, specifically "packages" - "The set of packages installed in the global profile, which is accessible at ‘/run/current-system/profile’"
<fusion809>I'll be .... I rebooted and now /root/.guix-profile/sbin is in the root $PATH. Might have been similar to what ya were talking about lfam. Seems odd though as I thought these commands came by default with GuixSD, they are found in most distros by default, but I suppose yas take barebones to a whole new level
<lfam>fusion809: We handle groups and group membership declaratively at the system configuration level.
<lfam>Please refer to the manual section User Accounts:
<lfam>Also, you don't need to reboot to check changes to your login shell. `bash -l` will invoke a new login shell. It's much faster for experimenting
<reepca`>notice how (at least in the default configurations) packages includes %base-packages. My question is why /root/.guix-profile/sbin has stuff in it if the root user didn't install it there - is /root/.guix-profile initially a symlink to /run/current-system/profile?
<ng0__>woo, after 5 months I ca ndelete the neo2 branch :)
<quiliro>i have noticed the partition was not bootable
<quiliro>i made it bootable and restarted
<quiliro>i will see the results
<lfam>reepca`: Not sure. If I uninstall all of root's packages that include an 'sbin/', then root's PATH no longer includes an entry for '/root/.guix-profile/sbin'
<reepca`>perhaps one of the few things fusion809 installed had a propagated-input that installed one of the mentioned binaries to /root/.guix-profile/sbin?
<bavier`>paroneayea: lots of exciting talk on HN right now :)
<reepca`>although if it had, that should have added it to the PATH anyway... my guess would be that a new shell wasn't started since the packages that put stuff in /root/.guix-profile/sbin were installed.
<quiliro>did not work...will try again with the boot activated...perhaps grub will then recognize it
<quiliro>or can i grub-update or grub-install?
<ng0__>guix system reconfigure, quiliro
<reepca`>updated guix and now the "failed to install locale" madness has returned...
<lfam>reepca`: Also update the locales in the environment where the daemon runs. So, on a foreign distro, do it as root and restart the daemon. On GuixSD, at the system level.
<reepca`>lfam: did that already
<lfam>Well, at least it doesn't break anything, right?
<reepca`>yep, that's always good.
<lfam>The locales / glibc versions need to stay in sync to avoid this issue. It's annoying but much better than it used to be, where incompatible locales simply caused glibc to abort
<lfam>Or, "when incompatible..."
<reepca`>I updated the installed locales (2.24 -> 2.25), still happening. Do I need to downgrade the installed locales, then?
<reepca`>how would one check which glibc is being used by the daemon and guix?
<paroneayea>bavier`: indeed :)
<paroneayea>mediagoblin and activitypub on HN homepage at the same time today!
<lfam>reepca`: I use something like to start exploring:
<lfam>guix gc --references $(realpath /proc/$(pgrep guix-daemon | head -n1)/exe)
<quiliro>guix system init /mnt/etc/bare-bones.scm /mnt --fallback
<quiliro>for the 15th time
<reepca`>hmm, that says the daemon's using glibc-2.24. Do the locales and glibc numbers indicate same versions?
<quiliro>in the 9th reboot
<lfam>reepca`: Yes, they need to match
<quiliro>after 7 days of cruising through the seas of GuixSD
<reepca`>that's strange, then - locale stuff is what prompted me to upgrade to 2.25
<quiliro>sometimes 24 hours in a row
<ng0__>quilior: if you have a system inited, you don't need to do it again. reconfigure is for the grub thing you wanted
<lfam>reepca`: You're still using the old guix-daemon, from Guix 0.12.0
<davidl>quiliro: are you using GPT or MBR for partition table?
<quiliro>ng0__: i think it is mbr
<quiliro>how can i test?
<reepca`>So wait, does this mean that if I'm at glibc-utf8-locales version 2.25 I'll have the daemon complaining about locales, and if I'm at version 2.24 I'll have guix complaining about locales?
<ng0__>parted /dev/sdfooX
<ng0__>then p
<ng0__>will show your partition table and information about it
<davidl>quiliro: idk, but if you just want it running you might be better off with MBR/DOS. If you use GPT you should possibly use separate /boot and set the grub_bios on flag with parted on the boot partition.
<ng0__>I'm attemoting to fix now
<lfam>reepca`: Not sure, but you should just use glibc and glibc-utf8-locales 2.25 everywhere. That should get rid of the warnings.
<davidl>like 1Mib to 3MiB partition for a boot with GPT.
<quiliro>ng0__: Partition Table: msdos
<reepca`>alright then, I'll look at the binary install instructions and see where to plop an updated daemon
<lfam>reepca`: Are you using Guix on another distro? How do run the daemon?
<reepca`>lfam: yep, Ubuntu, systemd stuff
<lfam>I mean, "How do you run the daemon?"
<lfam>Okay, then you just need to update root's guix package and restart the daemon. Make sure your systemd service file executes a symlink like '/var/guix/profiles/per-user/root/bin/guix-daemon' and *not* an absolute path to /gnu/store
<lfam>Our old service file had a bug that caused users to use old guix-daemons forever until they were garbage collected :/
<davidl>quiliro: I have had a very strange boot issue once which I solved after a very long time by "wipefs --all /dev/sda" and "head -c 3145728 /dev/urandom > /dev/sda; sync" and then repartitioning.
<reepca`>lfam: updating root's guix package here meaning guix pull, right? I've had a weird history with that - for the longest time, it wouldn't use guile2.2, and then I did "guix package -u guix" and now I think it does. Also, is savannah doing okay? It's been at 26KB/s for awhile here.
<lfam>reepca`: `guix pull && guix package -u guix`
<lfam>reepca`: Savannah only serves the code of Guix itself. The rest of what we serve comes from another server, which is pretty slow, and we work around that by putting a mirror in front of it. If the file you requested is not in the mirror cache, then you could experience slow transfers.
<lfam>Soon this will improve. We have a new, more powerful, machine that should improve the speed and reliability of substitute distribution
<reepca`>Said slow speeds were happening during guix pull
<lfam>Then it's Savannah being slow
<ng0__>is it worth fixing SLiM when it is abandoned and we will hopefully switch to a new, maintained, default login manager soon?
<sneek>I'll keep that in mind.
<ng0__>sneek: forget
<ng0__>sneek: forget is it worth fixing SLiM when it is abandoned and we will hopefully switch to a new, maintained, default login manager soon?
<sneek>Consider it forgotten.
<reepca`>sneek: botsnack
<mbakke>ng0__: what's broken with SLiM?
<mbakke>SDDM is supported in Guix, I've used it for a long time
<ng0__>nevertheless it is abandonware
<ng0__>"choice" of keyboard layout or the lack of it is 'broken' in SLiM:
<ng0__>also SDDM, which feels better to fix because it isn't abandonware
<mbakke>I can't log out right now, but will check if keyboard layout switching works with SDDM next login
<ng0__>i used it, i did not file the bug for SLiM *and* SDDM for no reason
<ng0__>all we can do at the moment is set console-layout, which does very little for login managers which rely on other ways to set it
<ng0__>should read the sddm-service again, maybe it is configurable and I missed it.
<mbakke>seems like it's still not properly supported upstream:
<ng0__>kde project complained in december about it again
<catonano`>good evening, guixers
<ng0__>i have this lxdm collecting dust.. maybe it has better support. or I try writing a service for the other login manager we recently got
<reepca`>lfam: pulled, updated, and restarted the daemon, still getting that warning.
<lfam>Okay, you'll have to keep debugging on your own :) These warnings don't bother me very much, since I don't need locale support in the daemon
<lfam>Or maybe somebody else has an idea
<snape>mbakke: is your comment on cpuid for both substitutes or only the second one?
<snape>because it seems to me that at least the first is obvious
<mbakke>snape: just the chmod thingy
<quiliro>reepca`: did you try installation of GuixSD as you said on physical machine?
<ng0__>Okay. I'm trying lightdm-service, unless someone else is already working on that
<rekado>ng0__: I have dvorak in slim by providing an evdev configuration section to the xorg configuration.
<reepca`>quiliro: got sidetracked fixing a locale problem. initing now.
<ng0__>rekado: would you mind sharing that? if you have no time to add it to the documentation (I think it should be there for slim layout changes) I can do it
<rekado>ng0__: sure, just a sec
<quiliro>is it mandatory to reboot after partitioning?
<efraim>lfam: I got the firefly hooked up finally, had to jump through hoops to flash it with their ubuntu image
<efraim>4gb of ram and emmc is much much faster than 2gb of ram and an sd card
<lfam>efraim: How's it runinng?!
<lfam>Or, running :)
<efraim>and the extra 2 A72(?) cores
<efraim>i ran make on the guix repo with guile-2.2 in 45 minutes
<efraim>compared with 6 or 8 hours on the odroid
<lfam>Wow, that's awesome
<efraim>with 2.0 it was about 15 minutes, on par with my x120e
<lfam>That's faster than my low-power x86_64 board.
<ng0__>so it's definitely worth the 120 USD
<rekado>quiliro: no.
<lfam>Does the board require a special kernel or can you use the Linux mainline?
<rekado>ng0__: the key part is that I’m changing the slim-service-type, overwriting its configuration
<rekado>to provide a new startx command
<efraim>its running 4.4.52, haven't gotten around to checking if its special or mainline
<ng0__>cool, thanks!
<rekado>which has its own xorg-configuration-file with extra configuration appended
<lfam>efraim: That's pretty recent for a special kernel. Only a couple months old
<quiliro>rekado: thank you
<rekado>ng0__: note that this snippet will kill middle mouse scroll on an x200s.
<lfam>It's from February
<quiliro>rekado: are you the only one that knows that or are you the most kind?
<rekado>(that’s configured via evdev as well, so the minimal evdev configuration breaks it)
<ng0__>hm.. I use middle mouse often, but I can find a way to restore it I'm sure
<lfam>efraim: I guess I'll decide between the firefly and this thing based on a Marvell SOC:
<rekado>ng0__: you can do that in the same way, i.e. by extending that xorg snippet for evdev
<quiliro>well... i have done this so many times, i have developed a scripts for installation
<davidl>quiliro: lol me too ^^
<rekado>ng0__: I use middle mouse very often, too, but I haven’t gotten around to fixing it yet.
<quiliro>davidl: publish yours
<davidl>ok :-) 10 sec
<rekado>I’m confused by the need for a script.
<rekado>you don’t need to repartition after a failed attempt
<rekado>I’m baffled that the same system needs to be reinstalled for several times.
<quiliro>rekado: in order to do repetitive typing
<rekado>are you documenting your problems? If so, please share your problem report with us with suggestions how this could have been avoided in your case.
<quiliro>i have done it about 19 times
<davidl>quiliro: where's the command to post to splurge from command-line?
<ng0__>there's this wegtpaste for example
<ng0__>or simple curl interacftion with
<davidl>ng0__:nice, thx.
<ng0__>the examples are bash, if you are on GuixSD this shouldn't be a problem.
<rekado><command> | curl -F 'sprunge=<-'
<quiliro><command> | curl -F 'sprunge=<-'
<rekado>(that’s what I found in the log)
<quiliro>curl is <command> | curl -F 'sprunge=<-'
<efraim><command> | wgetpaste
<ng0__>I'm sitting on a client for ptpb for a long time now due to the unresolved curl guix specific problem we have.. hope this will be fixed one day.
<rekado>what is that problem?
<rekado>(there are too many bugs to keep them all in mind :))
<ng0__>i think it was
<quiliro>rekado: that is the report I filed
<davidl>quiliro: not very pretty but..
<rekado>ng0__: ah, that’s actually … not really a bug (in the traditional sense) and also pretty easy to fix
<rekado>ng0__: but it will cause quite a few rebuilds
<rekado>ng0__: I guess I could fix it in core-updates
<rekado>ng0__: I suggest just patching libcurl.
<rekado>it’s ugly but since we have no default path we have to use an environment variable.
<ng0__>I think we had that discussion wrt darcs and pbpst
<ng0__>and patching the source is the quick solution I guess
<rekado>I think it’s the only solution that really works.
<ng0__> I need to rebase and update my pbpst
<rekado>users need this variable anyway for better control over certificates.
<ng0__>would be highly appreciated to have this feature-bug gone :) getting rid of old branches here is what I want.
<ng0__>I even helped to work on pbpst for systems which can not tup to thhe full extent, and then I had to pause because of cURL
<mbakke>the hydra evaluation of 'master' still fails with the same error:
<quiliro>davidl: thank you. mine is extremely specific for this machine and me....license for your script is?
<davidl>quiliro: GPLv3, not sure it counts as licensable though xD
<quiliro>of course it does
<efraim>firefly is compiling webkitgtk using all 6 cores
<rekado>mbakke: I think that was my fault
<rekado>mbakke: I may have fixed it with 3a62dd6b22c054f2e8d6df38d493def52020e714
<quiliro>davidl: i will certainly make use of the 4 freedoms with that script!
<quiliro>that is something i am capable of using them on....other stuff only 2 freedoms
<rekado>mbakke: I just asked hydra to reevaluate.
<lfam>mbakke: Did you recent Kodi update fix the subtitle vulnerability?
<mbakke>lfam: yes
<lfam>efraim: That's awesome. I didn't realize it could use the A53 and A72 cores heterogenously
<efraim>i wasn't sure it would be able to
<lfam>I had read some discussion of the SOC several months ago where it was claimed this would not work
<efraim>my understanding was that it shouldn't work
<efraim>it could be there's some black magic going on in the kernel
<lfam>It's one thing I like about that Marvell board: all the CPUs are the same
<lfam>But it's a lot more expensive
<ng0__>could the raspberry pi 3 work with some degraded gpu performance?
<lfam>We should evaluate these boards for a little while and try to get their makers to donate some to the build farm :)
<efraim>it has the sata slots ready, if i want that i would need to buy an M.2 to sata converter
<lfam>ng0__: I think it should work fine unless you want to play 3d games.
<lfam>Or do video editing / production
<ng0__>or use GNOME
<efraim>another possible problem with the RPi3 is it only has 1GB of ram
<lfam>Right... several packages require more than that to build.
<ng0__>but as a potential dd system on it which is ready to run it is cheap
<lfam>The real issue with the RPI3 is that it only has the A53 processors, which are really embedded / mobile processors. For the build farm, we really want some A72s
<quiliro>why :
<ng0__>oh, I was just thinking about general deployment once GuixSD can run on this arch
<lfam>But, for non-compute intensive work, I think it should be fine, and already much faster than the 32-bit ARM systems
<quiliro>$ python --version
<quiliro>Python 2.7.9
<quiliro>$ guix package -I python
<quiliro>warning: failed to install locale: Invalid argument
<rekado>quiliro: python3 only installs the “python3” executable
<quiliro>python 3 is installed but reports python 2.7.9
<rekado>quiliro: that’s an upstream decision
<rekado>(we just don’t challenge it.)
<quiliro>rekado: which is installed?
<rekado>quiliro: if you want the “python” executable to mean “python3” you can install “python-wrapper”
<rekado>quiliro: what does “which python” tell you?
<quiliro>i need python 3
<lfam>Where is Python 2.7.9 coming from? That's not a recent version.
<snape>davidl: I've read your script, and there's something I don't understand: why do you 'mkpart primary' on a GPT disk? There is the exact same thing in Guix, which is why I ask. I thought primary partition didn't exist with GPT.
<quiliro>$ which python
<rekado>well, that’s not from guix :)
<ng0__>I've sent a pullrequest to to include Guix in the updated list of systems where perl is available
<lfam>ng0__: I think for general deployment the RPI3 will be very popular :)
<ng0__>build farm will be more populated by fireflies I guess
<lfam>I'd really like to buy the 96-core Cavium servers, but that may be out of our reach :)
<ng0__>do you have a link?
<lfam>There are other implementations
<davidl>snape: that part was just copied from and edited minorly, I'm also pretty sure you don't need to use it like that since primary is just MBR and not GPT stuff as you say.
<ng0__>great scot
<quiliro>waiting for partition 'my-root' to appear
<quiliro>ERROR: In proceedure scm-error:
<quiliro>ERROR: failed to resolve partition "my-root"
<quiliro>those are the exact error messages
<rekado>quiliro: how many disks do you have that are labelled “my-root”?
<davidl>snape: also, I use set boot 1 on and also set 1 bios_grub on which is superflous too I think but that was just copied.
<quiliro>for GuixSD installation
<quiliro>i removed the other one long time ago...i only have one
<rekado>and you are certain that it actually has this label?
<quiliro>rekado: i mounted it with
<rekado>quiliro: you know, if the label thing doesn’t work for you, we might try the uuid.
<quiliro>mount LABEL=my-root /mnt
<quiliro>rekado: i used uuid an had a similar problem
<snape>davidl: mklabel gpt creates a new disk label, of type GPT
<quiliro>so i am absolutely certain it is right
<rekado>quiliro: and lsblk confirms that there’s really only one with that label?
<snape>davidl: nevermind my last sentence, I misread what you said
<rekado>quiliro: are you using full disk encryption?
<davidl>snape: yes. The word primary is an MBR thing, and I intended to use GPT so I don't think writing primary changes it, or you think it doeS?
<davidl>snape: ok :-)
<snape>No, I don't think it does, but it's confusing :)
<rekado>quiliro: I find this really weird. I have never seen these problems with a simple config like that.
<quiliro>lsblk does not show labels
<quiliro>only mountpoint
<quiliro>rekado: ^
<mbakke>quiliro: try "blkid"
<quiliro>/dev/sdb1: LABEL="my-root" UUID="4e7672ee-fa31-43ee-ad59-88372d1e8817" TYPE="ext4" PARTUUID="41bf2ef9-01"
<quiliro>/dev/sdb2: LABEL="swap" UUID="0bf37311-9b4d-4d29-8d2b-65b809db6de8" TYPE="swap" PARTUUID="41bf2ef9-02"
<rekado>quiliro: what’s sda?
<quiliro>the usb installation image disk
<lfam>I'm trying to build the latest version of python-cryptography, but it fails in the check phase when it can't find the 'packaging' module:
<quiliro>should i redownload installation image and re-dd it?
<lfam>I tried to comment out the requirement for this module, but then the module is actually used, so it fails anyways
<lfam>However, this module should be part of setuptools, so I'm confused :)
<rekado>quiliro: I don’t think anything is wrong with the installation image
<rekado>quiliro: (besides, you probably did check the signature on it before using it.)
<rekado>quiliro: could you please also show me “sudo fdisk -l”?
<quiliro>rekado: no :-X
<quiliro>i did not check
<quiliro>Device Boot Start End Sectors Size Id Type
<quiliro>/dev/sdb1 * 2048 33556479 33554432 16G 83 Linux
<quiliro>/dev/sdb2 33556480 39876479 6320000 3G 82 Linux swap / Solari
<rekado>quiliro: I’m very sorry you are having such a bad time with the installation :(
<reepca`>alright, init says it's succeeded, let's see how booting works! later
<rekado>quiliro: I cannot think of a good reason why GuixSD would claim that the partition cannot be found.
<rekado>my attempts to teach cmake how to cross-compile haven’t led to success yet.
<rekado>it’s calling the cross compiler, but it fails to link
<rekado>the error looks familiar; it’s like when you install “gcc” instead of “gcc-toolchain”
<rekado>arm-linux-gnueabihf-ld: cannot find crt1.o: No such file or directory
<quiliro>rekado: i mounted it with that label
<quiliro>sorry..i already said that
<rekado>that should be provided by the cross libc, but the cross libc isn’t even added to the inputs. Quite strange.
<rekado>quiliro: is that the only machine on which you can test this?
<rekado>quiliro: have you previously succeeded in installing it in a virtual machine?
<quiliro> i can go to another too...good idea...but i do not have another disk
<rekado>well, since the disk appears to be closely related to the problem I don’t think moving it to another computer would fix it.
<quiliro>i will get another disk and try another day then
<quiliro>i really appreciate all the help you people have provided
<rekado>quiliro: I admire your patience!
<quiliro>thank you :-)
<rekado>I hope we can figure this out and your experience will prevent others from suffering the same problems.
<quiliro>i will test it once more on the other machine
<davidl>eiro: hello
<rekado>I just cross-built bamtools for armhf (bamtools uses cmake).
<rekado>ACTION sends patch
<janneke>i tried 0.13 armhf on an armv7l machine and got Illegal instruction -- wrong arm?
<lfam>janneke: Which machine?
<janneke>lfam: i got a login from a friend, and have no clue about arm
<janneke>what can i do besides: uname -m?
<lfam>janneke: How about `cat /proc/cpuinfo`?
<janneke>lfam: ARMv7 Processor rev 2 (v7l)
<lfam>Hm... any ideas, efraim?
<rekado>mbakke lfam I’m confused about hydra again…
<lfam>rekado: What's up?
<rekado>when will hydra update from git?
<rekado>I broke cross-compilation earlier and fixed it today
<lfam>rekado: Typically, each evaluation checks out HEAD
<rekado>I started an evaluation about two hours ago, but it still fails with the same error
<lfam>However, wasn't mark_weaver talking about doing a special evaluation so we could get a substitute for guix@0.13.0?
<lfam>Perhaps something was changed that is not reflected in the web UI
<lfam>Do you remember the discussion? Perhaps on guix-sysadmin?
<rekado>yes, I do
<lfam>Woo-hoo! I finally figured out how to build the latest python-cryptography.
<mbakke>rekado: could it be related to cdb3f734cfb93a5689d9fec3558811fe652749f4 ?
<rekado>mbakke: no, I think it’s 7b3318e34f4e2743254a88b908859901db960e9a
<rekado>which I then fixed with 3a62dd6b22c054f2e8d6df38d493def52020e714
<rekado>it’s fine to cross build for mips, but the call of cross-gcc is wrong.
<rekado>it says “invalid keyword” followed by a package.
<rekado>this tells me that it was called as (cross-gcc … the-binutils-thing …), not (cross-gcc … #:xbinutils the-binutils-thing …)
<rekado>ng0__: building curl on core-updates takes very long, because I made a change on core-updates that results in a complete rebuild…
<rekado>ng0__: but I’m hopeful that we can close that bug soon.
<ng0__>woo :)
<ng0__>we only have armv8 support now, so cortex-a8 will not work as it is v7, right?
<ng0__>I'm trapped in this chart and omg its full of arms
<ng0__>yeah, definitely v7.. so nogo
<mbakke>rekado: I suspect the cross-stuff in build-aux/hydra/gnu-system.scm may be related
<mbakke>mips64el-linux-gnu is the first entry in %cross-targets