<sturm>Hi Folks. Where does syslog output go to by default? I'm trying to figure out why lsh is rejecting my public key auth, but can't find the logs. <sturm>(I'm pretty much running the example desktop config.) <sturm>Figured it out - lsh used the "~/.lsh" directory and a different public key format. <sturm>Hmm, no ls tools for converting private keys though... that's a shame. <sapientech>hi all, quick question regarding pure environments. When guix does not have a particular python package, what should I do to wrap it up in an environment? <sapientech>should I go ahead and use pip, and then link to it? or should i make my own guix package for the package? <sapientech>looking further into the docs it appears guix import could be something im looking for <sapientech>ah, okay so trying $ guix import pypi pygame, but getting an error i can't find online or in the logs: guix import: error: no source release for pypi package PyGame <sapientech>continuting on with dthompsons great writeup with guix and ruby, looks like he uses bundler in a guix env to grab a dependency <ng0>hi. i don't know if perl-net-psyc was already a subject of an email, but i found a bug in perl itself. this is currently being discussed/fixed/etc.. when perl with perl5 comes to fix it the way one person proposed now, can i include the fix as a patch? <sapientech>okay got things working for the most part: created a package.scm for the missing dependency. one question i have, however is that environment --pure seems to remove the dependencies i use in package.scm <ng0>sneek: later tell lfam: thanks for merging gmime :) <ng0>i can update my systems again <ng0>say, why do we have wicd and network-manager with GNOME? <ng0>for %desktop-services i mean <ng0>is network-manager not fully functional? <ng0>i know what wicd is, i have used network-manager on gentoo for a very long time and before that wicd <ng0>i wonder why we pull in both. is one a package which just pretends, to satisfy gnome? <ng0>can someone take a look at my system's config.scm? I don't know what's broken.. <sapientech>ng0: yes, right below it shows network-manager and the differences between both. or do you mean why are both installed, when one could suffice? <ng0>that is not my problem right now… <rekado>Info mode in Emacs 25.1-rc1 looks nicer. There’s a new face “Info-quoted” for embedded code. <rekado>I guess I can now use variable-pitch-mode in Info pages too. <ng0>i'm could now do the months delayed system update, but my config file fails me :/ <rekado>ng0: is /dev/sda3 really a swap device? <ng0>afaik and as far as parted knows, yes <rekado>did you configure the partition? <ng0> https://ptpb.pw/ltXv this did not change in the last months and also the config file did not change.. if it changed it must be some module i removed <rekado>ng0: re perl-net-psyc: Are you saying you want to apply a patch to the “perl” package? Or something else? <rekado>ng0: according to the log your new system configuration is in fact complete. It’s just restarting services that fails. <rekado>I don’t know why starting the swap service fails. <efraim>i saw that the guix devel snapshot got updated, was it also supposed to go from 0.10.0-0.abcde to 0.10.0-1.abcde? <rekado>could you show us the output of “sudo herd status”? <efraim>ACTION still has one more day of hectic move stuff, and then MAYBE back to normal tomorrow <ng0>rekado: i'm waiting on a change in perl5 on a bug I found,the unchanged perl5 makes perl5 segfault in several minor versions due to a change they introduced at some point. <ng0>it will be patched upstream, if they decide to, and it would be good to have it here then too <ng0>oh... swap is stopped <ng0>stopped: avahi-daemon and swap-/dev/sda3 <efraim>another gd cve, fixed by their latest release if anyone wants to take that. I can't get to it until tomorrow at the earliest probably <rekado>ng0: so… you would like to add the bugfix as a patch to our “perl” package? Many packages depend on “perl”, so it would have to be patched in core-updates. It wouldn’t be available until the next core-updates merge. <ng0>i am sure that it is swap because i set it up like this, andi have a swap every time i start. but "swsusp" is new to me, as i had "sw" used before. <rekado>in that case maybe it’s better to wait for it to be fixed by a new release. <rekado>ng0: what happens if you run the swapon command manually? <rekado>the log tells us that “swapon /dev/sda3” returns “Invalid argument” <ng0>re perl: it will depend on which version(s) they apply the fix. 5.24.0 is stable at the moment.. it happens at least with 5.22.0, 5.21.0, and one more <ng0>swapon: /dev/sda3: software suspend data detected. Rewriting the swap signature. <ng0>and now it's on again <ng0>isn't swap supposed to be just.. swap without problems? <ng0>i never had a swap file system with probelms <ng0>andif i run system reconfigure, the swap is off again <ng0>now it fails with device or resource busy <alezost>efraim: re guix snapshot: yes, I think Andreas forgot to update the revision number <efraim>I was wondering why it didn't update :) <ng0>so i turned swap manually off <ng0>and now it succeeded <alezost>efraim: I've sent a message about it to the ML <ng0>okay.. this worked. but i also have to start a thread on libreboot+guix with linux-libre 4.7, as we lack something parabola manages to do: recovering from suspend with certain devices. <ng0>i tested gnome and xfce, full recovery never happens with the version i used when i reported to the libreboot list. <ng0>should've been fixed with linux-libre 4.6.4 which i did not have at the time of testing, 4.7 did not fix it either for us <efraim>whoops, ran `guix gc' and I meant to run `git gc' <ng0>also 02:00.0 Network controller: Qualcomm Atheros AR928X Wireless Network Adapter (PCI-Express) (rev 01) just randomly decided to rf-kill itself with linux-libre 4.7 <ng0>in combination with recovery suspend not working, some of the people who do the kernel package should look into it <ng0>is my new host making bug-guix slow down on accepting the email? my address is known to the list ***Digitteknohippie is now known as Digit
<ng0>why does bug-guix list take almost 12 hours now and my bug email has not been whitelisted? <ng0>i can understand hours.. but half a day? infrastructure problems? <civodul>the first post is always delayed, but subsequent post show up in a few minutes at most <ng0>same for guix-devel .. and this is not the first message from this address. i wonder if it ever got through <ng0>i'll look on my side if something hang up <ng0>i should start to block hosts.. they make my logs unreadable with their failing attempts. <ng0>so far it looks like email got sent out <jlicht>civodul: I'm having some problems with my email provider right now, but RE: your tiny screencast: <jlicht>sneek: later tell civodul: the first command (installing hello) and then running it only works if someone has set up their guix properly (by setting PATH) <jlicht>sneek: later tell civodul: so it might lead to people 'installing' guix, and then trying the examples they saw on the front page (which won't work) <jlicht>When building older versions of node, I have the problem that <linux/limits.h> cannot be found <jlicht>Regardless of having the linux headers installed <jlicht>Do I still need to do anything to make sure gcc finds these headers? <davexunit>random question: has anyone had any decent experience learning/using a free software cad program? <bavier>I was looking into KiCAD recently, but haven't really used it much yet <ng0>i am sorting old emails out.. i saw that there was at least one email where the .po files were included, making it 8000+ lines.. even when it seems very obvious, we should point out that the generated files should be excluded and only the files one did apply changes in should be sent - and do not make your email 10000 lines long <ng0>oh stillgoing... 11k lines <ng0>or at least this looked like it, could also be my archive which is broken due to some copying <ijp>let's accept a plea deal <jlicht>#guix, now also offering legal advice :-) <ng0>has someone figured out vlc failures on core-updates? I can't find a working solution. <rekado>ng0: AFAIK suspend is broken in libreboot for a couple of devices. <rekado>that’s why I’m still using the 2015 release of libreboot <fps>hi. btw: what happens if the same guix build <package> is called simultaneously several times?