IRC channel logs

2020-05-11.log

back to list of logs

<jonsger>dftxbs3e: what are you working on? gcc-cross-boot0-wrapped@7.5 seems interesting :P
<dftxbs3e>jonsger, I'm trying to merge upstream master to my ppc branch
<jonsger>dftxbs3e: nice
<lispmacs>I need to change my gnome keyring password, but I cannot find the "Passwords and Keys" interfaced as described in the online documentation
<lispmacs> https://help.gnome.org/users/seahorse/stable/keyring-create.html.en
<dftxbs3e>rekado, I think I solved that issue helped by the diff - thanks
<dftxbs3e>It was a superfluous (var) instead of just var
<dftxbs3e>lispmacs, as a workaround maybe there's a cli for that?
<dftxbs3e>try guix search keyring?
<lispmacs>dftxbs3e: that does bring up the gnome-keyring daemon, but that is already in my path
<lle-bout>lispmacs: then just try: guix install seahorse (name of that utility)
<lle-bout>I can see it here
<lle-bout>btw I'm dftxbs3e just on a different machine
<jonsger>:P
<lle-bout>I got no bouncer so..
<jonsger>lle-bout: run the bouncer on your BMC :P
<lle-bout>jonsger: ehh.. I don't trust that piece really much for online software, it's got outdated kernel and libraries the entire time
<lle-bout>or you mean for local
<lle-bout>I'm fine tolerating the current state of affairs
<jonsger>local is maybe a better idea. I'm thinking of running local printing stuff on it, as its always on
<lle-bout>+ the added operational privacy it provides
<lle-bout>the Talos II BMC is nice but I wish it was equiped with better hardening / software minization and/or more regular updates
<lle-bout>minimization *
<lle-bout>+ there's no persistent space so
<lle-bout>you'd have to mod it from source etc. but that's offtopic so.. šŸ˜…
<jonsger>lle-bout: how much storage does it have? for guix pack stuff
<lle-bout>jonsger: hmm oh yeah.. well few hundreds MBs I'd say for firmware upgrades
<lle-bout>there's persistent space for dropping firmware updates AFAIK but it's in a location where you can't autoboot anything from it
<wdkrnls>Is there a simple way to find all packages provided by multiple channels?
<lfam>wdkrnls: `guix package -A` should list all available packages
<lfam>AKA `guix package --list-available`
<drakonis> hmm, there's no package namespacing in guix, right?
<drakonis>its always a single level, yeah?
***catonano_ is now known as catonano
<wdkrnls>lfam: That's just what I needed, thanks!
***jonsger1 is now known as jonsger
<jonsger>dftxbs3e, lle-bout: I'm trying to build new bootstrap-tarballs from https://gitlab.com/jonsger/Guix/-/tree/wip-master-powerpc64le. Its failing with https://paste.opensuse.org/view/raw/9860724
<lle-bout>jonsger: never got that
<lle-bout>I'm not trying to build new tarballs
<jonsger>its maybe because of glibc-3.21
<jonsger>glibc@2.31 I mean
<lle-bout>jonsger: I don't advice you try to build new tarballs, those are built at a certain commit and are sufficient
<lle-bout>no need to ensure working state of bootstrap chain onwards (it's hard to maintain)
<jonsger>I know
<lle-bout>it's an impossible QA matrix
<wdkrnls>Hmm... after updating my locale it not being set automagically in my profile despite glibc-locales being installed in my profile. After running "export GUIX_LOCPATH=$HOME/.guix-profile/lib/locale" everything works as expected. However, I don't want to have to run this every time. I'm sure this question has been asked before, I just can't find a better answer than running that command all the time.
<lfam>wdkrnls: You'll have to log out and in again
<lfam>Those variables are set on login
<lfam>You can test this with `bash --login` or equivalent for your shell
<wdkrnls>Hmm... doesn't help.
<lfam>Sigh
<Guixguy>Hey, if anyone has experience with doing a git-fetch, I need some help finishing up a package. Not sure what I'm doing wrong: https://paste.debian.net/1146079/
<lfam>Did you update the locales for *all* users?
<Guixguy>wdkrnls: What's going on?
<lfam>Guixguy: It would help if you told us what goes wrong
<lfam>Please help us help you
<wdkrnls>no, just for me. Do I need to update the system?
<lfam>wdkrnls: Yes
<lfam>If you want that warning to go away
<Guixguy>Here is the error messages: https://paste.debian.net/1146080/
<wdkrnls>Guixguy: I have probably the most common guix issue: locale not being updated without also updating the system as well as my user.
<wdkrnls>I'll give that a try.
<lfam>Guixguy: You need to at least provide a value for the "commit" field
<Guixguy>Ah, yes. a system reconfigure will do the trick for tha tone
<lfam>Guixguy: Try following the instructions here: https://guix.gnu.org/manual/en/html_node/Version-Numbers.html
<Guixguy>lfam: Thank you. I'm curious if you have "Please help us help you" hotkeyed as I see that often when reading irc logs.
<Guixguy>Ashamed to be on the end of one =P
<lfam>No, I just typed it out. The more info, the easier it is to help!
<Guixguy>In examples, I see the url of the git-reference points to the top layer. How do I specify a single package out of it?
<pkill9_>damn, i forgot another nice benefit of guix: able to update software while it's running without breaking it
<Guixguy>:-D
***pkill9_ is now known as pkill9
<lle-bout>pkill9: it's not really updated :P
<lle-bout>running process is still old
<lle-bout>samplet: oh hi samplet !!
<pkill9>yea lle-bout, i mean it's dependencies
<pkill9>and itself
<pkill9>i mean, the installe dversion
<malaclyps>is anyone else having problems using the emacs-guix package at the moment? I get a "service-sexps: unbound variable" error whenever I try to use it at themoment, even with an empty .emacs config.
<lle-bout>samplet: don't know if you heard that ppc64 is now bootstrapped
<pkill9>on a normal distro, updating it's dependencies while it's running swaps out file sunderneath it
<samplet>lle-bout: Wow! Thatā€™s pretty cool.
<lle-bout>pkill9: yes but it's still loaded so it should be OK but it could conflict yes
<Guixguy>malaclyps: I am not having any issues. I have it installed in my config.scm
<Guixguy>malaclyps: Sorry, ignore me. I confused emacs on guix with emacs-guix.
<malaclyps>Guixguy, dont' want to mess with your emacs, but if you install emacs-guix and try M-x guix, let me know if it works :)
<Guixguy>Sure, one sec
<lle-bout>samplet: it's been months but I'm looking at it again now
<malaclyps>i can't imagine it's really broken for everyone because I get the feeling ludovic and others use it a lot
<lle-bout>malaclyps: currently it's partly broken
<lle-bout>can't look up services for example
<Guixguy>malaclyps: I ran it and it said it loaded something at the bottom
<Guixguy>malaclyps: nothing opened in emacs
<malaclyps>lle-bout, huh, okay thank you!
<malaclyps>Guixguy, thank you to you too
<malaclyps>I shall report it as a bug!
<pkill9>how do you set up the cgit service to use ssl?
<lle-bout>malaclyps: I met that issue earlier too, as a workaround I used guix system search -- it searches services
<malaclyps>okay, recorded here: https://gitlab.com/emacs-guix/emacs-guix/-/issues/23 (I always thought emacs-guix was part of guix until now).
<ecbrown>emacs-guix has been failing for me on installing packages, complaints about #:use-substitutes
<raghavgururajan>Hello Guix!
<KE0VVT>hi
<rekado_>weā€™re again down to 2.8T free space on ci.guix.gnu.org
<rekado_>I wonā€™t intervene this time
<raghavgururajan>rekado_: What happend to guix-ci bot?
<rekado_>killed it
<apteryx>rekado_: hmm... would it be a good idea to lower our retention time from the default (1 month) to something lower, like 3 weeks?
*apteryx zzzz
<ryanprior>rip ci bot
<ryanprior>poor thing had anxiety
<xelxebar>rekado_: Regarding texlive package names, is the convention to mirror upstream's hierarchy and name them like texlive-<subdir>-<pkgname>?
<xelxebar>Most packages seem to follow this, but e.g. texlive-latex-babel is under generic/babel (though we have texlive-generic-babel-english etc).
<xelxebar>Though, things like texlive-ydoc seem to invalidate this hypothesis, since it's under latex/ydoc...
<lispmacs>hi, I have this oddity where gnome seems to have lost the ability to print unicode Hebrew characters
<lispmacs>but Emacs can display them fine
<lispmacs>is that a bug to report, or some config thing I need to adjust?
<xelxebar>lispmacs: By "print unicode Hebrew characters" do you mean that gnome GUI utilities are not properly displaying them?
<janneke>sneek: later tell civodul, oh bah, thanks!...small steps indeed :-)
<sneek>Got it.
<lispmacs>xelxebar: yes, gnome applications such as evolution and leafpad just show the empty square
<janneke>sneek: botsnack
<sneek>:)
<lispmacs>xelxebar: but if I insert the same unicode code into Emacs with Ctrl-x 8 ENTER, etc, the character displays fine
<xelxebar>lispmacs: What's your system locale?
<lispmacs>(operating-system
<lispmacs> (locale "en_US.utf8")
<lispmacs> (timezone "America/Anchorage")
<lispmacs> (keyboard-layout
<lispmacs> (keyboard-layout "us" "altgr-intl"))
<xelxebar>Okay, and just to make sure LANG set to the same?
<lispmacs>christopher@nightshade ~$ echo $LANG
<lispmacs>en_US.utf8
<xelxebar>How did you install your Hebrew fonts, btw?
<lispmacs>xelxebar: I don't recall ever installing any special Hebrew fonts, as it wasn't necessary in the past
<lispmacs>xelxebar: do I need to add something to the locale?
<xelxebar>lispmacs: I think your locale looks fine. Unfortunately, I don't know much about gnome, but my first hypothesis would be that there is some difference between your system profile (from which gnome typically runs) is different than your user profile (which ends up running your terminal and thus emacs).
<lispmacs>hmm... maybe default gnome font does not have Hebrew characters finished...?
<lispmacs>I think that is it. I see default font (at least on my system is Cantarell
<lispmacs>but character map program shows that font is missing the Hebrew letters
<xelxebar>Hrm, I wouldn't guess that's the problem, since AFAIK characters can be pulled from appropriate fonts as needed, but it's certainly worth a shot.
<xelxebar>Try something in the DejaVu family. That tends to contain a really good swath of unicode.
<KE0VVT>lispmacs: What issues are we having with Hebrew letters? I used to type Hebrew in Emacs, until I had to handle vowel markings. Emacs doesn't display them well.
<lispmacs>KE0VVT: emacs displays them fine, but not gnome applications, at least on my system
<KE0VVT>lispmacs: Interesting. Can I see how your Emacs displays the following text? מײַן נאÖøמען איז כּÖøלֵבÖæ
<lispmacs>xelxebar: I switch gnome to libertine default fonts, and am abble to insert the Hebrew characters now
<lispmacs>KE0VVT: it looks great, am trying to think how to send you a screenshot
<lispmacs>KE0VVT: can PM me your email...?
<KE0VVT>lispmacs: csh@bluehome.net
<KE0VVT>lispmacs: Though you could just upload to Framapic and put the link here.
<lispmacs>KE0VVT: just sent email. Haven't heard of framapic before, but have bookmarked it now
<malaclyps>Hmm, I am putting together some patches for new guix package updates. I'm doing it one package at a time, and want to test/lint the package definitions as changes to the guix gnu/package files. This means that guix has to rebuild itself, which takes a *long* time. I remember some reference in the manual to speeding this up -- does anyone have a suggestion?
<xelxebar>lispmacs: Hrm. Glad it's working, but that sounds like a broken fontconfig issue.
<lispmacs>xelxebar: hmm. do I need to rerun some fontconfig database command?
<xelxebar>malaclyps: How are you building? running make should just build files with newer timestamps than the last build, not the entire distribution.
*lispmacs going to bed
<xelxebar>lispmacs: Unfortunately, you've reached the limits of my (shallow) knowledge on how font rendering works...
<mroh>gn8 lispmacs
<KE0VVT>lispmacs: Framapic is a Lutim instance. If you have the resources, I'd suggest running your own. That instance will be going down at some point. For files, there is Lufi.
<KE0VVT>Both have excellent interfaces and are free/libre.
<KE0VVT>Going downstairs to write Guix installer and boot.
<calher>Writing ISO.
<malaclyps>xelxebar, I've got a checked copy of the guix source, which I have setup as a channel in ~/.guix/config/channels.scm. I edit the files in a branch there, and then do a guix pull.
<calher>It's nice to have a full monitor, keyboard and mouse, with a telephone to my right and a telegraph key to my left.
<calher>ISO written. Booting into Guix installer.
<calher>Ta-ta!
<xelxebar>malaclyps: Well, my guess is that substitutes are just not available. FWIW, I just do all my building directly in the guix environment guix.
<xelxebar>just do ./pre-inst-env guix <command> and it should work mostly as expected.
<malaclyps>xelxebar, this is just rebuilding guix because i'm changing a few package definitions -- it's not building subsitutes other than those.
<malaclyps>but maybe pre-inst-env is what I'm looking for, I've never used that.
<xelxebar>malaclyps: Did you check out the tutorial videos for package development? https://guix.gnu.org/videos/
<xelxebar>They give a decent overview of the standard development process. Basically, you just do everything from within guix environment --pure guix.
<xelxebar>My personal dev loop looks like this: edit gnu/packages/foo.scm and then make && ./pre-inst-env guix build bar
<xelxebar>Then once everything looks good, I just do a ./pre-inst-env guix package -i bar to install and do a final sanity check
<KE0VVT>In installer right now.
<KE0VVT>Guix has MATE now?!
<KE0VVT>Or should I still stick with GNOME?
<KE0VVT>I like GNOME, but I have a Core 2 Duo with 7 GB RAM.
<KE0VVT>Uh-oh. I'm having issues in the interactive installer in Partitioning. It sees no disks.
<KE0VVT>But /dev/sda exists.
<malaclyps>xelxebar, no -- thank you I'll take a look at that!
<xelxebar>np
<KE0VVT>:-(
<KE0VVT>Ugh... Guess I need to install the old way. But it's hard to get disk encryption set up that way...
<KE0VVT>Crap. desktop.scm does enc. by default. I thought NOT encrypting would make things easier.
<KE0VVT>This still does not explain why the guided installer does not see /dev/sda.
<KE0VVT>I guess I'll try doing this next weekend. :-(
<raghavgururajan>KE0VVT: You around?
<efraim>first stab at enlightenment-0.24 update done, always intensive when its substitutes and patches and services all in one go
<civodul>Hello Guix!
<sneek>Welcome back civodul, you have 1 message!
<sneek>civodul, janneke says: oh bah, thanks!...small steps indeed :-)
<mroh>good morning civodul!
<nixo_>Hello guix! Did deluge broke during last core-update merge or is my config broken? deluged cannot find libtorrent
<srandom>Has anyone used 'xorg-start-command'? how should it be used?
<srandom>I want to start my cwm using startx in tty
<xelxebar>This is so wierd. Putting together 6 different texlive packages, 4 of the hashes start with "0". To boot, no other packages in tex.scm have hashes beginning with "0".
<civodul>heh
<civodul>srandom: "startx" is not available on Guix System
<civodul>but i think you can configure gdm (or slim, etc.) to use a different tty
<lprndn>Hello guix!
<civodul>hey lprndn!
<janneke>hello civodul :-)
<lprndn>nixo_: Tried deluge and it can't find libtorrent either. So the package is probably broken...
<mroh>srandom: may this https://github.com/alezost/xdaemon helps
<xelxebar>civodul: The xinit package provides startx; does this not work for some reason?
<srandom>If Guix provides xdm, I might think about it, but gdm, sddm, slim, I really do nā€™t want to use them.
<pinoaffe>srandom: currently, guix supports gdm, sddm and slim
<civodul>xelxebar: i won't work as is because it needs to be told about the location of drivers, etc.
<pinoaffe>so you're out of luck
<nixo_>lprndn: thanks
<xelxebar>civodul: I see. Might decide to look into that.
<mbakke>nixo_: it could be because libtorrent was changed to use Python 3, but Deluge is built with Python 2
<nixo_>mbakke: According to https://dev.deluge-torrent.org/wiki/Plan/Python3, deluge is still not compatible with python3, so should we create a libtorrent-rasterbar-2?
<rekado_>xelxebar: no, the texlive names should just be texlive-<name>
<rekado_>xelxebar: those packages that include the sub-directory name are old-style packages that may actually be missing files.
<xelxebar>Thanks
<mbakke>nixo_: Ubuntu has a Python 3 version of Deluge: https://packages.ubuntu.com/focal/deluged
<mbakke>according to the Changelog, version 2.0.3 gained Python 3 support
<rekado_>srandom: lightdm is coming soon
<rekado_>see y
<rekado_>see https://issues.guix.gnu.org/issue/35305
<srandom>Will lightdm support starting .xsession or .xinitrc in the user directory?
<nixo_>mbakke: ok I'll try to upgrade it
<srandom>My prejudice against these dm is because they are too inflexible.
<mbakke>nixo_: great :)
<rekado_>srandom: lightdm-configuration has a field sessions-directories, which takes a list of directories where to look for sessionsā€™ .desktop files.
<rekado_>so you could put that in your home directory
<srandom>Great
<rekado_>but note that this is for .desktop files, not .xsession or .xinitrc directly.
<srandom>Well, the desktop file can be redirected to xinitrc. I can write xinitrc as a command.
<srandom>xinitrc is a shell script
<rekado_>yes, I just mean that itā€™ll need some manual plumbing
<bricewge>srandom, rekado_ : Actually you can directly specify a xinitrc by using the field lightdm-seat-configuration-session-wrapper
<rekado_>oh, good to know
<TZander>what do people think about the idea that guix knows its installed on a host system with the result that scripts and profiles change variables like PATH to append guix specific things instead of prepend.
<TZander>(not sure why it would even ever need prepend, but I guess thats done for a reason)
<cbaines>TZander, I guess prepending is done to avoid unreproducible behaviour with some search paths. I can imagine some search paths are searched from the start, so you want to prepend to ensure locations from the environment at the end are considered last.
<cbaines>Do you have an example of when you've had an issue with search path ordering?
<TZander>yeah, my kde no longer starts
<TZander>this is the native (archlinux) kde with in guix some kde based apps.
<TZander>I suspect the biggest issue is the very old libraries from guix being found before the native up-to-date ones.
<rekado_>append will cause failures
<rekado_>prepend has much fewer issues
<rekado_>but to find a better workaround we need to know the variable that causes problems
<TZander>ok, I'm out of luck then...
<rekado_>generally, I think Guix should prefix more environment variables
<cbaines>one strategy is to avoid the environment variables mixing in the first place
<cbaines>TZander, do you know what specifically from Guix you have installed that's causing issues, or what search paths the issues are with?
<TZander>I'm ok with it doing that on a guix system, it just doesn't work when you prepend guix stuff which is the opposite of 'always the latest' on a host system.
<TZander>cbaines: several things
<rekado_>TZander: can you please be specific? What libraries are ā€œvery oldā€ and the ā€œopposite of always the latestā€?
<rekado_>please report them to bug-guix@gnu.org if you can
<TZander>guix is not about having the latest version. It makes it easy to use several different versions of boost, for instance. This is what I mean where the design, the conceptual basis of the system is the opposite of most distos.
<TZander>So this is not about a bug specifically; even though Qt is old and KDE is old on guix.
<cbaines>Guix generally does aim as far as I'm aware to just have the latest versions of things
<cbaines>barring some exceptions, where the aim is to have the latest, plus some older versions (PostgreSQL, Guile, Ruby, ...)
<cbaines>of course, the degree to which that is true is mostly limited by the rate at which packages are updated
<TZander>No, you mistake "provides" with "having"
<YannD>Hi everybody
<TZander>If you go to a debian or archlinux you will find is very hard to download a binary package of any lib from 2 versions ago.
<TZander>It is completely supported on arch to roll back to a state as your machine was a year ago.
<TZander>Which is great, but also utterly incompatible with software from the host system.
<nixo_>mbakke: deluge-related good news and bad news: builds&run, but it depends on python-setproctitle which is currently broken (I had to work around the problem by breaking support for python 2 on it). Also, in order to pass some tests, it requires a newer python-twisted. The output for guix refresh -l -r python-twisted isn't promising (5793 packages would ensure 15616)
<cbaines>TZander, but Arch is the host system in this example. Did you mean to talk about roll backs with Guix or Arch?
<TZander>cbaines: ah, typo. Yes, arch => host system. Guix allows rollbacks of its profiles.
<TZander>not of archlinux software, indeed
<cbaines>I've not used KDE in a long while, but I've had similar issues with Gnome. When mixing Guix and Debian, sometimes I couldn't login. I think it was possible to address in some cases though by changing package definitions not to propagate certain packages that could interfere with software from Debian.
<cbaines>As a place to start investigating this though, it probably needs narrowing down to the environment variable, or Guix package that is leading to the issue.
<mbakke>nixo_: don't use -r with 'guix refresh -l'
<TZander>cbaines: after trying to find out why (and it just always failing) I ended up just moving the guix.sh out of /etc/profile.d and removing the lines in by bash_profile and now having a 'startGuix' function to start doing guix stuff
<nixo_>mbakke: then it's only 33-74, I'll try this afternoon
<bricewge>nixo_: I have a working deluge 2.0 but the patchset isn't finished, do you want it?
<cbaines>TZander, yeah, I realise not being able to login can be difficult to debug. At least I've had problems in the past with the GDM.
<nixo_>bricewge: mine is working, too, does your pass all tests?
<bricewge>Unfortunately not, few days ago I had ~90 failing...
<nixo_>i think some of them must be disabled as they require an internet connection, but I can't understand how many as the output is really terrible
<bricewge>nixo_: To build python-setproctitle on python 2 and 3, just remove the modified check phase
<nixo_>bricewge: yes, that's what I did, but I think this will break python2 build
<nixo_>btw, I really like the new deluge
<bricewge>I tried building it and it succeded
<bricewge>Yes the UI is nicer, I don't like not being able to scroll trough the 'Status, Details, ...' in the bottom part of the window
<bricewge>Best part is the sequential download option though
<bricewge>nixo_: I struggled so much with tests on this package, that I'm of the opinion they should be disabled. I can't find any other distro which are running them.
<nixo_>bricewge: we can do better than other distros ;) I'll try to run as many tests as I can
<bricewge>nixo_: You have a brave soul.
<nixo_>and some spare time
<nixo_>see you later people :)
<nikita`>i haven't send a patch in so long that I had to look up the email address and read into old logs how to format patch entries :O
<nikita`>hope to get back to sending at least a couple of patches per year from my backlog soon..
<xelxebar>\me fist bumps nikita`
*xelxebar fails
<xelxebar>Too much TeX these days...
*jonsger wonders how often one has to rebase and update a patch series until it gets merged ^^
<nikita`>hm. pkgsrc hat on: how do you handle ctan apparently not handling permanent files and replacing them in place? this seems to be an issue in pkgsrc right now and we have to track down packages and push sources to our ftp
<xelxebar>nikita`: Isn't that just an issue on Master? I'd expect the files in the tag/* hierarchies to be static.
<nikita`>probably?
<nikita`>you mean ctan now keeps tags?
<xelxebar> https://tug.org/svn/texlive/tags/
<nikita`>but what about individual ctan packages? that's the issue right now
<nikita`>oh nvm
<nikita`>i misread the issue. hm
<xelxebar>Yeah, texlive-build-system pulls sources from (string-append "svn://tug.org/svn/texlive/tags/" %texlive-tag "/Master/texmf-dist"), I'd believe we're reproducible.
<jonsger>debbugs is confusing: XXX-done closes a bug and set it to done, XXX-close marks it only as done :P
<xelxebar>So, closing depends on done-marking, but the XXX-* naming is backwards. Fun.
<nikita`>at least it's not gnats, which made me figure out how to deal with decoding attachments
<xelxebar>mbakke: Noticed that you are the one that bumped texlive to 2019.3 in January. 2020.0 has been out for about a month now, with 3367 commits over 2019.3. Would a bump be really disruptive?
<jonsger>does someone know why backspace is not working in vim on Guix System?
<mbakke>xelxebar: a TeX Live 2020 upgrade would be very welcome and can be done on a dedicated branch
<mbakke>I'm not volunteering though ;-)
<apteryx>jonsger: close is deprecated
<apteryx>use done
<jonsger>apteryx: ah but then you don't get this nice `close` icon at the end of the bug
<apteryx>in the Emacs client? I don't recall seing a close icon; I wouldn't notice perhaps, as I only lists opened items most of the times.
<apteryx>list open*
<rekado_>jonsger: in issues.guix.gnu.org we could also add the closed pill when ā€œ-doneā€ is used. Currently Iā€™m only matching on a Debbugs internal message that apparently gets added to the log when close is used.
<rekado_>who would like to adopt a forgotten bug/patch today?
<rekado_> https://issues.guix.gnu.org/forgotten
<civodul>excellent :-)
<PotentialUser-59>Can someone help me with the setup for flatpak?
<PotentialUser-59>when I install it with "guix package --install flatpak" and then try to add flathub via "flatpak remote-add --if-not-exists flathub https://flathub.org/repo/flathub.flatpakrep"
<PotentialUser-59>I receive errors
<PotentialUser-59>has anyone set it up?
<civodul>rekado_: how about using 'time->string' for timestamps? https://git.savannah.gnu.org/cgit/guix/guix-cuirass.git/tree/src/cuirass/templates.scm#n390
<civodul>PotentialUser-59: looks similar to https://issues.guix.gnu.org/35591 no?
<civodul>"The segfaulting behaviour is still a bug in my opinion" :-)
<mbakke>civodul: do you have a trick for mapping a PID to a build job? On Berlin, 63882 is using a lot of CPU time, occasionally uninterruptible, and I suspect that is what's holding the sqlite DB lock for more than an hour: https://ci.guix.gnu.org/jobset/guix-master
<mbakke>the process name is .guix-offload, so could be anything (presumably something huge)
<TZander>mbakke: cat /proc/63882/cmdline not helpful?
<cbaines>mbakke, do you see the PID in guix processes ?
***drakonis1 is now known as drakonis
<jfred[m]>I've also gotten Flatpak to work system-wide by running it as root
<jfred[m]>but yeah, it segfaulting is not super user-friendly heh
<jfred[m]>(probably using `--user` is the "right" way on guix, I didn't know about that option at the time)
<civodul>mbakke: dunno but i would do: sudo guix processes |recsel -e 'ClientPID ~ "63882"'
<civodul>and maybe strace the thing to get an idea of what it's doing
<rekado_>jfred[m]: is it segfaulting because of something we did to it or is it just an upstream bug? If itā€™s the latter, has the bug been reported upstream?
<civodul>jfred[m], rekado_: looks like https://issues.guix.gnu.org/35591
<rekado_>yes, I have that open
<bricewge>Updating network-manager to get vala binding from it, I'm wondering if I should separate libnm for it (either as an output or a proper package).
<rekado_>but it does not seem to include a discussion of the segfault
<bricewge>Distro like Debian or Arch already does it.
<rekado_>we arenā€™t patchig it.
*rekado_ tries upgrading it
<rekado_>weā€™re down to 2.4T, Iā€™m running guix gc -D < /gnu/big-stuff
<civodul>bah, ok
<civodul>i realize we forgot to update the GC job
<civodul>rekado_: what should be our daily threshold? 2.4T?
<civodul>i think it should be, say, 5 times our daily consumption
<jfred[m]>rekado_: IIRC the last thing I figured out wrt flatpak was that it was trying to access (write I think) files in /var/lib/flatpak, was unable to due to file permissions, then segfaulted. I just tested the same `flatpak remote-add` as an unprivileged user on a Debian 10 system and seem to be getting a prompt to escalate privileges
*jfred[m] uploaded an image: Screen Shot 2020-05-11 at 10.10.51 AM.png (452KB) < https://matrix.org/_matrix/media/r0/download/terracrypt.net/xEoFPXLqVbQHBXAGLCBWDkQR >
<jfred[m]>not entirely sure what facility they're using to do that
<jfred[m]>(should've updated the issue report once I figured that out in retrospect but it probably slipped my mind since I'd gotten it working)
<jfred[m]>ahh, it's policykit
<buffet>hey guix, i have a package (using meson) looking for "lua", but when i add it to the inputs field pkg-config only knows it as lua-5.3, is there a way i can "rename" it?
<cbaines>buffet, maybe patch the occurance of "lua" in the package
<buffet>eh i hoped that was avoidable
<buffet>cbaines, thanks
<cbaines>I don't know much about lua or pkg-config, so there might be a better way
<cbaines>I don't know if lua-5.3 is right in the Guix lua package, or if thats wrong
<raghav-gururajan>buffet: Is it pkg-config or configure that gives that error during build?
<buffet>raghav-gururajan, `meson.build:20:0: ERROR: Dependency "lua" not found, tried pkgconfig`
<raghav-gururajan>buffet: Ah okay, you can manually pass a config flag that overrides pkg-config.
<raghav-gururajan>buffet: Ah okay, you can manually pass configure-flag to override pkg-config.
<raghav-gururajan>buffet: So if you look at your package's configure script, there might be an option like --with-foobar-incl and/or --with-foobar-lib.
<buffet>raghav-gururajan, will check it, thanks
<pkill9>I thought configuring services was confusing, then i realised that the "config =>" part is only for inheriting from the pre-existing config when using modify-services
<raghav-gururajan>buffet: Oh shoot, you are using meson. I was telling you for gnu.
<raghav-gururajan>buffet: Anyway, there might be similar option (meson_option) in meson.build file, which you can pass as -Dfoobar.
<buffet>actually got another question. in the inputs list, i have a whole bunch of ("name" ,name). what is the string used for, if its not what the package gets made available as?
<cbaines>It's what the package get's made available as, but only in the Guile code for the build environment
<jetomit>is it OK to use wrap-program to set env vars that are not lists of paths? specifically something like (wrap-program "foo" '("MESA_GL_VERSION_OVERRIDE" = ("3.0")))
<buffet>ahh i see
<lprndn>Is there someone versed in dbus in guix? I'm searching for what we do with things in `share/dbus-1/interfaces...
<jonsger>$ git add --patch gnu/packages/messaging.scm
<jonsger>git: 'add--interactive' is not a git command. See 'git --help'.
<jonsger>what am I doing wrong?
<buffet>yet another question: is there a way to override attributes of a package, without copying the entire definition?
<buffet>something like nix's `(enableDebugging package)` would be nice
<buffet>as the debug output isnt enabled everywhere
<civodul>lprndn: the dbus-root-service-type thing takes care of making them visible to dbus-daemon
<civodul>so services that provide dbus interfaces must "extend" dbus-root-service-type
<pkill9>does anyone have an example config of an nginx server providing ssl?
<pkill9>i've set the ssl certificate and private key, and set listen ports to "80" and "443", but it says connection refused
<pkill9>oh interesting, the nginx config file doesn't have 443
<pkill9>the generated one
<pkill9>damnit, I was reconfiguring the wrong configuration file
<pkill9>hmm, it's still refusing to connect
<pkill9>ohhh, i need to manually restart the nginx service
<pkill9>that was all
***amiloradovsky1 is now known as amiloradovsky
<janneke>where do i access the running linux config.gz?
<janneke>is /proc/config.gz a debian thing?
<jetomit>janneke: no, but it has to be enabled in kernel config
<jetomit>CONFIG_IKCONFIG_PROC is the option
<lprndn>civodul: thanks! I think it's not what I'm looking for then...
<janneke>jetomit: thank you -- wouldn't it be nice to enable that?
<janneke>provenance, and all ;)
<jetomit>I agree ā€¦ I enable it in my configs, didnā€™t even notice itā€™s off by default :)
*civodul posts a neat 'guix pack -RR' improvement: https://issues.guix.gnu.org/41189
<lprndn>accountsservice looks in XDG_DATA_DIRS for dbus interfaces. It's used at least by lightdm. Currently, it doesn't find anything as this variable is not set. Should I just wrap accountsservice or are there better ways?
<PotentialUser-79>hello
<raghavgururajan>PotentialUser-79, o/
<PotentialUser-79>i have a problem with installer freezing at Wifi
<PotentialUser-79>i can connect to the network and ping before I start installation
<raghavgururajan>PotentialUser-79, Are you doing manual installation?
<PotentialUser-79>graphical
<PotentialUser-79>it gets stuck at "Connection in progress"
<PotentialUser-79>I have supported wifi dongle
<raghavgururajan>PotentialUser-79, You could intiate manual installation, just to get to the terminal and check with some nettools.
<bricewge>PotentialUser-79: It's a known issue, using a smartphone connected to the wifi and tethering to Guix via USB side step the issue.
<PotentialUser-79>I never done it before. But I will try
<PotentialUser-79>is there any link that describes the setup in more detail?
<PotentialUser-79>is there a way to hack the installer from the installation shell?
<PotentialUser-79>even If I do not have the wifi?
<lfam>PotentialUser-79: You can access the Linux console on another TTY with something like CTRL+ALT+F3
<lfam>Then, check dmesg for the device name of the network interface created when the phone is plugged in
<lfam>Then, set it up like a regular ethernet interface
<lfam>Something like `ip link set $INTERFACE up && dhclient $INTERFACE`
<lfam>Hopefully those commands are available in the installer image
<PotentialUser-79>can i do the same with existing wifi connection? I can connect to wifi even before I start installation
<PotentialUser-79>is there way to skip wifi connection in such case?
<lfam>You'll need to use wpa-supplicant for wifi and it's not quite as easy
<lfam>I probably can't explain it to you
<PotentialUser-79>i did it and it works
<lfam>Great!
<PotentialUser-79>i need to skip the step asking me for wifi connection, because I have the wifi already
<lfam>Sorry, I don't know
<PotentialUser-79>i will try manual install
<Gooberpatrol66>What's the best way to get useful backtraces in guix?
<Gooberpatrol66>for example, if you do guix system disk-image --target=INVALIDTARGET /etc/config.scm, it prints guix/ui.scm:1936:12: In procedure run-guix-command:, even though the actual error is printed in the function glibc-dynamic-linker, in a different file
<bricewge>Gooberpatrol66: +1
<civodul>Gooberpatrol66: it's a regression since the switch to Guile 3 i think, though it's probably not Guile that's to blame
<civodul>but yeah, annoying!
***jonsger1 is now known as jonsger
<civodul>i looked at it a bit and didn't spot the problem
<civodul>we must be re-throwing from a 'guard' handler, which runs after the stack has been unwound
<Gooberpatrol66>what can i read to be made aware of when it gets fixed in the future?
<pkill9>does cgit add to the configuration of nginx? or spawn a whole new nginx/
<pkill9>?*
<pkill9>e.g. does it extend the nginx service or create a new instance of it
<pkill9>i hope it's the former
<bricewge>pkill9: It extend it if I recall correctly
<pkill9>cool
<bricewge>It's the case for most of the service
<PotentialUser-79>how do you solve mismatched parentheses when you get to editing /mnt/etc/config.scm
<lfam>PotentialUser-79: You have to match them up correctly
<lfam>If you can share your file I can fix it for you
<PotentialUser-79>yes, but how on earth you can do it without paredit
<PotentialUser-79>?
<lfam>Once you learn the language, you can do it by hand somewhat easily
<PotentialUser-79>how do I share that file from such limited system?
<lfam>You could take a photo and share that
<TZander>PotentialUser-79: colors in vim are useful too...
<PotentialUser-79>no chance, the file is longer than one page. can i ssh to the installer?
<lfam>Two photos?
<lfam>Whatever you can manage
<PotentialUser-79>i would prefere ssh and do it my self
<lfam>Okay
<lfam>The parentheses are part of the language, like semicolons or equality signs in C
<PotentialUser-79>i know Lisp
<lfam>Oh okay
<lfam>Basically the same thing, then. Just count them out so that each field of the operating system starts and ends correctly
<lfam>There shouldn't be too many layers of parens for config.scm
<TZander>any, just in case you use vim, use `set showmatch`.
<TZander>but, yeah, you need a proper text editor.
<civodul>PotentialUser-79: the installer has ssh, see https://guix.gnu.org/manual/en/html_node/Keyboard-Layout-and-Networking-and-Partitioning.html#Keyboard-Layout-and-Networking-and-Partitioning
<rekado_>guix gc -D < /gnu/big-stuff has finished ā€¦ 2.1T free space.
<rekado_>not enough
<rekado_>ā€œguix pull --newsā€ has a formatting problem
<rekado_>ā€œ[ā€¦] weighs in at approximately 60??MiB, a quarter of what it used to be[ā€¦]ā€
<rekado_>maybe thatā€™s because of missing new localesā€¦?
<jonsger>hm. Is there a way to make a guix graph smaller? atril segfaults with the resulting pdf and can not zoom enough into the postscript file :(
<rekado_>jonsger: with huge graphs I often just look at the dot file directly
<rekado_>I would love to have a PDF viewer that can zoom deeper
<rekado_>evince also becomes very sluggish when zooming in, and thereā€™s a limit
<cbaines>I think I've found using the SVG ok for large graphs
<rekado_>you can also try the d3 output
<jonsger>cbaines: how can it output an svg?
<cbaines>Maybe you can render dot to SVG... I forget
<jonsger>convert: delegate failed `'potrace' --svg --output '%o' '%i'' @ error/delegate.c/InvokeDelegate/1958. lol. pdf -> svg is not working
<civodul>rekado_: did you try xdot?
<lfam>Yes, use SVG
<civodul>it's really cool!
<lfam>`guix graph foo | dot -Tsvg > out.svg`
<civodul>guix graph foo | xdot -
<lfam>It will still take a while but it won't crash your computer
<civodul>rekado_: the question mark in "guix pull --news" must be a no-break space
<lfam>Thanks for xdot, that's really nice
<bricewge>Does the d3 actually work, I never managed to use it successfully. xdot is nice indeed.
<bricewge>s/d3/d3 output/
<Blackbeard>What's the command to check if packages are outdated ?
<pkill9>guix refresh i think
<rekado_>bricewge: why yes!
<rekado_>itā€™s a different visualization, though
<bricewge>rekado_: I endup with a html file (http://ix.io/2lG6) how can a view the graph
<rekado_>open it in a browser
<bricewge>Icecat returns an empty page.
<rekado_>ah, thereā€™s a #f where a script should be
<rekado_>did you use pre-inst-env?
<reepca`>seconded on xdot being really nice, helped navigate control flow graphs a lot
<PotentialUser-79>no luck with ssh connetio to maual installer, will try the phone method
<pkill9>does the certbot service allow you to change details provided by the certificate, e.g. organisation
<pkill9>?*
<bricewge>rekado_: No just ā€œguix graph -b d3js shepherd > /tmp/shepherd.htmlā€
<rekado_>yeah, that wonā€™t work :-/
<rekado_>gotta use pre-inst-env because we donā€™t actually install d3js anywhere
<rekado_>this can be fixed
<bricewge>rekado_: Which package contains d3js?
<bricewge>That would be great, or at least written in the manual
<lfam>It's in the Git repo
<PotentialUser-79>how do i use my phone to bypass the problem of guix bing stuck on network connection. Alternatively how can I hack the installer?
<bricewge>lfam: That's odd but the output is really fancy
<bricewge>PotentialUser-79: The installer is using connman and I'm used to network-manager so I can't really hep you on that front.
<mbakke>PotentialUser-79: do you have another computer with Guix installed?
<mbakke>IIRC the connection issue was fixed shortly after the release :-/
<PotentialUser-79>no, no i am trying to install over usb0
<bricewge>Normally you just plug your phone by usb. In the phone enable the USB modem or sharing connection trough USB. And that's it
<bricewge>You end up with a new interface with a IP from the DHCP server of the phone.
<lfam>mbakke: Can we make a new installer image?
<lfam>Just cherry-pick the fix?
<lfam>I mean, I know there's more to it than "just" that one thing
<jonsger>civodul: can we add the xdot example to the manual. Thats the best
<mbakke>lfam: let's do a 1.1.1 release with the recent core-updates goodies (and bugs..) instead :-)
<civodul>jonsger: yes, that's what i was thinking
<civodul>let me do that
<PotentialUser-79>i am at Manual partitioning stage and how can I be sure which partitions will be formatted?
<mbakke>PotentialUser-79: you need to format the partitions manually
<PotentialUser-79>The message say We are about to format your hard disk. How do i choose only one partition to format?
<pkill9>how do you get certbot to do the initial certificate generation etc?
<mbakke>PotentialUser-79: oh sorry, I did not realize you were in the installer, I thought you were installing manually in a shell
<cbaines>pkill9, how have you configured certbot to handle the verification?
<PotentialUser-79>I want to cry ;-(, I have got stuck on manual installation because I could not edit the scm config file. Now on graphical install I can not select partition to install guix.
<pkill9>cbaines: dunno, i copied the example from the manual and removed 'deploy-hook'
<PotentialUser-79>if I go back to manual installation how do i edit the scm config file on another machine?
<civodul>PotentialUser-79: you can select partitions before you get to the "We are about to format" dialog
<ngz>A package of mine has this embarrassing cmake flag: -DDEFAULT_ASIAN_TTF_FILE=\"/usr/share/fonts/truetype/arphic/bkai00mp.ttf\" What should do you suggest to use instead?
<PotentialUser-79>so why after that it does not say which partitions it is going to format?
<PotentialUser-79>why does it say it wants to format my disk
<PotentialUser-79>does it mean WHOLE DISK?
<pkill9>cbaines: how do you configure it to do that?
<cbaines>pkill9, I think the default is to use the HTTP challenge. Assuming you've got NGinx or whatever webserver up and running, serving files from the webroot certbot is configured with, I think the system reconfiguration prints out the script to run
<pkill9>i thought it handled it
<pkill9>i don't see the script printed out
<civodul>ngz: refer to the actual font, assuming that font plays a crucial role in the package
<cbaines>I belive the certbot wrapper script is /var/lib/certbot/renew-certificates
<pkill9>yes it is, thanks
<pkill9>the manual says it's supposed to handle the initial creation of this, but it doesn't
<mbakke>lfam: I believe the main font problem from the core-updates merge is that that bitmap fonts are no longer supported by Pango (which GTK+ uses): https://gitlab.gnome.org/GNOME/pango/issues/386
<mbakke>see also https://fedoraproject.org/wiki/BitmapFontConversion
<ngz>civodul: AFAIK, we don't have arphic (?) font. And I'm no specialist of CJK fonts. So I wonder what font could replace it. Moreover, some fonts packages I checked provide .ttc, not .ttf. I don't know if that's the same.
<ngz>civodul: Or do you mean something else?
<civodul>ngz: that's what i meant; or maybe just don't pass that font?
<civodul>s/font/flag/
<mbakke>lfam: gs-fonts are bitmap fonts, not sure if converting them will give satisfactory results
<NieDzejkob>PotentialUser-79: when in doubt, make a backup ;)
<lfam>Thanks for investigating mbakke
<ngz>civodul: this would mean cn_CN users would not be able to use the package.
<NieDzejkob>I'm not sure why you want to edit the file on another machine, specifically, but if the network is up, you can use many different methods
<NieDzejkob>say, pipe the file through netcat
<NieDzejkob>or scp it over if you have sshd with password authentication on the other machine
<PotentialUser-79>@Niedziejkob, i tried to use tramp to edit the file
<Blackbeard>Guix pull gives me an unexpected http status code: 504
<PotentialUser-79>would sftp work to transfer it to another machine?
<NieDzejkob>why wouldn't it?
<PotentialUser-79>i'will try and see what happens
<pkill9>i got my domain registered with letsencrypt woo
<civodul>ngz: or that they would have to install fonts by themselves, no?
<civodul>Blackbeard: yeah looks like Savannah is down or something
<lfam>Savannah is rebooting ATM
<ngz>civodul: You are probably right. I'll leave it blank. Thank you!
<lfam>Hang in the #savannah channel for notifications
<PotentialUser-79>@Niedzejkob, ostatnia proba
<Blackbeard>civodul: ok thanks
*lfam gets way too familiar with alsa-lib
<pkill9>I want a list of all packages that you would run from a desktop launcher, and then have guix automatically install them if they don't exist
<pkill9>so you just open software you don't have installed like it's installed, heh
<pkill9>and installation is basically just syncing
<cbaines>pkill9, I've seen this happen in Gnome, you effectivly include package search results in the shell, so they're mixed in with software you have installed
<cbaines>Not working on Guix yet of course
<NieDzejkob>PotentialUser-79: daj znać jak ci pĆ³jdzie, w takim razie :D
<PotentialUser-79>mam pomysl
***familia_ is now known as familia
<lfam>Is there a way for C code to export environment variables besides setenv(3)?
<lfam>Maybe "export" is not the correct word. I mean, to set a variable in the context of the running program
<lfam>I am working with alsa-lib, testing a feature that would make it read and use an environment variable that would contain a list of paths. In most cases, it works as expected, but there is one codepath that fails, and the environment variable seems to truncated when fetched with getenv(3)
<lfam>So I'm wondering if the program is setting the variable somehow, but there aren't any setenv calls
<NieDzejkob>it could be modifying the strings returned by getenv or messing with environ calls
<lfam>I'm printing the value of the pointer returned by getenv immediately after the getenv call
<lfam>But it only shows the first entry in the variable, when it should contain two entries
<lfam>Colon separated entries
<lfam>It works in other parts of the codebase
<lfam>Quite confusing
<NieDzejkob>there's also the option of firing up the program in gdb, enabling reverse debugging, getting to the point where the variable is not what you expect, setting a watchpoint on it, and continuing in reverse
<lfam>I was afraid of that (don't have any experience with gdb)
<lfam>Would that work for environment variables? The C variable has a short lifetime and I'm printf-debugging it at every step
<lfam>I know what's wrong. I didn't realize that strtok_r would mutate the environment
<leoprikler>strtok is evil
<PotentialUser-79>I am trying to crate config https://pastebin.com/uHkNmEUe
<PotentialUser-79>i get error about mapped devices in config.scm:20:14
<PotentialUser-79>I am trying to use it on HD with MBR
<rekado_>PotentialUser-79: where has your definition of mapped-devices gone?
<PotentialUser-79>I thought it was not needed for my case
<rekado_>then you should remove the ā€œ(dependencies mapped-devices)ā€
<rekado_>because that refers to mapped-devices, which is now undefined
<PotentialUser-79>still the same error
<PotentialUser-79> https://pastebin.com/TvdPCYFk is this what you meant?