libera/#devuan/ Wednesday, 2019-01-16

* gnarface is suddenly having deja-vu about this00:01
arminhttps://lists.freedesktop.org/archives/systemd-devel/2019-January/041971.html18:57
arminwahahaha18:57
xrogaancould this be solved? http://dpaste.com/35GRG2A18:58
xrogaanoh, maybe it's due to that mail.18:59
KatolaZxrogaan: what does "apt-get upgrade" say about that?18:59
KatolaZaptitute tends to play a bit too smart, and sometimes its predictions might be off...19:00
KatolaZ(sorry, I meant "apt-get dist-upgrade")19:00
xrogaanupgrade ignores anything that breaks something.19:00
xrogaanoh19:01
KatolaZdist-upgrade19:01
xrogaanThe following packages have been kept back:  libsystemd0:i38619:01
xrogaani386 and x64 versions mismatch19:01
KatolaZok fine19:01
KatolaZwhy would you need an i386 version of libsystemd0 at all?19:02
xrogaanPaquet : libsystemd0 Version : 232-25+deb9u719:02
xrogaanVersion : 232-25+deb9u8 État: installé (232-25+deb9u7), upgrade available (232-25+deb9u8)19:02
xrogaansomething something pulse audio19:02
xrogaansomething something steam19:02
xrogaanIDK19:02
KatolaZok19:02
KatolaZis the mismatch a problem then>19:02
KatolaZ?19:02
xrogaanit's why i386 fails, I'll take a look at debian's tracker to see if it's upstream19:03
gnarfaceSteam isn't great at figuring out what it needs19:06
gnarfaceit could be missing more i386 libraries than just that one19:07
xrogaani'm not missing anything19:07
xrogaanThere is no stretch-security in that "à'é_çtéà"ç_'(è19:08
gnarfacehmm. there was only once or twice, i think, that somehow a wine package pulled in a i386 version of something i had to remove to make sure it used the amd64 one when it was supposed to19:08
gnarfaceto make sure steam would not prefer the i386 one over the amd64 one19:08
xrogaansteam needs sdl i386, which requires pulseaudio, which requires systemd19:09
xrogaanSome stuff from steam doesn't work with only 64bits software. It's very annoying.19:09
gnarfaceyea, i know19:09
gnarface90% of the time 80% of it works19:10
xrogaanI can't find the build log thing for debian19:10
gnarfacefor Steam?19:10
gnarfacei don't think there would be one19:10
gnarfaceoh you're talking about something else19:10
gnarfacewhat build log thing?19:10
xrogaanYou're not following.19:12
xrogaan<xrogaan> could this be solved? http://dpaste.com/35GRG2A19:12
xrogaani386 version mismatch with x6419:12
xrogaanso, maybe it's being held upstream19:13
gnarfaceoh19:14
gnarfacei didn't read your paste.  those are NOT the same versions19:15
gnarfaceit's complaining about a version issue19:15
xrogaanI KNOW19:15
xrogaanMaybe devuan's pipeline broke or something, or maybe debian's maintainer just gave up.19:16
gnarfacewell, debu8 is oldstable and debu7 is ... deprecated19:16
gnarfacewhat version of devuan are you running again?19:16
xrogaanascii, the stable one?19:16
KatolaZxrogaan: we are not forking libsystemd019:16
xrogaanI don't know where to see if debian's packages are proper.19:17
KatolaZthose packages come straight from Debian19:17
KatolaZxrogaan: tracker.debian.org19:17
xrogaanyeah, I get nothing.19:17
gnarfacehttps://pkginfo.devuan.org/cgi-bin/d1pkgweb-query?search=libsystemd&release=ascii19:17
gnarfacexrogaan: check out the versions in this search... it looks like there may be evidence of distro or version-mixing in yours...19:18
KatolaZgnarface: that won't tell much19:18
KatolaZsince pkginfo is currently based on amd64 only19:19
KatolaZ:\19:19
xrogaangnarface: are you high?19:19
gnarfacewell, i note that his paste says it wants the version from ascii-security, but it's trying to install a version that isn't there19:19
gnarfacexrogaan: not yet19:19
gnarfacexrogaan: are you?19:19
KatolaZxrogaan: there is no reason to treat like that somebody who is trying to help you, mate...19:19
gnarface232-25+deb9u8 < where is it getting this from?19:19
KatolaZhttps://tracker.debian.org/pkg/systemd19:19
xrogaanWhat's the difference between "[ascii-security] libsystemd0-232-25+deb9u7" and my "232-25+deb9u7" ?19:19
gnarfacesome security patches19:20
xrogaanit's the i386 version, not the one listed in the devuan tracker thing.19:20
gnarfacewell i guess they might not be the same but they usually are...19:20
gnarfaceusually version discrepancies only happen with the ARM stuff19:20
xrogaanthe way it is, x64 is supposed to be debu8 but it isn't. I want to know why.19:21
xrogaanand this isn't helping: https://packages.debian.org/source/stable/updates/systemd19:22
gnarfaceas far as i can tell from this though, it's NOT supposed to be debu8... that's what i meant by evidence of distro/version mixing19:22
xrogaansmells like somebody fucked up somewhere.19:22
KatolaZxrogaan: again19:22
KatolaZ:\19:22
xrogaanDude, it is supposed to be debu819:22
xrogaan> Package: libsystemd0 (232-25+deb9u8) [security]19:23
KatolaZdude, we are trying to help19:23
KatolaZcan you see it?19:23
KatolaZo_O19:23
gnarfacehmm. well it's true that packages.debian.org does list 232-25+deb9u8 for libsystemd and libsystemd-dev... and that's the one in stretch-security19:23
xrogaanand, well, look: security.debian.org/debian-security/pool/updates/main/s/systemd/libsystemd0_232-25+deb9u8_amd64.deb19:24
gnarfaceit's not a different version on any architecture though19:24
xrogaansurprise, it exists.19:24
KatolaZxrogaan: which mirror are you using?19:24
KatolaZbecause the package is in the repo19:24
KatolaZis in ascii-security19:24
KatolaZin Devuan, I mean19:24
KatolaZhttps://pkgmaster.devuan.org/merged/dists/ascii-security/main/binary-i386/Packages.xz <- xrogaan19:25
KatolaZit's there19:25
xrogaanI should have that> deb http://deb.devuan.org/merged/ ascii-security main non-free contrib19:25
KatolaZxrogaan: which mirror are you using?19:25
KatolaZxrogaan: silly question: apt-get update?19:25
xrogaanyes, update doesn't change19:26
KatolaZxrogaan: apt-cache policy libsystemd019:26
xrogaanI don't know *which* mirror I'm using. It's the round robin.19:26
KatolaZ(please)19:26
KatolaZxrogaan: all the mirrors are updated19:26
KatolaZxrogaan: apt-cache policy libsystemd019:27
xrogaanhttp://dpaste.com/11QH3N019:27
KatolaZthat is simply not possible19:27
KatolaZsince the package is in ascii-security19:27
xrogaanascii-security/main ?!19:28
KatolaZPackage: libsystemd019:28
KatolaZVersion: 232-25+deb9u819:28
KatolaZInstalled-Size: 68819:28
KatolaZMaintainer: Debian systemd Maintainers <pkg-systemd-maintainers@lists.alioth.debian.org>19:28
KatolaZlook at the link I posted above19:28
gnarfacehmm. this paste mentions backports...http://dpaste.com/11QH3N019:28
KatolaZgnarface: it's not in backports19:28
KatolaZit's in ascii-security19:29
KatolaZthe snippet I pasted comes from the Packages.xz in ascii-security19:29
gnarfaceKatolaZ: there is a version in backports though too... if he mixed in some backports versions of stuff, theoretically they could be causing this indirectly, no?19:29
KatolaZand that file is the same across all the mirrors behind deb.devuan.org19:29
xrogaanKatolaZ: I may be hitting a borked mirror, no clue which.19:29
KatolaZxrogaan: that's not possible19:29
xrogaanI don't know how to know.19:29
KatolaZI know19:29
KatolaZall the mirrors are updated19:29
KatolaZand synced19:29
KatolaZjust checked again19:30
KatolaZgnarface: that's a good point19:30
KatolaZxrogaan: maybe you got the package from backports?19:30
xrogaanwhich one?19:30
KatolaZthe one you have currently installed...19:30
gnarfacesome of them, hard to say19:30
gnarfacedpkg -l |grep bpo19:31
KatolaZlibsystemd0....deb9u719:31
gnarface?19:31
xrogaanno19:31
KatolaZbut again, that's not installed according to the paste19:31
KatolaZxrogaan: the package is there19:31
etech3doing a test on a 64 bit base install only19:31
etech3can /etc/systemd be removed?19:31
KatolaZetech3: why should you? o_O19:32
etech3clean machine19:32
KatolaZ-_-19:32
KatolaZyou can only fear what you don't understand...19:32
gnarfaceit might break something obscure.  i forget what, but i removed it once and the lack of some symlink choked some video game for some reason... i forget the details19:32
golinuxarmin: Great find! https://lists.freedesktop.org/archives/systemd-devel/2019-January/041971.html19:33
xrogaanmy installed backports packages: http://sprunge.us/iWnh5J19:33
xrogaangolinux: was posted here 40 minutes ago19:33
gnarfacehostile, accusatory, not-detail-oriented, AND rotating pastebin links?  nah, i'm out.19:34
xrogaanhow do I remove the local list package cache?19:34
KatolaZxrogaan: apt-get install libsystemd0=232-25+deb9u819:34
golinuxxrogaan: Of course.  Repeated for truth19:34
xrogaangnarface: I use dpaste manually, sprunge is when I use pastebinit.19:35
KatolaZxrogaan: ^^^19:35
KatolaZgolinux: I don't see how that is a good or a bad news at all19:36
xrogaanKatolaZ: it obviously do not work :P19:37
golinux<KatolaZ> since pkginfo is currently based on amd64 only.   WTF!19:37
KatolaZgolinux: feel free to amend that :P19:38
KatolaZxrogaan: you obviously know better, so you probably don't need help19:39
KatolaZthe package is there19:39
xrogaanwould removing /var/cache/apt/pkgcache.bin force a proper "update"?19:39
KatolaZthere might be something wrong in your config19:39
KatolaZe.g. a pin19:39
KatolaZagain, the package is *there* in the repo19:39
KatolaZit should appear in apt-cache policy19:39
KatolaZright after an apt-get update19:40
KatolaZif it doesn't, then you have a pin somewhere19:40
xrogaanthere is a /etc/apt/preferences.d/avoid-systemd19:40
xrogaanonly for Package: systemd-sysv19:40
KatolaZbbl19:40
xrogaanI know it's in the repo, we established that.19:42
KatolaZxrogaan: hold on19:42
KatolaZthe amd64 version is deb8u719:42
KatolaZthe i386 version is deb9u819:43
KatolaZthe amd64 version is deb9u719:43
xrogaanapt-mark showhold returns nothing19:43
KatolaZmmmhhh19:43
xrogaanKatolaZ: yes, that's what I have.19:43
xrogaanhttps://packages.debian.org/stretch/amd64/libsystemd0/download < the repo url finishes with `debian-security stretch/updates main'19:44
KatolaZbut it's the same in Debian19:44
xrogaannothing to do with devuan, right? It's not as simple as a misconfigured source.list?19:45
xrogaanI need to go to bed soon :/19:45
KatolaZsleep tight19:46
gnarfacexrogaan: not that you deserve the answer, but the fact of the matter is that while this could have been caused upstream, the vast majority of cases like this come from mixing distros or mixing repo versions.  if you're sure you never did, but you still used backports - think again.19:47
xrogaanwho cares about systemd anyway ( ̄へ ̄)19:47
gnarfaceif you trace carefully all the dependencies linked to those backports you might be able to figure out exactly where the block is, but it might also be easier to just remove them all so you can do a clean update && upgrade19:48
KatolaZxrogaan: it could also be a temporary glitch on the debian cdn side19:49
KatolaZI mean, some of the security mirrors might not be yet updated19:49
KatolaZthis could happen19:50
KatolaZand would be solved by waiting a few hours more, normally19:50
xrogaanThat's what I thought too, I noticed this issue yesterday.19:52
xrogaanIt's not critical anyway.19:53
xrogaan'night and thanks19:53
armingolinux: made my day actually :)20:11
etech3doing a test on a 64 bit base install only21:38
etech3can upstart  be removed?21:39
nemohuh... I don't know much about devuan's package management, but seems like that would complicate packaging services?21:40
etech3just trying to keep a clean machine21:41
nemoetech3: although, wouldn't surprise if everything still uses init.d for compatibility/least surprise21:41
nemoetech3: I guess without knowing the decisions devuan packagers made, if everything in /etc/init is also in /etc/init.d you're probably fine...21:42
nemojust a guess tho21:42
nemoalso would probably have to make sure startup proceeded normally21:42
etech3may do a remove of upstart and see what happens21:42
nemoheh21:43
nemoTIAS21:43
etech3test machine21:43
nemomaybe use a vm ☺21:43
nemoah21:43
nemoetech3: might want to try installing some services after removing it21:43
nemoapache etc21:43
etech3good idea21:43
nemoetech3: I'd suspect if you want hardcore customisability you'd be better off on gentoo rather than devuan21:45
nemoetech3: devuan is still debian and has favoured stability and standard environments for a long time21:45
* nemo uses both personally21:45
nemoetech3: every time I've tried to make significant overrides to debian package list it's always ended up in tears21:45
nemounless some fork like devuan is doing the hard work for me ☺21:46
etech3gonna try and report21:46
etech3if it breaks I get both halves21:47
etech3HEY no toast!!!21:56
etech3remove upstart crud and installed apache2 reboot no problem21:56
etech3so far21:56
etech3gonna add to my install checklist if it pands out21:57
etech3pans21:57
sixwheeledbeast^upstart?21:57
etech3yep21:57
etech3base install only for headless machine21:58
sixwheeledbeast^I don't understand why it would be included if your not using it.22:00
etech3try updatedb ; locate upstart22:00
* gnarface doesn't have it here either22:03
etech3I have a list of extra tools and trying to find which one may pull it22:05
gnarfacei'm curious now too22:05
gnarfaceit's never been the default init in debian.  it was in ubuntu though for a while i think...22:06
etech3started with looking at refracta but that is not installed yet22:06
gnarfaceuse any PPAs?22:06
gnarface3rd party repos?22:06
etech3no22:06
gnarfacehmmm22:06
etech3gonna redo the install and test some more22:09
sixwheeledbeast^I liked Upstart as an init. It's an Ubuntu thing, it's no longer available in the latest Ubuntu or Debian version repos AFAIK.22:10
nemognarface: oh? I'd heard that when debian when systemd, the vote was upstart vs systemd - I kinda assumed you guys went upstart22:10
nemoI have a bunch of stuff in /etc/init on this clean devuan install22:10
gnarfacenemo: no, devuan stayed with sysvinit, which predates both of them22:11
gnarface(was the default in debian wheezy and earlier)22:11
nemo$ ls /etc/init | paste -sd'|'22:11
nemoanacron.conf|irqbalance.conf|networking.conf|network-interface.conf|network-interface-container.conf|network-interface-security.conf|slim.conf|ssh.conf|startpar-bridge.conf|usb-modeswitch-upstart.conf22:11
nemoguess some packages just install stuff there just-in-case...22:11
nemoor maybe the init.d scripts read those configs? *shrug*  I thought /etc/init was an upstart thing tho22:12
gnarfacenemo: there might be some overlap between sysvinit and upstart... i seem to recall someone complaining about maybe a compatibility layer of scripts shared between the two?22:12
_abc_Hi fsmithred. Is this mostly relevant to making things like refracta2usb manually? The boot loader part, at least? https://www.ab9il.net/linux/grub2-iso-booting.html22:12
nemognarface: oh for sure there's overlap.  upstart kinda was an augmentation of sysvinit.22:12
sixwheeledbeast^I imagine ChromeOS is still using Upstart? but Canconial dropped it to follow Debian.22:13
nemognarface: one nice thing about it. progressive enhancement and things still working in existing locations.  I'm really surprised debian went w/ systemd22:13
nemosixwheeledbeast^: chromeos is gentoo, so who knows. gentoo can do either one.22:13
nemosixwheeledbeast^: you'd think systemd would appeal to google's android-y soul22:13
_abc_nemo: why surprised? debian has become a clusterf* supplying ubuntu's every need22:13
gnarfacenemo: you can guarantee money slid under the table and heads rolled to make it happen22:13
etech3That is what I'm thinking22:14
nemoin that case redhat was probably supplying a good bit of the cash/pressure ☺22:14
gnarfacethey do have motive and opportunity...22:14
nemodon't think ubuntu had as much invested22:14
_abc_No need for real cash. Just stopped saying "we'll destroy you"22:14
gnarfacewell, redhat can donate monetarily significant development man-hours too22:14
gnarfacethat can pass for money in an open source context22:15
_abc_Now it's "resistance is futile, you will be poetteringed"22:15
_abc_I would very much like rh not to donate any more poisoned apples22:15
gnarfacei find myself just wondering how anyone could really be surprised RedHat is evil22:15
_abc_Which means almost all they did. There's a long list and about 3 things in it are valuable for everyone, not just themselves and their own job security.22:15
etech3rh=ibm22:16
gnarfaceindeed.22:16
_abc_https://community.redhat.com/software/#operating-system read this and weep. Identify the 5 things they made which are actually nice and do not break anything22:17
gnarfacehehe, i used to like ntsysv22:17
gnarfacethen i found out it's just symlinks22:17
_abc_Apropos does anyone know anybody who actually uses WIMAX? It's a huuuge part of the kernel build and I've yet to see it used anywhere, once.22:18
sixwheeledbeast^Those are contributions lists not complete projects22:20
CriggieNever - its more of a long-haul wireless ethernet right ?22:20
gnarface_abc_: i don't know.  802.16 has been around for a while and i was really anxious to see ISPs start using it for a long time, but one by one everyone i heard talking about it just vanished, closed up shop, or got bought out by someone bigger.  no deployments so far ever, that i've heard of personally.   :(22:20
nemo_abc_: well in all fairness, the first thing they made that was actually nice was rpm22:21
gnarfaceCriggie: yea it's supposed to have a range of 30 miles without satellite22:21
nemo_abc_: prior to that, it was kind of a PITA on linux22:21
nemo_abc_: even if I do have a softspot for slack/gentoo style build-your-own22:22
nemo_abc_: getting beyond that to 5 though? hm...22:22
gnarfaceCriggie: (it would evaporate the primary monopoly foundations in huge swaths of the country here for cellular and landlines so it's not hard to imagine why nobody has managed to make it to market with one)22:23
sixwheeledbeast^WIMAX will have to die in the future over 4 and 5G mobile.22:23
nemognarface: SpaceX is kinda targetting that no?22:24
gnarfacewe should probably be having this whole conversation in #debianfork22:24
gnarfacenemo: no idea actually22:24
nemohttps://en.wikipedia.org/wiki/Starlink_(satellite_constellation)22:24
nemognarface: their goal is global broadband22:24
gnarfacenemo: oh, yea, but that's not 802.16 i don't think. i think that's just basically regular old satellite internet.22:25
_abc_Using 802.16 ? SpaceX?22:25
gnarfacewe have regular satellite internet but mostly just small players that only serve rural areas22:26
_abc_Because frankly I am fed up editing kernel configs and rebuilding. WIMAX stack is all modules and hefty.22:26
_abc_gnarface: but not WIMAX.22:26
gnarface_abc_: yea, you don't need WIMAX22:26
gnarfaceyou can leave that out of your build22:26
gnarfaceyou can probably leave out the 10GB ethernet drivers too :-p22:26
_abc_Not really.22:27
_abc_I sometimes need that22:27
_abc_Rarely22:27
gnarfacereally?  lucky you22:27
_abc_Not my gear, testing.22:27
_abc_And it's really a pain to get everything together to test a unit.22:27
_abc_GBICs etc22:27
_abc_It's certainly not lucky or pleasant or anyhting.22:28
gnarfacewell, that's probably related to the reason why the kernel has drivers for everything, even stuff almost nobody uses22:28
_abc_10GB ethernet on a modest laptop using external adapter is sort of like having a train carriage attached to motorbike22:28
gnarfaceheh, i bet22:28
Criggiegnarface: 5G has a range of fuckall, so its for dense urban only.22:30
nemognarface: traditional satellite internet uses high flying geosync satellites and is very high latency22:30
gnarfaceyea, i know.  to both of those statements22:31
nemognarface: the spacex constellation is closer to the orion satellite phone, only with tons more satellites22:31
nemo'k22:31
nemognarface: just saying you'd get a lot better performance than "regular old satellite internet"22:31
nemodownside is the satellites have to be constantly replaced22:31
gnarfaceoh, i see.  but 802.16 could have really helped a lot of last-mile coverage at lower latency.  and it's been basically sitting on the shelf for a conspicuous amount of time...22:31
gnarfacei doubt even their fancy new satellites could improve on it22:32
nemo"However, SpaceX ultimately decided to keep the satellites at a relatively low 550 kilometers (340 mi), due to concerns about the space environment."22:32
nemognarface: only 550km each way, and probably fewer router hops22:32
Criggiegeometry says 550 km if its right overhead22:33
Criggiehow many satellites are there ?22:33
nemoCriggie: a LOT ☺22:33
DonkeyHoteiin 2009 i picked up a municipal 802.16 signal in downtown san antonio, but i didn't use it because i already had ev-do service i already paid for22:33
DonkeyHoteii mean 2010, sorry22:33
nemoCriggie: "Initial plans as of January 2015 were for the constellation to be made up of approximately 4000 cross-linked[37] satellites"22:33
Criggiemmm ok if it was ~20 then you'd have low prob of a sat overhead, but if its 200 there's a lot more chance22:33
nemosooo 1100km means speed of light to/from satellite would be 4ms of latency22:35
nemoplus whatever relaying would be needed22:35
CriggieI got more than that between home and my upstream gateway22:38
nemoCriggie: by comparison traditional satellite internet is more like 250ms of latency22:39
gnarfacewell i'll be impressed if it works that well22:39
gnarfacei'm not buying 4ms until i see it with my own eyes22:39
nemognarface: I have no idea if spacex is claiming 4ms - after all there's the relaying.  but they have a massive advantage over "regular old" ☺22:40
gnarface~250-300ms is more along the lines of what i'd expect from a satellite22:40
nemognarface: yeah. but geosync is soooo much further away. these satellites are basically flying through the earth's atmosphere, that's why they need to be constantly replaced22:40
gnarfacehmm. yea, could work i guess.  in theory22:40
gnarfacei figured they'd have to be low enough to be aircraft actually though.  like those balloon things Google (was it Google?) wanted to make22:41
gnarfacebut i am not a rocket scientist22:42
CriggieLoon22:42
gnarfaceyea, that's the one, Project Loon22:42
gnarfacethey were just gonna make some dirigible drones and put wifi repeaters in them22:42
gnarfacei think i heard they're actually doing it somewhere, too...22:43
gnarfacenot around here though22:43
nemognarface: I got the feeling google's wifi projects were just there to put pressure on existing network providers23:00
nemognarface: they sure didn't put much effort into them23:00
nemospacex has no such motivations, at least at present23:00
nemognarface: kinda reminds me of what google did to mozilla with initial webm/ogg23:01
gnarfaceyou might be right.  google fiber seemed to evaporate pretty quickly too23:01
nemonegotiate better mp4 licensing for youtube23:01
nemothankfully AV1 seems to finally be getting some traction23:01
nemognarface: worst one could say about the spacex project is that it might be similar to other musk things. just a way to get a bit more investor capital23:02
se7enI came here last month speaking about issues with my /usr directory filling up to fast. The temporary solution was to simpily uninstal supertuxkart23:56
se7enNow it's filled up again23:56
Springlebutthi. i just installed ascii from dvd iso + network mirror over ethernet connection. now i need to get my tp-link n150 wireless dongle (TL-WN722N) working. after some googling i enabled non-free repo and installed firmware-atheros package. but the device still does not work. this is a popular wifi interface im sure devuan people use it. any advice thanks23:57

Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!