IRC channel logs


back to list of logs

<rekado>I assumed it would be some conflict with XDG_* vars, but these aren’t set
<apteryx>so they probably need to be set?
<rekado>I don’t know, maybe
<rekado>but there are none in $GUIX_PROFILE/etc/profile
<rekado>so I wouldn’t know what they should be set to
<rekado>it’s also something where I’d give up as a newcomer
<apteryx>perhaps you need to install gtk or the likes that sets it
<apteryx>and some icon theme
<apteryx>agreed it's hard
<rekado>that seems … like a bad idea in general? I have a vague memory of problems relating to the installion of gtk+ in a Guix profile.
<rekado>the Application Setup section may need to be expanded once we’ve got a good recipe with a fix
<apteryx>please :-)
<rekado>we’ve had a couple of reports relating to icon themes in the pats
<rekado>I wonder if this could perhaps be taken care of by a profile hook
<rekado>i.e. auto-detect the “situation” somehow™ and add a suitable icon theme unless an icon theme exists already
<trofi>civodul: could it be that guix does not handle gracefully situation when daemon does not support compression methods provided by substituter? I think I see a nar in zstd and daemon does not supports it (no zstd bindings):
<civodul>trofi: this code filters supported compression methods
<civodul>so it should be fine
<civodul>or did you experience otherwise?
<trofi>civodul: AFAIU when filter length is 0 last match crashes
<civodul>currently ci.guix provides zstd + gzip + lzip, it's that generous
<trofi>substitute: Throw to key `match-error' with args `("match" "no matching pattern" ())'.
<trofi>I think in I see a user reporting the substitution server that provides only zstd
<trofi>Here is a bit of nar metadata i extracted by adding a few print statements:
<civodul>ah got it: the servers does only zstd, but that daemon does not understand zstd
<civodul>so at the very bottom of the file, we need to add a clause for the empty list
<civodul>and throw from there
<trofi>aha. i'll need to wrap my head around exceptions in scheme and where they get caught for these archives. thank you!
<bdju>all my stuff is crashing like crazy... :/
<bdju>not sure if it's a nasty coincidence or if there's something related between them all
<bdju>qutebrowser, gajim, sometimes nheko, had a sway freeze the other day
***jave_ is now known as jave
***califax- is now known as califax
<lfam>I'm still having trouble with Guix System based on the last few days of commits. It hangs after boot
<sneek>lfam, you have 2 messages!
<sneek>lfam, mhj[m] says: Guix system list-generations has this: commit: bbe52d3b86c06857eca2722fada77be060b145f5 with the 5.15.17 kernel
<sneek>lfam, civodul says: + nckx a system built from 43dd34c7777a212c99a97da7a2c237158faa9a1b fails to boot on my laptop (hangs before syslogd is started)
<lfam>It becomes more or less unresponsive
<lfam>sneek: later ask mhj[m]: Is that generation based on commit bbe52d3b86c working or not working?
<sneek>Will do.
<lfam>Hm, seems we have a problem
<nckx>OK, I just logged in (morning all) to see if it was safe to update yet. I guess not.
<lfam>It's hard to understand what goes wrong. The system seems to work for a minute or so and then becomes unresponsive.
<nckx>It's not one of those spooky ones that persists after a rollback, is it?
<lfam>Sometimes it reaches the login prompt and I can login, sometimes not
<lfam>Both me and mhj[m] have no trouble rolling back from GRUB
<nckx>OK, thanks.
<lfam>My system is pretty bare bones, no X or Wayland. But I do have dbus-service
<lfam>I feel like these dbus errors are new
<lfam>I have dbus and elogind services, as concessions to modernity
<lfam>The only other non-base services are gpm, openssh, ntp, wicd, and a custom substitute-url for the guix service
<lfam>But, I doubt those services could really break anything
<apteryx>seems your network is broken
<apteryx>or perhaps it's just not up yet
<apteryx>it's weird that this is new
<apteryx>what changed? the kernel?
<lfam>The network isn't up, but that's expected, because the interfaces or either disconnected or unsupported by linux-libre
<lfam>I'm using linux-libre to aid debugging
<lfam>If those error messages are only about the network being down, then they aren't relevant to whatever this bug is
<lfam>Well, time to bisect
<char[m]>which package is ldd a part of?
<lfam>char[m]: glibc
<char[m]>lfam thank
<char[m]>is it guix acceptable to pass the LIBRARY_PATH environment variable to ldd -rpath?
<dust_>I'm getting 'Too many heap sections: Increase MAXHINCR or MAX_HEAP_SECTS' when invoking guix pull. I'm looking at irc logs for solution. I have 24GB of ram, so ram is not the problem I believe. Can anyone help ?
<sneek>dust_, you have 1 message!
<sneek>dust_, abrenon says: I program in Haskell but haven't contributed anything haskell-related to guix yet
<lfam>dust_: Are you pulling to any particular commit, or just the default (latest)?
<dust_>lfam: latest
<lfam>Okay. And what CPU architecture do you use?
<lfam>32-bit or 64-bit?
<lfam>Alright, I'm trying it now, also on x86_64
<iyzsong>i just pull successful on x86_64, doing a system reconfigue now..
<iyzsong>dust_: are you on a foreign distro?
<lfam>It worked for me, going from commit d331bd0 to c1043fd on Debian
<lfam>But, I didn't build anything: all substitutes
<lfam>dust_: Were you using a large value for --max-jobs or --cores?
<dust_>I didnt use any flags. I was making some package defs for some haskell packages. Some of them showed this error while building. Later I did guix pull and it gave me the same.
<dust_>This is my Guix version : c1043fd60d273c61c80ecfd31fa3a4ea1b24639a
<lfam>Were you trying to `guix pull` with your custom package definitions?
<lfam>dust_: Ah, you should have started with that info :)
<lfam>There's an infinite loop in your packages, most likely. So you are running out of memory
<dust_>lfam: Oh. My bad.
<lfam>If you want, you can share your packages on <> and we can help find the issue
<dust_>lfam: What best time in a day to have a conversation in guix irc chat. Most of the time I feel like I'm talking to myself.
<lfam>European and American day time
<lfam>The loop is probably a package dependency loop
<lfam>That's my guess based on experience
<dust_>lfam: That fixed my issue. I commented every this that didnt build.
<dust_>Thanks a lot.
<lfam>We are happy to help!
<dust_>lfam: I have asked this before in the chat. I have made 20+ package defs for haskell. And one of them seems to use different version that guix channel provide. And this is for testing. So should I skip testing ?
<lfam>dust_: The choice is yours. You should test your packages "in practice". Check if they work for your use case
<dust_>I'm pushing it to main Guix channel.
<lfam>Then you should run the tests
<lfam>We always run the tests in Guix
<gnoo>how can i change %default-profile-hooks so that each guix invocatoin uses that value?
<gnoo>is there a config file or similar?
<dust_>lfam: If I were to build a different package version then I have to make pac def for every depending versions.
<apteryx>gnoo: I don't think that's user customizable thus far; you'll have to change in in the sources
<iyzsong>i think by using 'guix build' instead of 'guix package', you can build the <profile> record (Declarative profile) with custom hooks..
<lfam>dust_: Maybe you can use 'package-input-rewriting':
<gnoo>apteryx: thanks, should i define a derived pacakge for guix with a patch?
<apteryx>not sure what are your goals; if it's just to experiment, you could edit the sources and simply use ./pre-inst-env guix do something
<gnoo>there are some actions that i don't care in %default-profile-hooks so i want to optimize them out
<gnoo>hmm, maybe i can just use guix as a wrapper for (cd ~/guix/ ; ./pre-inst-env guix "$@" ) ?
<pingpongball>Is there AUR like thing similar to arch, void-packages like in voidlinux?
<pingpongball>i i want to install like robo3t ,
<pingpongball>how many packages are in guix with unoffical made for guix?
<bdju>pingpongball: other users host channels you can add to get additional packages. nonfree stuff can't be discussed here, fyi.
<pingpongball>:b  oh i forgot non-guix
<bdju>I just use the default channel atm so can't be much help
<pingpongball>:)  fine
<dust_>lfam: , here is code
<dust_>I'm having problem in ghc-http-media
<gnoo>oh my god, the guix Makefile is 7257 lines long
<gnoo>at that point, why not demand GNUmake and make that much shorter??
<gnoo>when compiling guix, i get: ice-9/eval.scm:293:34: In procedure abi-check: #<record-type <origin>>: record ABI mismatch; recompilation needed
<gnoo>i'm trying to build v1.3.0.rc2
<gnoo>before make, i did: git checkout -b stable v1.3.0rc2
<vivien>gnoo, you need to make clean and make again
<vivien>The Makefile is not written by hand, it it generated from with Automake
<gnoo>oh, ok, i'll edit the then, thanks
<civodul>Hello Guix!
<cbaines>morning o/
<brendyn>With this new way of listing inputs, how do i specify a particular output of an package?
<civodul>cbaines: like this: `(,foo "lib")
<civodul>brendyn: ↑
<civodul>cbaines: it looks like bordeaux.guix only does lzip, is that right?
<civodul>would be nice to have zstd
<civodul>it makes a noticeable difference performance-wise:
<cbaines>cbaines, indeed. I'd prefer to not just store everything compressed as zstd and lzip to avoid taking up lots more space
<cbaines>if it's possible to work out when clients would prefer zlib, it would be nice to provide zlib compressed nars in that case
<dust_>Its is okay to use hackage-uri instead of uri in origin record ?
<dcunit3d>if i'm getting "bad use of '=>' syntactic keyword", how do i fix that?
<rekado_>dcunit3d: we need more context to understand what’s going on
<rekado_>dust_: probably, yes. In (gnu packages cran) we’re using cran-uri everywhere.
<dcunit3d>the command i'm running is `guix reconfigure -L ~/.config/guix/systems ~/.config/guix/systems/hersai.scm` and that system depends on ./base-system.scm
<dcunit3d>the code is here:
<dcunit3d>the error is occuring in the %dc-desktop-services which uses the hashrocket => macro, but i just don't know enough about scheme in order to figure it out.
<dcunit3d>in hersai.scm, i have (define %hersai-desktop-services (modify-services %dc-desktop-services ....) which mimics the call to (modify-services %desktop-services ... ) in base-system.scm
<cbaines>dcunit3d, you're not importing the module that provides modify-services, which is (gnu services) I believe
<dcunit3d>thanks cbaines, that was it. now i'm onto another issue lol but it's progress.
<dcunit3d>wow that was nowhere near the direction i was looking in.
<jpoiret>hello guix
<gnoo>i'm installing guix with './pre-inst-env guix package -i jq' and it's downloading 'subversion-1.14.1 3.4MiB' for some reason
<efraim>civodul: %final-inputs-riscv64 is a hack, it adds (list gcc-final "lib") to %final-inputs. The %current-system check has to come in (guix build-system gnu) because of when it re-evaluates %current-system from x86_64-linux to riscv64-linux.
<efraim>I do wonder if it's related to some of the issues we've been seeing with guix deploy on the aarch64 boxes getting x86_64 binaries
<efraim>also, the true fix™ for %final-inputs-riscv64 is for me to track down and fix gcc correctly linking to gcc:lib, similar to this:
<civodul>efraim: could you add a comment next to %final-inputs-riscv64? :-)
<civodul>normally, is always found by dlopen because we modify GNU_USER_TARGET_LIB_SPEC in gcc.scm
<civodul>that modifies the built-in GCC "spec" such that it passes -rpath=gcc:lib/lib
<civodul>so i suspect the problem is that we're not modifying the right file for riscv
<civodul>see gcc.scm:241
<AwesomeAdam54321>Does anyone know how to solve this problem? 'gnustep-config --objc-libs' needs make at runtime, and it works since gnu-make is a propagative-input of my gnustep-make package definition
<AwesomeAdam54321>But when gnustep-base has gnustep-make as an input, during configure 'gnustep-config --objc-libs' outputs nothing, and that only ever happens if make isn't available
<civodul>efraim: following up: could you check "gcc -dumpspecs" on riscv with Guix's gcc?
<jpoiret>AwesomeAdam54321: can't you patch gnustep-make to directly refer to the make of gnu-make, instead of relying on PATH?
<efraim>civodul: sure to both
<jpoiret>that way you won't need to propagate make
<jpoiret>the issue here is that since gnustep-make is not propagated by gnustep-base, gnu-make is not installed to the profile either
<efraim>civodul: output of 'gcc -dumpspecs' from gcc-final:
<attila_lendvai>nckx, have you finished your experiments with channel dependencies? i'd like to delete the test commits from my channel...
<efraim>on Debain ld-linux is at /usr/lib/, so I'm guessing with a name like that I'll need to tell it to search a little harder before I can remove all the (add gcc:lib) phases I've added in commencement.scm
<efraim>gcc:lib is at /gnu/store/cgxghxvqds82d64mjg01wy0vb597vx88-gcc-10.3.0-lib
<gnoo>package `foo@1.25.0' has an invalid input: #<package intltool@0.51.0 gnu/packages/glib.scm:539 7efd5206b1e0
<gnoo>what does this output mean?
<AwesomeAdam54321>jpoiret: I should try that, thanks
<attila_lendvai>it'd be very helpful if someone could push this simple change to service configurations:
<efraim>but having stared at gcc previously I don't like startfile_prefix_spec
<civodul>efraim: it does have -rpath=/gnu/store/cgxghxvqds82d64mjg01wy0vb597vx88-gcc-10.3.0-lib/lib so that part looks good
<civodul>and it has -lgcc_s as well
<efraim>in *lib should it have -Lgcc-lib also?
<efraim>I've pulled up 'gcc -dumpspecs' from x86_64 also to check myself
<efraim>*startfile_prefix_spec is empty on x86_64
<efraim>*startfile_prefix_spec is also empty of aarch64-linux
<efraim>and empty on powerpc-linux
***Fd9a_ is now known as Fd9a
<efraim>civodul: similar with alpine linux, musl and /lib in place of /lib64. I expect to be poking around there a lot soonish™
<tschilptschilp23>Hi! I just set up a fresh headless guix and am having troubles with guix home. I am very confident, that I have 'gnu packages guile' in the 'use-module' section of my configuration, however I keep getting 'unbound variable guile, did you forget to put ...'! Does anyone have a hint for this?
<efraim>tschilptschilp23: try 'guile-3.0', I don't think we actually have 'guile' itself defined
<tschilptschilp23>Probably worth mentioning, the above is a VM. My Laptop does not even need 'gnu packages guile' to install guile via guix home... But maybe there are some remainders somewhere from my pre-huix-home setup, that I am not aware of.
***attila_lendvai_ is now known as attila_lendvai
<jeko>Hey Guix ! I have the package `guile` installed in my profile but I can't see its info documentation in Emacs info mode
<jeko>Do I need an extra step to make it available from Emacs ?
<tschilptschilp23>efraim: thank you, your comment made me double-check. All on my side - I just didn't quote the packages!!!
<civodul>jeko: hi! if you're on a "foreign distro", make sure INFOPATH contains ~/.guix-profile/share/info
<jeko>civodul: hey! INFOPATH does contains ~/.guix-extra-profiles/jeko/jeko/share/info and I can see into it
<tschilptschilp23>But now I run into another thing- at guix home reconfigure the last step ('building /gnu/store/xyz-home.drv") fails with 'guix home: error: mkdir: Permission denied'. I guess one could sudo that, but that's not exactly what I want to Do...
<jeko>civodul: Ooooooops!!! I found it !
<jeko>civodul: was lost between other Guile libs info docs ^^"
<jeko>Guile Config, Guile Hall, Guile Reference, Guile-Gcrypt, …
<jeko>I did not read the list carefully because I thought it was only libs
<civodul>jeko: ah yes, it's easy to get lost in that list!
<roptat>hey, time is short, remember to submit proposals for the Guix Days before Feb. 8!
<civodul>yeah! :-)
<jeko>can't wait for the talk list for Guix Days
<tschilptschilp23>OK, found it thanks to bug-report 50941. Setting XDG_RUNTIME_DIR in the home-configuration does what I was searching for on a 'headless' (%base-services only in the 'global configuration') system.
<AwesomeAdam54321>jpoiret: setting the GNUMAKE variable to configure solved the problem with gnustep-make, but I've yet to figure out why gnustep-base still fails
<abcdw>hey guix!
<attila_lendvai>i wish guix package -u had a switch to only upgrade what is available as a substitute
***robin__ is now known as robin
<leinad>Recently, while browsing the Guix I wondered why some names are prefixed with %, like %default-argument-handler. Is this some sort of naming convention?
<leinad>s/Guix/Guix sources/
<civodul>leinad: yes, usually to indicate a "special" global variable
<attila_lendvai>leinad, it is. in lisp %foo is something that is private and usually has a public, slightly different version under the name foo. in Guix it's more like a general "this thing is special in some way".
<leinad>I see, thanks :)
<MysteriousSilve4>is it just me who couldn't access the savannah cgit?
<akonai>i think is down currently
<akonai>at least for me, going to just results in a timeout
<leinad>yep, I noticed this, too
<raingloom>came to check if it's down for others, first message confirms. :D sorry Ekaitz, i wanted to debug that librecad failure. guess it'll have to wait.
<ekaitz>raingloom: no worries mate
<ekaitz>it's probably something related with libboost
<raingloom>yup. my first hunch was either boost was upgraded and librecad uses and old API, or librecad was upgraded and our boost is old.
<ekaitz>raingloom: something like that should be, yes
<ekaitz>i'd fix it myself but I have to study gcc internals and it's going worse than i expected :)
<ekaitz>also updating texlive just filled my disk so I have to upgrade my packages in small blocks now wtf?
<raingloom>oof. yeah no offense to texlive but fuck texlive. it takes up an obscene amount of space.
<ekaitz>i had no issues in the past, but now i started my package update with more than 20 GB free and boom
<raingloom>i was so glad when i could delete the profile it was in. hopefully i won't need it for this semester.
<whound>Can anyone look into the series of patches that I have submitted ? From 53683 to 53692, 53684 supposed to be the first one.
<sneek>Welcome back whound, you have 1 message!
<sneek>whound, podiki[m] says: I have all of the haskell gi (and taffybar) package defs, but working on cleaening up to submit them; it needs a different gobject-introspection for cairo (I think same error you're seeing)
<MysteriousSilve4>"no offense to x but fuck x" :/
<whound>podiki[m]: Can you help me in making GHCjs pac def ?
<ekaitz>i found an integer overflow on size calculations done by guix gc?
<civodul>ekaitz: tell us
<ekaitz>civodul: I just sent a bug report
<civodul>ah good
<civodul>the estimates done by "guix gc" are often off due to deduplication
<ekaitz>take a look to this civodul
<ekaitz>those are a lot of megabytes
<ekaitz>A LOT
<civodul>but you know, the GC is *extremely* efficient
<ekaitz>more than my disk size, that's good
<ekaitz>and then texlive comes and it doesn't fit in my disk space
<civodul>sure but texlive is infinitely big, that's why
<trofi>how much do you have out of curiosity?
<ekaitz>77G for / (57 used)
<trofi>*nod*, reasonably small
<ekaitz>not that small imho, but yeah
<ekaitz>I have to expand it a little but now I'm afraid of doing it and breaking my files :)
<ekaitz>(and I also forgot how to do it)
***A_Dragon is now known as NotAFurry
<ekaitz>trofi: wanna remind me how to do it? :)
<trofi>which filesystem you are using? in the olden days i used gparted to resize filesystems with data from any contemporary livecd
<trofi>resize2fs should be able to deal with it (but it's a low level tool)
<ekaitz>which means danger
<ekaitz>i'll try with gparted later :)
<trofi>Should be as easy as dragging the ruler around:
<ekaitz>thanks for insisting on this, it's something I should've changed a while ago
***NotAFurry is now known as A_Dragon
<ekaitz>btw I have a secondary hard drive here i could use, but I don't know how to configure everything correctly for guix
<ekaitz>i could move the / to that drive and expand the home in this one
<gordon1>so my current-guix profile uses some version of guix, how can i customize that?
<gordon1>i just want to change the package definition
<leinad>not super sure what you mean but how about `guix pull --commit=...'
<zap>Hello guix!
<apteryx>gordon1: you could specify an exact commit in a channels file
<gordon1>i want to modify how it is compiled, not the commit
<gordon1>looks like nf2l97kmj5iccxcsdk5zvy6ysjs2dq4d-guix-f00ed4365 depends on guile-avahi, i want to remove that input
<civodul>gordon1: that's not something you can customize
<civodul>looks like you really want to get rid of Avahi though, perhaps you could make your case on the mailing list?
<gordon1>yep, looks like that's my only option
<gordon1>but in general it sounds a bit weird that i have no way of influencing how guix is getting build
<rekado_>gordon1: you do have an influence on that. Just not when you use an unmodified Guix package from the default channel.
<rekado_>the implied order in your question, however, is never how it would be done. You wouldn’t edit whatever has been built. You’d rather make it build from different instructions.
<gordon1>rekado_: sure, maybe i just formulated my question incorrectly, i'm perfectly fine of being able to specify my modified package somewhere in config to make it build new version of guix, the issue is here that i didn't find that place where i can specify my modified package, it looks like the only way of doing that is specifying a for of guix's repo with modified (guix packages package-management), no
<gordon1>way of doing it in the config
<gordon1>at least i havent found way
<gnoo>gordon1: hey! i'm doing exactly that, added a commit and then am building guix to install
<gordon1>gnoo: did you just do your local fork and added it there?
<gnoo>yeah, i git cloned the repo and changed the (source) to (source (git-checkout (url "file:///path/to/clone")))
<gnoo>with everything else inherited
<gnoo>i mean it's still building xD
<gnoo>here's the definition, really simple
<gordon1>yeah, see, in ideal world i would like to avoid that
<apteryx>civodul: copying /var/cache is going to take a couple more days, it seems. do you think we could mount what we have (> 10 TiBs), shadowing the curret /var/cache, and keep the sync in the background? Or would that cause too much disruption?
<gordon1>that will introduce a lot of extra work
<gordon1>of maintainig your own fork
<gnoo>i think modules are self-contained so using a config file to change it (which is what i want) seems hard
<gnoo>i was using `./pre-inst-env guix' for a while ago but it basically can't do anything like building packages or even guix pull
<gnucode>hello guix!
<gnoo>o/ gnucode
<gnoo>is your name josh by any chance?
<gnucode>My experiment with using only %base-services as for my main workstation is going interestingly. hahaha.
<gnucode>gnoo: yes. May I ask how you knew?
<gnoo>your username (which is different from nickname) shows that. it also shows your ip
<gnucode>maybe it's gnoo's job to "knew". that's a silly pun.
<rekado_>gnoo: I use pre-inst-env all the time to hack on Guix. It certainly can build packages.
<gnoo>hmm, i did pre-inst-env guix -L ~/.cache/guix/my-packages/ package-name and it spitted out errors which didn't happen with system guix
<gnucode>ok. do you just have a good memory of my IP? Or do you have a program that gives you an estimate on where I am based on my IP?
<gnoo>gnucode: you can do 'whois <ip>' and it'll show you the geographic region, i think
<gnoo>or at least your isp
<gnucode>gnoo that's pretty cool.
<gnoo>okay, it's been 10 mins and guix build is stuck in "updating checkout of 'file:///home/me/src/guix/'..." should i be worried?
<gnoo>also cpu usage is off the charts
<gnucode>gnoo if you are building guix from source, it does take a while for it to least for me.
<gnoo>rekado_: guix build: error: /home/me/.cache/guix/my-packages/pidgin-sipe.scm:10:2: package `foo@1.25.0' has an invalid input: #<package intltool@0.51.0 gnu/packages/glib.scm:539 7f34738d71e0>
<gnoo>rekado_: this is the error i got with pre-inst-env
<gnoo>gnucode: no i mean it's still not cloned
<gnoo>and it's local so that's too much
<gnoo>i did a 'git pull ~/guix' in /tmp and it did that in about 3-5 seconds
<gnucode>that's weird.... guix pull just have me an SSL error: unexpected status code 502...that may be an HTTP error code...
<gnoo>with guix build it's still in: receiving objects 75% [#################################################### ]
<gnoo>gnucode: the gnu servers are down i think
<gnucode>gnoo, oh. that's a new one.
<gnucode>also my autocomplete (I'm going doom emacs) is pretty annoying right now. Typing "one" then RET results in "clone1".
<gnucode>there's probably a clone1 in the chatroom. ahhhhaa.
<rekado_>gnoo: we don’t have a foo package, so I guess you’d have to fix your problem on your side.
<gnoo>rekado_: i stupidly removed the package name thinking talking about propretary packages is not allowed. but it is actually free software
<gnoo>the actual error is: guix build: error: /home/me/.cache/guix/my-packages/pidgin-sipe.scm:10:2: package `pidgin-sipe@1.25.0' has an invalid input: #<package intltool@0.51.0 gnu/packages/glib.scm:539 7f34738d71e0>
<gnoo>this happened when i had: (inputs (list intltool))
<gnoo>but it worked fine with system guix
<gnoo>while building guix, can i tell it to only build the english docs?
<apteryx>is savannah having problems? can't seem to fetch from
<gnoo>yes, gnu servers seem to be down
<apteryx>I've reported in in #savannah in case
<gordon1>gnoo: can you pls elaborate how do you use this?
<gordon1>i thought to replace guix in system you have to replace guix channel in channels.scm
<gnoo>sorry, i forgot to mention, you should probably remove "file://" form that
<gordon1>that's fine, just tying to figure out how to use this code
<gnoo>first i want to build the package and see if it actually works then only i will change system guix to use this one
<gnoo>nono, you can just add a (replace) for guix definition in package-management
<gnoo>i think
<gnoo>in gnu/packages/package-management
<gordon1>ah, that's just a safety measure
<gordon1>ok, so if i want to use my own repo (with changes in gnu/package/package-management) i have to change it in channels.scm
<gordon1>ok, then i'm confused
<gnoo>you can just do guix package -L ~/path/to/package-dir <package-name>
<gnoo>wait that's wrong
<gnoo>let me copy from my history
<gordon1>that's for just installing it, isn't it?
<gnoo>guix package -L ~/.cache/guix/my-packages/ -i <package>
<gnoo>you can stuff almost any guix package with -L
<gordon1>and if i want to replace every instance of guix in my /gnu/store it has to be channged in channels.scm, isn't it?
<gnoo>guix command
<gnoo>hmm, i don't really understand
<gnoo>you mean for packages that have guix as a dependency ?
<gordon1>well, my issue is that something in the system has guix as dependency, yes
<gordon1>one of such things - guix-daemon-service-type where i changed it
<gordon1>other place - guix home which has guix hardcoded
<gordon1>third place - profile itself that pulls and buildss guix from the repo specified in channels.scm
<gnoo>i haven't envisioned that far but i think this will solve that? info "(guix) Security Updates"
<gnoo>after a guix system reconfigure, it should install your version of guix
<gordon1>are you suggesting to try and graft guix itself?
<gordon1>ah, no i still need to change the very definition of guix itself to make it work i guess, to add (replacement foobar) property
<gordon1>so looks like there is no way around of forking guix
<gnoo>yeah, i already made a checkout of v1.3.0rc2 and have one commit on it :P
<gnoo>removing the xdg stuff %default-profile-hooks
<gordon1>gnoo: did you manage to make guix home work w/o xdg_runtime_dir dependency?
<gnucode>It would be nice if there librevox had an API to search/download various audiobooks. Trying to search their website on a console is turning out to be difficult...
<gnoo>gnucode: i have this in my config because sway demands xdg_runtime_dir
<gnoo>export XDG_RUNTIME_DIR="${XDG_RUNTIME_DIR:=/tmp/me}"
<gnoo>mkdir -p "$XDG_RUNTIME_DIR"
<gnoo>(in ~/.profile and ~/.bash_profile respectively the latter sources the former)
<gnoo>gnucode: sorry, that was for gordon1
*gnoo afk
<rekado_>gnoo: why such an old checkout?
<mhj[m]>Hilo guixers~
<sneek>mhj[m], you have 1 message!
<sneek>mhj[m], lfam says: Is that generation based on commit bbe52d3b86c working or not working?
<apteryx>if you miss 'journalctl -b', put this in your ~/.bashrc: alias bootlog='tail -n+$(grep -n "Linux version" /var/log/messages | tail -n1 | cut -d: -f1) /var/log/messages | less'
<mhj[m]>sneek: tell lfam later: The commit bbe52d3b86c doesn't work for me
<sneek>lfam, mhj[m] says: later: The commit bbe52d3b86c doesn't work for me
<mhj[m]> * sneek: later tell lfam: The commit bbe52d3b86c doesn't work for me
<mhj[m]>Uhh, am I using the right arguments for the bot lol
<csantosb>Ey #guix, I'm building a container with 'guix shell --container', and including gcc-toolchain.
<csantosb>Under /gnu/store I get gcc 11.2 as well as gcc 10.3, both with -lib directories.
<csantosb>Problem is I get an error about GLIBCXX_3.4.29 missing.
<csantosb>My software compiles (and finds) the wrong gcc 10.3 lib, and not the correct 11.2.
<csantosb>How can I set the correct gcc libs ? I have located the folder, but the random hash makes it unusable.
<csantosb>I mean, setting LD_LIBRARY_PATH solves the problem, but, how to guess the right one ?
<sneek>jgart: Greetings!
<char[m]>When does a patch need to be registerd in
<podiki[m]>sneek: later tell whound sorry, don't know anything about ghcjs, I would start with guix import and/or looking at other haskell packages to get started
<sneek>Will do.
<gnoo>rekado_: i was searching for some "stable" version, something people put a tag on so that i'll have to do less maintenance of it
<dongcarl>I know I'm behind but just read "The Big Change", what an incredible improvement to the input syntax. The Guix community's commitment to developer experience is truly laudable!
<civodul>dongcarl: hello! it's a pleasure to read those kind words :-)
<lfam>I'm still bisecting the "hangs after boot" bug, but it's not looking good for version 5.17 of the kernel
<sneek>lfam, you have 1 message!
<sneek>lfam, mhj[m] says: The commit bbe52d3b86c doesn't work for me
<lfam>There's only about ~30 commits left to bisect and most of them are in radio.scm and bioinformatics.scm
<trofi>you can also do 'git checkout specific-commit' and run a 'git bisect bad/good' against it if you have a hunch on what causes a regression. might save a few steps
<lfam>Yeah, after the current run finished, I'm going to give it a hint like that
<apteryx>if anybody is tired to reboot to update their udev rules, you may want to review!
<attila_lendvai>someone has set up a guix channel that is promised to be more agile in accepting quationable quality packages. can you please remind me what to search for? i can't find the mail.
***califax- is now known as califax
<attila_lendvai>drakonis, that's it, thank you!
<drakonis>there's an irc channel with the name of the org
<rekado_>the wireguard service on the aarch64 nodes on my desk doesn’t start any more
<rekado_>says “Configuration parsing error” – but nothing has changed
<rekado_>(I’m still booting the very first system generation)
<rekado_>“Name or service not known: `'”
<rekado_>maybe it’s no longer running on
<rekado_>uhm, never mind…
<rekado_>my toddler got into the room and unplugged all network cables :)
<podiki[m]>apteryx: udev reloading sounds great, would be a nice addition
<podiki[m]>(speaking of, I have a few udev rules packages I should submit)
<civodul>lfam: that hang-after-boot bug, it's most likely due to the kernel, no?
<civodul>apteryx: hi! do compressed debug sections break .gnu_debuglink rewriting?
<paladhammika>hello guix! :)
<paladhammika>I have interest in contributing to the guix wiki ( what is the background of this wiki?
<lilyp>note that it's a third-party wiki and we have no say over it
<lilyp>worse than that, it's hosted by a person whose largest contribution is IRC beef
<paladhammika>i see. so completely unconnected and hosted by a person with questionable inclinations.
<apteryx>civodul: I questioned myself too about it, but I don't have a firm answer. Note that due to 48907 we have to revisit this anyway.
<the_tubular>Are the Savannah server back up ?
<nckx>Works here.
<nckx>(Hi apteryx :)
<apteryx>podiki[m]: it's automatic if the udev.d directory doesn't change place; udev uses inotify to notice itself
<apteryx>nckx: hello! :-)
<the_tubular>Nice, thanks :)
<the_tubular>What happened exactly ?
<apteryx>not sure, the VM got unresponsible with high load; a reboot apparently fixed it
<apteryx>thanks to bandali!
<apteryx>rekado_: haha, thanks for the laugh
<efraim>helping clean up
<efraim>I remember when mine got big enough to help me remove the outlet protectors. Those tiny fingers make prying them out so much easier
<lfam>nckx: Any chance you'll be able to look at (linux-libre 5.16 doesn't work with our default-initrd-modules)?
<lfam>If "It would be great if we didn't have to manually edit the
<lfam>configs each time we add a new kernel series" is too hard to accomplish, maybe we could set these config options via the Scheme interface rather than the config files
<apteryx>efraim: that didn't sound right ^^
<lfam>sneek: later tell mhj[m]: I filed a bug about the "hangs after boot" problem: <>
<apteryx>lilyp: thanks for the comments/review! I've sent a fresh batch of comments back and added the doc I mentioned locally.
***iyzsong- is now known as iyzsong
<zap>Is anyone also using manifest for emacs? If I update profile emacs is using, how to switch to the new profile generation without restarting emacs?
<lilyp>zap: load your subdirs.el
<lilyp>that'll prepend the new profile onto %load-path
<lilyp>ehh... load-path of course
<lilyp>(we should really document that in the cookbook)
<zap>lilyp: :)
<zap>lilyp: Wait is subdirs.el part of emacs?
<podiki[m]>yeah, that would be a nice thing in the cookbook, emacs management with guix (and things like guix management from emacs?)
<podiki[m]>I'm wondering how I would share that with a computer that wouldn't use emacs from guix, just for package management, but normally I want emacs in the same profile
<podiki[m]>maybe some scheme in the manifest to conditionally include the emacs package or not
<podiki[m]>or just set emacsloadpath manually
<zap>lilyp: ah found it. Thank you for the tip
<pinoaffe>zap: I personally have a somewhat hacky setup where I change the relevant environment variables of my running emacs instance and the values of the emacs variables `exec-path` and `load-path`
<jeko>yooo guixters !
<apteryx>podiki[m]: why not use your emacs environment with guix everywhere? that's much easier
<rekado_>zap: I have a solution that works for me but is probably both primitive and overly complex. I have a manifest in .emacs.d, which installs to .emacs.d/.guix-profile, and I use direnv so that when I enter the directory that profile’s etc/profile is loaded.
<rekado_>there I start emacs, so EMACSLOADPATH is set to the location of the profile
<rekado_>when I install new packages they are readily available in the running Emacs, because EMACSLOADPATH doesn’t need to change.
<zap>pinoaffe: yea I do change these manually time to time but I'm getting tired of it :)
<zap>rekado_: Hm.. This is exactly what I do (appart from direnv). I just load it with "guix environment -p ~/.emacs.d/.guix-profile -- emacs"
<zap>Im not sure what subdirs.el does but it contains full paths to store
<zap>ah but it is inside the profile...
<pinoaffe>zap: I'll send ya link to my setup soonish, once I push the current version to git
<the_tubular>I'm not sure I'm foilowing your conversation, would you mind sharing your guixy emacs config so I cant take a look at it?
<zap>pinoaffe: Cool will be interesting to see your config
*zap got to go now
<the_tubular>I'm looking for a way to manage emacs with guix-home and I haven't figured it out yet
<ryanprior[m]>just heard about hvm, a lisp-y pure functional language runtime:
<ryanprior[m]>might be a neat thing to use with Guix
<rekado_>the_tubular: I have yet to try Guix home