12[00:08:43] <mrjpaxton> Question: How would I report an out of date Debian package? (I'm also asking for Unstable in #debian-next, but haven't gotten a response back yet.)
13[00:08:43] *** Quits: teclo- (42@replaced-ip) (Quit: Lost terminal)
23[00:13:47] <mrjpaxton> Okay. Yeah, it's for telegram-desktop. There's quite a bit of features that are unsupported. Thing is, I've tried out the Flatpak version of it, but the problem is that it freezes when opening a file, so I've filed a bug on Github for that one, too.
333[03:04:48] <rant> kiwi_36: and do you have other filesystems like /var or /usr?
334[03:06:19] <rant> kiwi_36: what is the size of this filesystem?
335[03:06:39] <kiwi_36> it says /usr is 11GB, and /var is 27GB
336[03:06:40] <kiwi_36> even though / is 12GB
337[03:07:10] <kiwi_36> should be 12GB
338[03:07:32] <kiwi_36> rant: oh let me restate that, i think /var is actually 5G
339[03:07:40] <rant> I am not sure if you answered my questions or not
340[03:07:52] <kiwi_36> i ran it a second time with "sudo du -sh /var", and that gave me 27G
341[03:07:53] <rant> I am asking you if everything is all on the rootfs or if you have seperate filesystems
342[03:08:40] <kiwi_36> rant: everything is in vda1 (part /)
343[03:08:54] <rant> normally I'd recommend ncdu or baobab but I can't do that in this case cause your fs is already full and you can't install anything
344[03:08:54] <kiwi_36> (not totally sure what you're asking)
346[03:09:26] <rant> I am trying to gather the nature of your situation..
347[03:09:42] <kiwi_36> rant: yeah i get "no space left on device" when i try to install ncdu
348[03:09:45] <rant> it is possible to make other filesystems and mount them onto the rootfs
349[03:10:08] <wuseman> Maybe you can wipe /home/user/.cache or /tmp to get some freespace
350[03:10:12] <rant> if for example you had a vda2 mounted on /var then the size of /var is irrelevant to the rootfs being full cause /var isn't part of the rootfs
351[03:10:21] <themill> kiwi_36: do you have a separate /home too there?
352[03:10:40] <rant> we cant know what to investigate if we don't know what all is on the rootfs and what its overall size is
353[03:10:49] <themill> (a pastebin of "df -h" would be useful to see)
354[03:11:17] <rant> if "everything" is on rootfs and its only 12GB you quite simply need to redo the whole system.. cause you just simply did not allocate enough space
355[03:11:25] <kiwi_36> i have a "debian" user in /home
356[03:11:29] <kiwi_36> 3.2G
357[03:11:52] <themill> kiwi_36: no-one is asking about du, it would be better to see "df -h"
370[03:17:14] <rant> well that changes things a bit.. a desktop install would take 3-6GB and need twice that just to have enough space to do a dist-upgrade. but for a basic server type system or something for a specific purpose you can probably get away with a 12GB rootfs
372[03:18:16] <rant> kiwi_36: we do still have problems with the info you've provided, /usr cannot be 11GB and /var 5GB if they are both on the rootfs which is only 12GB
373[03:18:16] *** Quits: CaptainN (zelda@replaced-ip) (Quit: I have to pee!)
382[03:20:29] <kiwi_36> does du -sh give directory size?
383[03:20:40] <rant> yeah, I'm not familiar with what exactly openstack is.. there are so damn many similar technologies now I'm not very familiar with them
384[03:20:51] <rant> but does seem like you're confusing output
394[03:25:01] <Church-> So question, do I need to use apt-daily or can I just set a cron/timer to call unattended-upgrades at a certain time? And secondally is there an easy way to validate it's completed running?
395[03:25:53] <rant> kiwi_36: do you have the nc command? netcat?
396[03:26:21] <kiwi_36> rant: not installed unfortunately
406[03:27:57] <rant> for i in /*/; do sudo du -sh "$i"; done
407[03:28:08] <rant> paste that on paste.debian.net
408[03:28:40] <rant> if you have to type the password each time you run sudo, you'll need to sudo -v first to avoide typing your password on each iteration of the loop
432[03:34:36] *** Quits: frgo (~frgo@replaced-ip) (Remote host closed the connection)
433[03:35:05] <rant> sounds to me like this small chat application is buggy or misconfigured cause no "small application" should've ran away and filled a 12GB volume
434[03:36:04] <kiwi_36> yeah i properly need to comb through each directory and see what's taking up all this space
435[03:36:35] <rant> you got about 2.5GB in /home/ unaccounted for and /var is probably 4-5GB larger than it should be
436[03:37:18] <rant> I'd check /var/cache, /var/cache/apt/, and /var/log/ first
437[03:37:29] <rant> downloaded packages sit in /var/cache/apt/archives
438[03:37:51] <rant> they aren't needed and can be removed manually or with an apt clean or apt-get clean
444[03:38:40] <rant> I have a full desktop install and my entire /var is only 1.7GB
445[03:39:01] <rant> yes you have a buggy or misconfigured piece of software that filled your system with logs it would seem
446[03:39:15] <kiwi_36> i remember now, i found this last week, totally forgot about /var/log
447[03:39:18] <abrotman> or an actual problem being logged
448[03:39:24] <kiwi_36> can i remove /var/log?
449[03:39:28] <kiwi_36> or clear it?
450[03:39:30] <rant> abrotman: out of space on / :D
451[03:40:10] <rant> kiwi_36: you /can/ but as abrotman says, you might want to figure out what the problem was that something felt the need to log 5GB worth of info on
452[03:40:28] <rant> to do that, you gotta have a look at the logs
453[03:41:21] *** Quits: datastream (~datastrea@replaced-ip) (Remote host closed the connection)
454[03:41:45] *** Quits: behanw (uid110099@replaced-ip) (Quit: Connection closed for inactivity)
573[04:58:33] <rant> jim: my buffer shows you were conversating around this time yesterday and were told that they are a sysadmin on east coast, and would most likely be active during normal business hours.. this however is not a social channel, if you have an issue you can just ask a question and keep personal contact in personal messages or social channels
591[05:07:10] <jim> rant, ok. I am given to understand that some sound card hardware has those 3.5mm sockets that flip a bit somewhere when something's connected to them... I'm looking to find out how to read those bits
594[05:08:08] <rant> jim: well the jack is physically a switch.. the part that connects to the tip of the jack is pushed back when its inserted and when that happens it moves off one contact onto another
604[05:10:33] <rant> cause on some machines jacks are also multi-function.. my laptop only has one.. and its both input and output.. but doesn't switch between the two, it just supports jacks with 3 and 4 contacts
605[05:10:55] <rant> i.e. can't be just a microphone port.. has to be a headset with mic
606[05:12:18] <jim> so, the #alsa folks have a script that collects data about the system it's run on, including those bits... when I present the output of that script on a pastebin, some folks were able to read that listing and determine where I had my headphone plug connected
607[05:12:32] <jim> so it's not just a theory, it's fact
611[05:15:19] <rant> jim: bottom line for here is, this is not a debian support question.. its kernel/alsa development related.. if you have a legit concern, try lkml, or alsa channels
612[05:15:36] <jim> SerajewelKS has demonstrated pretty extensive knowledge about the sound driver, and has been helping me get through this
613[05:15:46] *** eb0t_ is now known as eb0t
614[05:16:13] <rant> yeah well, you can't use this channel for that.. if someone /wanted/ to make themselves available to milk their knowledge they'd give you a means to do so..
618[05:16:43] <rant> for more in depth development type things you serve yourself and everyone else best if you use a development channel or mailing list
619[05:16:55] <rant> more detailed stuff is much better done on mailing lists anyhow
620[05:16:56] <jim> you were not part of that conversation, so I guess you don't have that information
625[05:17:40] <rant> and the channel /is/ pubilically logged if need be
626[05:17:47] <rant> dpkg, logs?
627[05:17:47] <dpkg> Logs are what you should always check when you have a problem, they live in /var/log/ . For #debian logs, ask me about <irclog>.
628[05:17:55] <rant> dpkg, irclog?
629[05:17:55] <dpkg> from memory, irclog is #debian on <freenode> is logged at replaced-url
630[05:18:51] <rant> thats my point, things where anyone would have to have seen old convo and use extensive knowledge is beyond the scope of this channel
655[05:37:49] <rant> yes, I know you view it that way, and I'm trying to tell you its not helping you or anyone else. I've been here decades and never ignored anyone, I'm not gonna start now. Folks sit here expecting when they see activity in this channel for it to be a question they can weigh in on.. so having things outside the scope not only lessens the ability to support more apropo questions, but its not getting you the
656[05:37:55] <rant> help you need and in a timely fashion
657[05:38:08] <rant> mailing lists are surprisingly quick in response and reach a wider more advanced audience..
660[05:40:40] <rant> your line of questioning is hardware/kernel/alsa related, it is not debian user related in even the most remote sense as you asked no question related to debian use of such things
686[05:59:16] <petn-randall> rant: If they run Debian, users are free to ask whatever they want about their system in this channel. This includes the Linux kernel and alsa stack provided by Debian.
690[06:00:59] <rant> I am not questioning their freedom to do so, I am saying that trying to milk individual users via this channel even when they're not active and not asking specific questions but wanting to carry on in depth development debates are better suited for other means
693[06:03:40] <jim> what is it you're trying to protect?
694[06:03:45] <rant> I'm not trying to argue here.. I'm just trying to tell you how to get better served without developing potentially disruptive habits
697[06:04:22] <rant> you could've sent a private message for that, and you can contact developers who can carry on the kinds of detailed conversations you are looking for
704[06:06:10] <jim> I'd advance that (a) my habits are my business, and (b) habits that have not formed yet are vaporware. having said that, thanks for suggesting ways that would better serve me
705[06:06:13] <nefsation> hi there
706[06:06:23] <jim> ZaZaGX, hi
707[06:06:36] <ZaZaGX> i just got the debian 10 to work on my laptop
708[06:06:41] <jim> nefsation, hi too
709[06:06:50] <ZaZaGX> it took like gave up like 3 times lol
710[06:06:53] <jim> ZaZaGX, works good?
711[06:07:01] <nefsation> im having some trouble after upgrading my home server to buster
712[06:07:25] <nefsation> ill be really happy if someone points me in the right direction
713[06:07:35] <jim> nefsation, ok, tell us (in some detail) what the trouble is
745[06:16:45] <jim> nefsation, so you see that's a conflict... debian won't let you downgrade while owncloud won't let you upgrade... I'd say that (while debian has many positive attributes) debian is presently not a good choice for this particular application
746[06:18:06] <jim> it's unfortunate that you don't have a complete filesystem backup, maybe if you did, you could restore from it
753[06:21:06] <jim> I'll also note in passing that php has a history of not being developed in a stable way... also a history of invalidating millions of lines of code on a periodic basis
754[06:21:28] <nefsation_> so im not the first
755[06:21:33] <nefsation_> hahaha
756[06:21:37] <jim> no, of course not
757[06:23:06] <nefsation_> ill laugh because irony
758[06:23:33] <jim> (in comparison, other languages are not so unstable, including in their development... in particular an old language, TCL, has developers that are extremely careful to not invalidate any code
759[06:23:51] <nefsation_> mmmm
760[06:24:03] <jim> and they've been very successful at it. so php and tgheir developers don't have an excuse for this
761[06:24:17] <nefsation_> is there any other option? like other private cloud solution
762[06:24:17] <Deihmos> is there a way to get the network connection speed on a headless debian server?
763[06:24:31] *** Quits: itamarst (uid165457@replaced-ip) (Quit: Connection closed for inactivity)
764[06:24:43] <Deihmos> on windows i can see if i cam connected 100 Mbps or 1Gbps. Is there a way to see it on debian?
776[06:28:22] <rant> <nefsation_> doing this will lead people to think you are quoting them
777[06:28:39] <nefsation_> rant sorry
778[06:28:51] <nefsation_> im not used to irc yet
779[06:29:02] *** debhelper sets mode: +l 1500
780[06:29:13] <rant> nefsation_: no worries, just letting you know.. highlighting is best done with the nick at the beginning of the line without the < and with a single delimiter like a , or :
791[06:30:46] <rant> nefsation_: most clients do it automatically with tab completion, I just type ne and press tab and it does it automatically for me
792[06:31:23] <nefsation_> im using hexchat for the first time in a long time
793[06:31:32] <BazookaTooth> :)
794[06:32:33] <nefsation_> well, anyone knows a good way to backup files from db? mysql
797[06:33:12] <Deihmos> anyone did a backup using acronis?
798[06:33:38] <Deihmos> i normally use clonezilla but not being able to mount an image isn't good
799[06:33:41] <rant> I'm no SQL/DB expert but I do know that you should always export the files first as I've worked on situations where people just copied the files then tried using them on a different db version and it didnt work
800[06:33:57] <rant> exporting makes files capable of being properly imported on other db versions
807[06:37:48] <rant> Deihmos: yeah, not seeing anything like that in debian.. you could make this relevant by asking something like "How can I foo in Debian so I can bar and baz?"
808[06:37:50] <nefsation_> rant, 0k ill read about it. Hence im the only admin here i dont think it would be a problem. less than 1gb or so
811[06:39:03] <rant> nefsation_: yeah well I'm strongly advising this.. there many other reasons I cant accurately articulate why you want to export the DB and import it back rather than just backup the files
812[06:39:23] <nefsation_> well
813[06:39:31] <rant> I can't because I'm not that knowledgeable about DB as I avoid them whenever possible, I prefer to use sqlite as its simpler :P
814[06:39:36] <nefsation_> actually i just want the files back
815[06:40:03] <nefsation_> saw phpmyadmin for that purpose
829[06:48:16] *** Joins: Antoine (~Antoine@replaced-ip)
830[06:48:56] <jim> nefsation_, some dbs have a command that reads everything in the database and produces an sql listing, which when run on an empty database, loads the database with the earlier content
831[06:49:00] <kreyren> nwm it's debian-next issue with openrc
832[06:49:03] <alkisg> kreyren: systemd-logind?
833[06:49:26] <kreyren> alkisg, nah systemd-pam apparently based on replaced-url
834[06:49:31] <nefsation_> jim, ill look into it
835[06:49:37] *** Joins: Abdullah (~ak@replaced-ip)
854[06:55:27] <kreyren> alkisg, gentoo developers are not responsible for packaging you build it from source
855[06:55:29] <alkisg> It's not just one runscript. How about information like "who is logged in currently? is it safe to unmount x user's sshfs home dir?"
856[06:56:10] <alkisg> Anyways, it would take time to convince you that supporting various init systems takes time :)
865[07:00:07] <kreyren> alkisg, > "who is logged in currently? < -> afaik you have session manager for that, `is it safe to unmount x user's sshfs home dir?"` - just use umount lol?
868[07:00:53] <alkisg> kreyren: you're missing the point, I mean "systemd has APIs that some programs need and may decide to depend on them"; and I'll stop at that.
869[07:01:06] <kreyren> w/e
870[07:01:33] <kreyren> just make quality code and keep it open-source and i'm gut
883[07:14:44] <jim> kreyren, if you ever feel like you'd like a detailed overview on systemd, man systemd happens to be a good read (I was surprised by this)
1025[08:45:26] <jim> kreyren, not sure, but people might feel devuan is different from debian, in ways they're not aware of, so answers they might give would be guesses (and so not helpful to you)
1029[08:47:31] * jim is not even sure how similar they are (or if they're exactly the same, differing only in which sets of packages are installed from debian's archive)
1037[08:49:17] *** Joins: kila (~kila@replaced-ip)
1038[08:49:38] *** Quits: tsujp_ (~tsujp@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
1039[08:49:58] <alkisg> Can anyone manage mksquashfs -regex to exclude things under ^/proc/.*$ ? This works, but it also excludes /other/dir/proc/.* ...
1046[08:53:05] <rant> devaun is different from debian in ways us Americans are all too aware of.. it offers init freedom very much the same way the US offers freedom.. you are free to use any init system you want, except systemd :P
1107[09:21:01] <rant> or other such things.. usage polls and testimonials aren't really within the scope of this channel
1108[09:21:46] <rant> you can either google and read reviews and comparisons online, or perhaps go to a channel specific to such software and ask questions about it
1161[09:32:18] <rant> dpkg, why isn't virtualbox in testing
1162[09:32:18] <dpkg> virtualbox is not in testing for the reasons listed in replaced-url
1163[09:33:59] <gggg> "I would like to see it in Debian, but since people working for Oracle might risk to get punished for not following the Oracle policy, I think we are not sure we can continue giving a CVE free package for Stable Releases."
1164[09:34:15] *** Joins: lane (~laurent@replaced-ip)
1165[09:34:16] *** Quits: Treylis (~quassel@replaced-ip) (Remote host closed the connection)
1267[10:24:32] <ZaZaGX> i heard good things about Mallvad vpn, how is it?
1268[10:25:06] <Makaveli7> no complaints so far. it's fast, it has a kill switch and netflix works
1269[10:25:16] <Makaveli7> I'm using it via wireguard
1270[10:25:27] <BCMM> k_sze: what do you mean? like, compiling the interpreter with different cflags?
1271[10:25:48] <k_sze> BCMM, different options passed to the `configure` script.
1272[10:26:50] <k_sze> Python's `configure` script has a `--with-optimizations` option that will perform profile-guided optimizations during the build.
1274[10:27:35] <ZaZaGX> i just googled wireguard. still not sure what it is
1275[10:27:37] <EoflaOE> Makaveli7: Is Mallvad vpn paid or free
1276[10:28:30] <Makaveli7> it's paid
1277[10:28:43] <colo-work> "a kill switch"?
1278[10:28:52] <k_sze> BCMM: I tried that on Raspbian Buster (which is largely Debian Buster for armv7, for the Raspberry Pi series of single board computers): replaced-url
1279[10:28:54] <EoflaOE> Makaveli7: OK. A subscription?
1281[10:29:31] <ZaZaGX> i use ProtonVPN. it offers free vpn but the speed is basic
1282[10:29:42] <ZaZaGX> i decided to pay for it
1283[10:29:56] <BCMM> k_sze: how did you do that? did you just download python sources from upstream, or did you start by modifying debian's source package?
1284[10:29:59] <Makaveli7> yes you can get subscription
1285[10:30:07] <EoflaOE> ZaZaGX: Nice. I use that, but not able to pay. It delivered enough speed.
1286[10:30:09] <BCMM> (or raspbian's source package, if that's different in any way)
1287[10:30:10] <k_sze> python sources from upstream
1288[10:30:10] <Makaveli7> I got it for a month to try
1289[10:30:18] <EoflaOE> Makaveli7: Nice.
1290[10:30:21] <k_sze> directly from python.org
1291[10:30:30] <BCMM> k_sze: any particular reason not to try using your distro's source package?
1292[10:30:56] <BCMM> k_sze: the objective is to change configure options, not to get a newer version, right?
1293[10:30:57] <ZaZaGX> i got the bundle. Protonmail and ProntonVPN. instead of paying 15 bucks. it cost 12 bucks a month
1294[10:31:26] *** Quits: v01d4lph4 (~v01d4lph4@replaced-ip) (Remote host closed the connection)
1295[10:31:46] <Makaveli7> that's nice. I heard about protonvpn
1301[10:33:24] <BCMM> k_sze: that way, if there's some weird tweak needed to make it build on debian, you can get a copy that's already had that done
1302[10:33:25] <ZaZaGX> so wireguard has stronger security?
1303[10:34:58] <BCMM> ZaZaGX: wireguard is relatively new, so is considered to be a bit less audited than some other solutions. however, the protocol is also much simpler than some other vpns, and it uses existing encryption standards
1304[10:35:05] <Makaveli7> I don't think it's about security. it's just faster and has lesser code than openvpn
1305[10:35:11] <BCMM> so there's kind of not a lot of places that a security bug could hide
1324[10:39:31] <BCMM> k_sze: any package in debian's own repository has to exist as a "source" package. you can't just upload a build that some developer managed to compile once on their own system
1325[10:40:16] <rant> that includes binary data.. you have to provide the means for the generation of the data
1327[10:40:47] <BCMM> k_sze: debian binary packages are automatically built by debian's build servers. this is how raspbian is able to exist - by almost completely automatically compiling debian source packages for the Pi's specific arm microarchitechture
1328[10:41:05] <g0zzy> Can anyone think of anything that boots UEFI and has memtest on it? I want to test this RAM before installing Buster.
1329[10:41:27] <rant> g0zzy: memtest is built into the linux kernel, has been for a long time
1335[10:42:04] <ZaZaGX> would debian bullseye testing be a good choice?
1336[10:42:11] <BCMM> k_sze: so if you want to modify a package on your system, your best starting point is probably the source package, because it can automatically be built in to a working binary. if you use a source package properly, you don't have to engage in trial-and-error to work out how to make the package build on your system
1337[10:42:20] <BCMM> because the source package is already set up to build with a single command
1338[10:42:23] <g0zzy> Ah great
1339[10:42:41] <kreyren> Why does my debian system takes +- 60 sec to trigger any command ?
1403[11:17:12] <dpkg> well, dont break debian is replaced-url
1404[11:17:30] <kreyren> ratrace, that's apt specific not devuan issue
1405[11:17:50] <kreyren> and my debian is used to beeing broken! i have the power
1406[11:18:06] *** Quits: encod3 (~encod3@replaced-ip) (Remote host closed the connection)
1407[11:18:25] <ratrace> kreyren: but using devuan repos is not supported here
1408[11:18:56] <g0zzy> alkisg: where did you see any docs? I've just been doing memtest=9 (It didn't work)
1409[11:19:01] *** debhelper sets mode: +l 1531
1410[11:19:35] <alkisg> g0zzy: I didn't find any docs :D then I read the source, memtest.c, and I tried memtest=1000000, which delayed a lot, so then I tried plain "memtest", which did 17 passes
1411[11:19:46] <alkisg> g0zzy: dmesg | grep memtest => tells you if it ran or not
1412[11:19:50] <kreyren> ratrace, the question was about managing different repos on apt
1437[11:25:20] <g0zzy> alkisg: So what about combining what you tried with that?
1438[11:25:21] <BenNZ> ratrace: i had the same issue updating stable to buster recently
1439[11:25:38] <BenNZ> ratrace: altho i was using debian repos
1440[11:26:00] <alkisg> g0zzy: I don't understand what you're asking; I've read kernel parameters before and I did make memtest work properly, it just was not as useful as I expected; what is your question?
1470[11:31:26] <alkisg> g0zzy: afaik it was never there, as it's proprietary. You can download the free commercial version of memtest.efi from replaced-url
1471[11:31:48] <alkisg> The open source version is a much older one, that doesn't support efi
1496[11:36:08] <ratrace> BenNZ: can you pastebin your sources.list?
1497[11:36:22] <ayekat> files shouldn't just randomly disappear in /bin - something is definitely weird
1498[11:36:27] <ayekat> maybe a botched usrmerge install
1499[11:36:32] <mpodien> Is it possible to continue reading with less where you left off after a reboot? is this possible using dtach? do socket files survive a restart?
1500[11:37:10] <BenNZ> ratrace: the issue is because the server uses buster backports in the release file instead of testing , not much i can do about it except use buster in my sources
1501[11:37:16] <Kocane69> alkisg: what does debsums -s do?
1502[11:37:55] <alkisg> Kocane69: it checks all the files in your system (or in the specified packages) against their initial md5sums, so it's a good indication of debian installations that are broken due to bad disks
1503[11:38:02] <alkisg> E.g. it would report /bin/systemctl missing
1504[11:38:22] <alkisg> So if you see that it will report tens of broken files, it's a disk issue; etc
1507[11:39:12] <Kocane69> alkisg alright.. right now it's just listing a ton of "changed file"
1508[11:39:13] <kreyren> fixed the issue by renaming the release since Release file in testing-backports is using bullseye instead of testing.. suck it ratrace!
1509[11:39:37] <alkisg> Kocane69: did you pass the -s parameter? pastebin a few lines...
1510[11:39:50] <ratrace> kreyren: forgot your meds today?
1511[11:40:07] <kreyren> ratrace, fetching my drugs atm
1512[11:40:12] <alkisg> Kocane69: debsums -s, when ran as root, should ideally display nothing
1513[11:40:15] <Kocane69> alkisg: yeah I did. It stops with missing files in some kernel drivers, and in the end "debsums: missing file /bin/systemctl (from systemd package"
1514[11:40:20] <Kocane69> Oh..
1515[11:40:40] <alkisg> Kocane69: pastebin the output, so that we see how bad it is
1516[11:40:46] <ratrace> kreyren: you'll still break your debian if you try using devuan packages
1526[11:43:25] <ratrace> kreyren: why would you even want to do that btw, what do those packages offer that you can't do in debian, or can't use devuan straight?
1527[11:43:43] <kreyren> ratrace, yep since debian is unable to handle my openrc configuration
1528[11:44:01] <kreyren> since i need consolekit2 and some maintained version of openrc
1529[11:44:20] <kreyren> but also i need fast updated by debian for rest of the packages
1530[11:44:30] <kreyren> *fast updates of packages on debian
1531[11:44:36] <ratrace> so then why don't you use devuan straight?
1552[11:52:26] <ayekat> calling people assholes if they tell you that you're asking questions in the wrong place seems more asshole-y to me
1553[11:52:31] <alkisg> Kocane69: it seems like your disk has started being corrupted since 4.16... I'd suggest an fsck while the disk isn't in use, then reinstallation of all broken packages, and when you finally have debsums -s clean and fsck clean, monitor it for a few days in case it errors again,
1554[11:52:31] <alkisg> btw, do you see any disk or ATA errors in dmesg?
1564[11:55:41] <afidegnum> hello, using debian with i3, when i ran start_navicat is not able to run. $PATH is not exported, part of the script is not run even with sudo. what can i do? here is the installation script replaced-url
1566[11:56:24] <ayekat> it's like asking for ubuntu help - sure, it uses mostly the same tools, but who knows how the default config differs from debian?
1568[11:59:00] <ratrace> kreyren: my argument was that your problem stems from trying to use non-standard repos; you wouldn't have it if you used standard, supported repos; that's all
1569[11:59:57] *** Quits: dutchfish (~wil@replaced-ip) (Quit: Splash, leaving the bowl)
1570[12:00:21] <kreyren> fair enough, but you should still provide the help needed with it if end-user wants to use this configuration instead of screaming 'not supported'..
1571[12:00:44] <ratrace> kreyren: and if you think that's being an ass, perhaps you'd like to invite entire Freenode to seek support in #debian? it's all linux anyway, amirite? ;)
1583[12:05:54] <ayekat> kreyren: debian is the creator of apt the *software*, but devuan is the maintainer of apt the *package* - the debian maintainers have no control over how devuan has packaged or configured apt on devuan
1584[12:06:00] *** Quits: nexgen (~nexgen@replaced-ip) (Remote host closed the connection)
1596[12:09:40] <ayekat> kreyren: no, it still doesn't make it relevant for both - and using debian's apt doesn't mean there isn't still devuan-specific configuration lying around on your system
1597[12:10:22] <ayekat> if you know the specific differences, and you ask an apt-related question that is not a support question, it's fine to ask here
1598[12:10:23] <kreyren> ayekat, in that case it would made it relevant for both devuan and debian if said system is using configuration from debian and devuan
1604[12:11:23] <ayekat> kreyren: no, because debian people don't know anything about devuan configuration - so if you mix'n'match configs, only devuan people will know (if any at all)
1614[12:14:17] <kreyren> like i understand that you dont like frankendebian users in general since it makes it difficult to help them, but i dont see a reason why woudn't you provide said help so that i would be able to use debian's part of my franken for my usecase..
1615[12:14:45] <kreyren> ratrace, afaik only init is different and i'm currently running openrc on debian from debian which is pain to use..
1616[12:14:54] <kreyren> rest is pretty much same or very simmilar
1617[12:15:00] <ayekat> kreyren: oh that's simple: because even on debian, mixing distro releases is unsupported
1618[12:16:04] <ayekat> if you mix stretch repos with buster repos, expect things to break (someone linked to the "don't break debian" article) - the same applies to mixing different releases from different *distros*
1619[12:16:23] <kreyren> ayekat, but you have testing debian which literally needs pinning from stable, experimental etc.. to work around present issues + it's anti-gnu since gnu encourages to support every configuration of system
1620[12:16:44] <Kocane69> alkisg: i'm not seeing any disk errors regarding to my system SSD i don't think.. however I am seeing some on one of the disks in my zfs pool
1621[12:17:02] <kreyren> ayekat, breakage is expected i can work around that usually and when i can't i expect to ask here or on #debian-next to get more info to fix said issue
1622[12:17:13] <Kocane69> alkisg: is there any way I can scan the disk while it's in use?
1623[12:17:31] <alkisg> Kocane69: dunno, I'm not using zfs
1624[12:17:44] <kreyren> else you are turning this into LFS hell even when i am LFS user it's still annoying, wastes time and is harmful to FOSS..
1625[12:17:49] <ayekat> kreyren: pinning stable/experimental falls into the same category - you can do it, but expect things to go awry at some point
1626[12:17:53] <Kocane69> alkisg: oh the zfs is irrelevant to this, it's only for storage, not system.
1627[12:18:06] <alkisg> Kocane69: I don't know of any proper way to fsck ext4 online
1628[12:18:09] <ayekat> also, I'm gonna stop, because now it gets just religious
1630[12:18:45] <ratrace> it's just a lot of noise ; or else we could invite all users of ubuntu and debian derivates to ask here ; they all use apt after all
1631[12:19:16] <ratrace> (which was the original statement ; devuan uses apt so #debian is relevant to ask about unsupported use case)
1640[12:20:43] <Kocane69> alkisg: hm, okay. I guess I can make it run a fsck on boot.. the trouble is that the server is remote
1641[12:20:50] <Kocane69> no IPMI or anything.
1642[12:21:13] <kreyren> ratrace, you already have most of the debian's community helping ubuntu plebs because the solutions are simmilar.. and it should be taken as relevant since devuan can NOT help with debian-specific issues and vice-versa
1644[12:21:55] <ratrace> if they are, then they're supporting in #ubuntu and relevant derivative channel ; otherwise this cahnnel would be very very noisy
1645[12:22:22] <ayekat> also, devuan *should* be able to help with debian-specific issues, since they're the one applying the changes to debian
1648[12:22:53] <alkisg> Kocane69: I don't know how to properly fsck -f automatically on boot even on cases where user input is needed (fsck -a refuses to fix errors) without user input; I think you should re-ask this question in this channel without directing it to me, so that anyone may answer
1649[12:22:56] <ratrace> kreyren: and speaking of asses, the only ass here is you, demanding support for very much unsupported configuration ; you're literally demanding and calling "ass" anyone who points that out ;
1650[12:23:00] <kreyren> ayekat, assuming that devuan's community on irc is almost 10 times smaller and i'm prefering debian packages over devuan i doubt it
1656[12:24:30] <ayekat> kreyren: "the community of my distro is too small" is an age-old (and invalid) argument - if lack of community support is an issue for you, you may need to pick a different distro
1658[12:24:41] <kreyren> ratrace, i'm calling you ass for assuming it beeing devuan issue even when it was debian issue since said system was running 100% debian packages at the time of question and you insisted on asking in #devuan which was provided beeing not relevant assuming conflict within the apt and it's configuration for testing naming on debian repository..
1661[12:25:28] <ayekat> kreyren: also, again, you can ask apt questions here - but if you ask about an unsupported case (which is: mixing repos from different distros), you can't expect help here
1662[12:25:33] <ratrace> kreyren: it cannot be debian issue when you're using unsupported third party repos ; why is that so hard to understand
1664[12:26:48] <ratrace> same if you pointed at an Ubuntu PPA (hey, it's all apt, amirite) and then demanded help over some package that fails to install :: i mean... duh...
1665[12:28:09] <kreyren> ayekat, which is acceptable assuming it beeing unsupported but said question was 100% debian's issue and even if it was frankendebian then any form of help should be encouraged assuming debian's repositories and software used..
1674[12:30:46] <kreyren> and it was still using 3rd party repos on debian.. i doubt you would sent me to WINE with this issue if i ware using their repos.. -> i'm wasn't using devuan at the time of asking
1677[12:31:37] <alkisg> I think the point of argument is: (1) how can I add a ppa? (2) instructions (3) breakage ==> people here don't want to even support (2), before it gets to (3), and it's their right to do so. It's not a channel for supporting "any kind of usage of debian software", but only "supported usage of debian software"
1678[12:31:45] <kreyren> themill, ok
1679[12:32:18] * kreyren goes back to breaking debian
1680[12:33:02] * ayekat just noticed they were already here in March with a weird ubuntu/debian/bedrock/gentoo mix
1681[12:33:56] <kreyren> ayekat, i'm messing with it for longer xD but it was never a problem providing support for it on debian.. in #gentoo developers pretty much get ptsd from seeing my portage configuration and refuse to help
1682[12:34:39] <Kocane69> If I use forcefsck in / to run a fsck on reboot, does anyone know if it saves the output somewhere?
1683[12:35:03] *** Quits: BlueByte (~walther@replaced-ip) (Quit: This computer has gone to sleep)
1684[12:35:27] <jelly> kreyren, while Debian purports itself to be a "universal operating system", and that means you should be able to install any sort of software on it, some ways of getting said software are better and some are worse, some more likely to cause problems with other components, and some less likely
1686[12:36:01] <Kocane69> alkisg: isn't it weird that a command like reboot also disappears?
1687[12:36:17] <kreyren> jelly, i was asked to stop and i don't want to block #debian with this issue we can move in #debian-op or something.. your info is noted
1688[12:36:18] <Kocane69> I reinstalled systemd, rebooted server and the systemctl/reboot is once again gone..
1713[12:45:12] <dpkg> To instruct your system to execute <fsck> on the next boot: "touch /forcefsck". Useful for checking the root filesystem for errors (via /etc/init.d/checkroot.sh). Alternatively, supply the -F option to shutdown(8) to create the /forcefsck advisory file, or include "forcefsck" at the kernel command line.
1714[12:45:24] <jelly> okay, that's a bit outdated...
1715[12:45:33] <Vanfanel> Hi!
1716[12:46:16] <J_C> Vanfanel: hello there
1717[12:46:26] <jelly> there's also a systemd-specific way of doing it via a boot parameter, I can't remember it right now
1721[12:47:19] <themill> initscripts still do it though from rcS.d
1722[12:48:04] <ayekat> huh, didn't know about that forcefsck... TIL
1723[12:48:26] <Vanfanel> I am trying to do an apt-get update on a chrooted "stretch" system, but it gets stuck at "Waiting for headers". The system has internet access and name resolution works. I can even ping to "deb.debian.org" with no problems at all. But still, its stuck at 0% in that "waiting for headers" message. Any idea on what the problem could be?
1724[12:48:40] <kreyren> Can i use variables in sources.list ? like `deb $url blah blah`
1739[12:53:52] <Kocane69> jelly: or I can throw in "fsck.mode=force" in the kernel parameters, no? I still can't figure how I can see the result of it, though.
1744[12:57:06] <Vanfanel> I can even wget files from github, etc... but apt-get update is stuck at "waiting for headers". My sources.list contains just this one: "deb replaced-url
1749[12:59:21] <themill> Vanfanel: do you have firewalling that might get in the way? apt uses SRV records and then goes off to a CDN, it doesn't do http to deb.debian.org
1756[13:02:14] <Vanfanel> themill: the system is accessed via chroot, since its an ARM64 debian stretch system. It has no firewalls. The host system also does not have any firewalls to my knowledge: its a Lubuntu system on x86_64. How can I know if theres something on my network that could be causing the apt-get update fail on the chrooted arm64 system? Any tests I can do?
1834[13:20:21] <Psyndrome> They are in the history but are so many to click one by one
1835[13:20:23] *** Quits: msdoos (~ms@replaced-ip) (Remote host closed the connection)
1836[13:20:24] <themill> vague questions get vague answers
1837[13:20:31] <jelly> Kocane69, ideally you'd see all the logs of things done during initramfs times somewhere in journal after boot, including fsck, but I can't test or confirm if that's true for debian 10 right now
1838[13:20:48] <Vanfanel> themill: I found a possible solution online, that consisted on chmod 666 /dev/null. Now I am not stuck at "waiting for headers" anymore, but I get instead an error with Release.gpg
1839[13:21:04] <ubq323> seems like an replaced-url
1840[13:21:05] <themill> Vanfanel: oh, now that's unexpected.
1856[13:23:54] <Vanfanel> themill: now I get "Terminate called after throwing an instance of 'std::runtime_error' ": what(): random_device::random_device(const std::string&)
1857[13:24:08] <jelly> Vanfanel, show ls -ld /dev/{null,zero,full}
1898[13:29:45] * jelly has not had to deal with a static /dev in a while and may not remember the right MAKEDEV incantation
1899[13:30:09] <Vanfanel> also, I did "mount --rbind /dev dev/" after chrooting. Isnt that bind mounting /dev?
1900[13:30:17] <ubq323> Psyndrome: how did you install these packages? if they are all dependencies or installed automatically then there's an easier way to get rid of them (for instance)
1901[13:30:28] <Vanfanel> or should I do differently BEFORE chrooting?
1902[13:30:38] <jelly> Vanfanel, after chrooting inside chroot? That won't do much.
1906[13:31:41] <jelly> you'd need to do that in the chroot dir, but outside of chroot.
1907[13:32:06] <Psyndrome> They are fonts and i installed them just by clicking one by one in synaptic manager and very few of them was linked as multiple
1908[13:32:25] *** Quits: jubo2 (~juboxi@replaced-ip) (Remote host closed the connection)
1909[13:33:37] <ubq323> does synaptic allow you to un-click them one by one?
1910[13:35:35] <Psyndrome> yes but i have to search them one by one
1912[13:35:54] * kreyren merged beowulf on his debian with openrc and it runs like a charm
1913[13:37:29] <Vanfanel> jelly: I understood the problem. Once I bind mounted /dev, I had the same /dev contents that the host system has. Now everything works.
1914[13:37:41] <Vanfanel> jelly: thanks for your help!
1915[13:37:57] <Vanfanel> themill: thanks a lot! Its solved now! :)
1920[13:41:44] <Kocane69> jelly: i really cannot get any output from this fsck
1921[13:42:07] <Kocane69> I can easily install other packages and use them, but if I install systemd there goes a minute then systemctl exe disappears again
1980[13:56:17] <ZaZaGX> how long is debian buster supported up too?
1981[13:56:29] <jim> so about that... I've been stopping pulse with systemd, then starting jackd, and it starts, but when I try a jack client (like jack_metro), I don't hear it (so jack isn't connecting to the right thing, or the headphone plug isn't connected to the right jack)
2058[14:16:44] <jim> that's when I did a apt autoremove, and I guess that screwed things up for cadence (which probably doesn't have its package metadata right)
2059[14:17:33] <jim> after that, I could bring up the cadence panel but the user interface elements on that panel were all grayed out
2060[14:17:43] <Psyndrome> Thanks for the suggestions see you
2061[14:18:07] <jim> after that experience, I thought, no more cadence
2091[14:34:14] <andi_> hello can somebody be so polite and pastebin the content of apt-get --print-uris --yes install wine | grep ^\' | cut -d\' -f2 >downloads.list the download.list file?
2092[14:34:33] <andi_> i don' thave a debian installation on my hands and need some packages for a offline install
2093[14:34:38] *** Quits: tagomago (~tagomago@replaced-ip) (Remote host closed the connection)
2108[14:36:45] <andi_> i have a debian installation up and running i just don't have access to the internet on the machine
2109[14:36:51] <jim> and the machine is not connected to the net/
2110[14:37:00] <jim> so no
2111[14:37:32] <jim> andi_, ther's apt-zip and apt-offline
2112[14:37:37] <andi_> yeah sadly i don't have a debian with internet availible
2113[14:38:02] <Kocane69> alkisg: turns out it was due to some idiotic scripting of mine in a monit conf.. not sure why I didnt notice it before now since it's been a while since I made that script. All in all, pretty embarresing but atleast the disk is OK
2125[14:44:48] <kilobyte_ch> I'm trying to build a minimal debian rootfs for Arm. For that I'm using multistrap. With Debian 9 this works. But with Debian 10 I'm getting conflicts with multiple packages: replaced-url
2130[14:46:04] <greycat> Why do you want to prevent that, and how do you intend to prevent the users who WANT to use that shell feature from simply bringing their own bash?
2176[15:03:30] <opv> i don't have control over potentially vulnerable web applications, to name an example
2177[15:04:32] <BCMM> i don't understand...
2178[15:04:38] *** Quits: MrShared (~mrshared@replaced-ip) (Remote host closed the connection)
2179[15:04:43] <BCMM> oh i see
2180[15:04:56] <BCMM> post-exploitation mitigation or somesuch
2181[15:05:04] <opv> correct
2182[15:05:13] <BCMM> i'd been imagining this as giving shell accounts to users for a specific purpose, and trying to stop them using it for other purposes
2188[15:06:23] <karlpinc> opv: There's probably something radical you can do with SELinux. :-|
2189[15:06:24] <BCMM> opv: but yeah, if somebody malicious gets a shell, there's a million and one ways they could open a tcp socket
2190[15:06:31] <opv> i assume i could use tools like apparmor or SEL to prohibit the replaced-url
2191[15:06:57] <half-beard> ratrace, would you be able to advise me please on getting my new ubuntu install to be bootable?
2192[15:07:05] <greycat> ask #ubuntu
2193[15:07:11] <half-beard> sorry I meant debian
2194[15:07:19] <half-beard> lol
2195[15:07:22] <greycat> *plonk*
2196[15:07:30] <ratrace> and where did you pick me out of....
2197[15:07:33] <karlpinc> !errors
2198[15:07:33] <dpkg> If you don't tell us the exact error messages you get, we can't tell you what's wrong, and it's also useful to know exactly what command you're typing. Please look for the *first* error that occurs, as this is often the cause of later errors -- ask me about <localized errors> too. Please don't paste in the channel, use a pastebin instead; ask me about <pastebin>.
2199[15:07:35] <greycat> So it's going to be one of THESE days, is it.
2205[15:08:47] <half-beard> this is what I've done: GPT1:4M Grub. GPT2:40G Windoze, GPT3:512M ext4 /boot, GPT4 LUKS. Then I mounted luks as myluks then created a zfspool on that. Then created a 10G ZVol in the zpool and formatted it with ext4 Now Debian is busy installing in that.
2206[15:08:57] <karlpinc> opv: A good first step is to not allow the replaced-url
2207[15:08:58] *** Quits: MrShared (~mrshared@replaced-ip) (Remote host closed the connection)
2222[15:11:09] <half-beard> Some ZFS guy told me he did this
2223[15:11:11] <ratrace> half-beard: just debootstrap with proper zfs datasets
2224[15:11:33] <ratrace> half-beard: then you'd have to ask that some zfs guy to help you with that quite unorthodox (if even possible at all) setup :)
2228[15:12:14] <half-beard> well, if I rsync the fresh install into zfs, would that not be the same end result as debootrapping?
2229[15:12:38] <ratrace> half-beard: kinda, just make sure you adjust fstab, network and other configs, watch out for clashing UUIDs
2230[15:12:53] <half-beard> hmm
2231[15:13:06] <karlpinc> half-beard: The problem might be grub. Somehow the boot loader has to find the kernel (Although you do have a separate /boot so....)
2232[15:13:10] <ratrace> you'll need zfs-initramfs package for root on zfs
2233[15:13:32] <half-beard> ratrace, thanks, will install that
2234[15:14:00] <half-beard> ratrace, do you have an example /etc/crypttab that I can refer to for getting my luks partition decrypted on boot
2235[15:14:07] <ratrace> half-beard: but that's "zfs dataset", not zvol ; i don't think zvols are supported by initramfs-tools
2236[15:14:12] <half-beard> (this is is a workstation, not a dropbear thing)
2237[15:14:23] <opv> karlpinc: absolutely, we already do that. now looking at how to improve the situation
2238[15:15:02] <half-beard> ratrace, just to clear any miscommunication, the zvol is temporary, it's only while I'm installing.
2239[15:15:07] <ratrace> half-beard: "sda3_crypt UUID=.... none luks" adjust where necessary, man crypttab for info on columns
2276[15:26:47] <jim> half-beard, here's what you don't understand yet... there have been a lot of projects that try to take what debian started, and try to add, repurpose, reshape or otherwise change it... and debian is already a fairly large distribution...
2282[15:28:22] <jim> half-beard, if you're running one of the projects that are derived from debian, there's a hidden expectation that the person you'd be asking to help you, knows the derived project equally... and that's a big expectation
2285[15:29:47] <jim> half-beard, what dist do you run now, or do you want to run?
2286[15:29:49] <ratrace> difference between ubuntu and debian, for zfs, is that zfs packages pre-built zfs ; debian needs dkms, and currently there's a bug which requires you to install packages in very specific order, spl-dkms first, then zfs-dkms, then zfsutils-linux, which fails, modprobe zfs, then apt install -f ; none of that is needed on ubuntu
2287[15:30:03] <ratrace> so if this is really ubuntu, the above recommendation would fail
2299[15:31:34] <BCMM> opv: was going to suggest something on those lines... if there's a finite white-list of functions that something should perform, restrict it on that basis, rather than attempting to blacklist specific malicious actions
2300[15:31:43] *** Quits: dvs (~hibbard@replaced-ip) (Remote host closed the connection)
2301[15:31:57] <jim> greycat, frankly, I was fairly surprised that debian has added additional packaging systems... I would have thought that was fairly insane
2307[15:32:28] * tarzeau is happy with xfs and btrfs (for the compression part)
2308[15:32:37] <kreyren> Looking for zoom software on debian
2309[15:32:42] <jim> I don't know and not really wondering at the moment :)
2310[15:32:49] <tarzeau> and exfat
2311[15:33:00] <tarzeau> kreyren: homepage of it?
2312[15:33:03] <greycat> I think it's something related to SNS.
2313[15:33:07] <colo-work> tarzeau, because, for instance, replaced-url
2314[15:33:13] <kreyren> tarzeau, nah something that is able to zoom on something on screen
2315[15:33:31] <tarzeau> kreyren: a magnifier?
2316[15:33:41] <kreyren> tarzeau, correct
2317[15:33:43] <ratrace> people want zfs because they perceive btrfs as broken ; and it's really about "why do people want a CoW volume manager that protects against bitrot and has easy to use snapshots"
2318[15:34:10] <kreyren> tarzeau, ideally something that doesn't follow mouse cursor so that i could use 8K scren resolution for better workflow
2319[15:34:33] <kreyren> tarzeau, and something that is full screen
2320[15:34:45] <tarzeau> kreyren: you've got an 8k screen?
2321[15:34:58] <kreyren> tarzeau, yep
2322[15:35:05] <kreyren> not currently in use tho
2323[15:35:19] <tarzeau> because with linux it'd be of no use, right :)
2324[15:35:23] <greycat> I've seen a bunch of talk about zfs so far, and AT NO POINT while reading any of it did the words "easy to use" ever cross my mind.
2325[15:35:26] <tarzeau> it's even hard to use 4k screens with linux
2334[15:37:34] <kreyren> ratrace, ideally using shortcut to zoom in cursor's current position, but never follow mouse cursor..
2335[15:37:52] <ratrace> I got a 35" curved AOC screen ; I deliberately did not want 4k because I did not want to need a magnifying glass (a literal, physical one), to hunt down what it says on the screen :) 8k? if that thing is not at least 60" then you'll need a microscope, not a magnifying glass
2336[15:38:17] <ratrace> kreyren: maybe you meant tarzeau ? ;)
2383[16:02:16] *** Joins: Abdullah (~ak@replaced-ip)
2384[16:02:30] <Kocane69> ratrace it's just a server, where I need em for hardware accelerated transcoding
2385[16:02:31] <Kocane69> vaapi
2386[16:02:50] <Kocane69> I guess a startup script that chmod 666 on the files could help but...
2387[16:03:51] <ratrace> or put the user that needs access to them, in those respective groups
2388[16:03:52] <greycat> Kocane69: in Debian, management of the ACLs of hardware device nodes (sound cards and such) is done during login by the systemd stuff.
2389[16:04:34] <greycat> Kocane69: Logins are tied to a concept called "seats", etc. A console login is supposed to grant access to the console hardware. An ssh login is not.
2428[16:22:36] <tarzeau> comes from coreutils. i wonder why
2429[16:22:57] <tarzeau> !pwsh
2430[16:23:54] <greycat> dir(1) is GNU only, not POSIX... I can only guess it was written to help DOS users adjust to their new environment gently, but who knows
2434[16:25:42] <Wulf> gamblr: 7.7E9 people on the planet, 3E6 people in Nevada. 1538 people here. A (too simple) formula would be 3E6 / 7.7E9 * 1538 = 0.6. Assuming that most people here are not from Asia and Nevada is a reasonable location for #debian people, there is a really high chance. So let's assume "yes".
2448[16:30:30] <ratrace> half-beard: no, that's just bug in packaging
2449[16:31:17] <half-beard> jim, well I've been running 18.04 as my development machine's OS because it seems to be most compatible with most software I've wanted to install, then debian for all my containers, VMs, embedded devices, chroots on android and dedicated servers
2451[16:31:56] <half-beard> but now 18.04 is old and I feel that ubuntu users are not particularly technical and have been enjoying the level of expertise in #debian type of conversations
2452[16:32:08] <half-beard> So I'm thinking to run Buster KDE for my dev machine now
2464[16:33:48] *** Parts: jozef (~jose@replaced-ip) ()
2465[16:34:03] <half-beard> alkisg, I've been thinking of trying root on ZFS
2466[16:34:11] <half-beard> so that's why I wiped
2467[16:34:15] <jim> I would have suggested you back up first... but, here we are...
2468[16:34:23] <half-beard> Also the possibility of a 40G windoze partition for Overwatch
2469[16:34:29] <alkisg> I'm thinking of migrating a school from ubuntu to debian. Apart from the firmware and adding contrib/non-free to sources.list, what else should I keep in mind?
2470[16:34:32] <jim> do you have buster installation media?
2471[16:34:40] <jhutchins_wk> Strange to receive a point release notice more than a month in advance.
2472[16:35:08] <ratrace> alkisg: that ubuntu has a lot of ubuntu-specific kernel patches for hardware enablement
2473[16:35:27] <alkisg> (the school computer lab will be maintained by the IT teacher, I'll only help when they have issues...)
2474[16:35:28] <half-beard> ratrace, what method do you recommend for installing buster on zfs?
2477[16:36:06] <alkisg> ratrace: thanks; I guess I'll see that part upon installation... are peripheral devices like printers/scanners/usb sticks affected?
2478[16:36:13] <half-beard> do you mean with the above guide?
2480[16:37:01] <jhutchins_wk> alkisg: Do some test machines first, see what you're actually dealing with rather than trying to anticipate everything through speculation.
2483[16:37:51] <alkisg> jhutchins_wk: we using ltsp, so we only install to 1 machine and netboot all others; I'll see about hardware compatibility when I do that first installation; I was wondering about other things that I won't see in the first hour...
2484[16:38:38] <alkisg> E.g. at some point gnome was using cantarell in debian, and didn't even have greek glyphs; I filed a bug report about it 5+ years ago but I'm not sure if it was solved...
2485[16:38:45] <ratrace> half-beard: i suppose so ; i haven't read it in detail though at a glance it looks a bit.... overreaching
2503[16:42:41] <horribleprogram> Aug 01 10:39:43 horribleprogramDebian sshd[2176]: pam_unix(sshd:session): session opened for user horribleprogram by (uid=0)
2504[16:42:49] *** Quits: sebatron (b92ed455@replaced-ip) (Remote host closed the connection)
2505[16:43:15] <horribleprogram> lemme try restarting the service
2506[16:43:24] <half-beard> ratrace, I thought it looked a bit overboard
2507[16:43:42] <half-beard> ratrace, what are the basic steps you perform when you do a debian zfs install?
2508[16:44:16] <half-beard> ratrace, would you say a debian live ISO is required or could I do it from an ubuntu one? since it's debootstrapping, ubuntu live should work?
2512[16:44:42] *** Quits: towo` (~towo@replaced-ip) (Remote host closed the connection)
2513[16:44:44] <half-beard> I'd be debootstrapping buster onto the zfs
2514[16:44:58] <ratrace> half-beard: essentially you need section "2.4b LUKS", you don't need a separate bpool, separating /var and friends into datasets is okay, but you'll need legacy mounts, systemd has a race condition since it takes zfs-mount.service to mount if not legacy
2523[16:47:02] <half-beard> Yeah, I'm currently questioning why I should even run root on ZFS at all
2524[16:47:04] <ratrace> for that, there's a zfs-mount.service that mounts the datasets, but that service is started after systemd requires /var, /usr and other important bits, thus you need legacy mounts via fstab which are done much earlier in boot
2525[16:47:09] <half-beard> makes life a lot more complicated
2526[16:47:32] <half-beard> Perhaps I should just stick with LVM for my roots and just keep using ZFS for VM's
2527[16:47:42] <ratrace> half-beard: zfs on linux is pretty much broken ; especially due to its license, its future compatibility with the linux kernel is questionable
2528[16:47:49] <ratrace> half-beard: btrfs ain't that bad
2529[16:48:13] <half-beard> yeah, I'm thinking because zfs on linux is so uncertain... I like being able to boot reliably.
2535[16:48:42] <half-beard> At least if there's an issue I can still boot.
2536[16:48:46] *** Quits: isr60 (~u0_a115@replaced-ip) (Remote host closed the connection)
2537[16:49:07] <ratrace> it'll be interesting to see how zfs goes with debian 10.1 ; it incldues 4.19.38+ kernel which breaks zfs, which needs simd patched out ::: there's a patch with newer fixes for 5.x and ported kernels, dunno if it'll be included, it's deemed "new feature"
2538[16:49:22] <ratrace> so yeah... interesting times ahead for zfs on debian
2539[16:50:56] *** Joins: jozef (~che797@replaced-ip)
2540[16:51:20] *** MathAmphetamine is now known as CrystalMath
2558[16:59:40] <jhutchins_wk> ratrace: Because you have a deliberately unrealistic notion of available space based on projected compression efficiency.
2559[16:59:41] *** Quits: CombatVet (~c4@replaced-ip) (Remote host closed the connection)
2560[16:59:54] <ratrace> jhutchins_wk: you havent' used zfs then :)
2561[17:00:02] <ratrace> zfs is used by many professional/enterprise storage appliance vendors ; albeit based on FreeNAS :: on FreeBSD ZFS reigns supreme
2562[17:00:12] *** Parts: jozef (~che797@replaced-ip) ()
2565[17:01:08] <ratrace> jhutchins_wk: on btrfs compression is hard to track because btrfs deals with extents and can actually have very mixed profiles within single filesystem (single, dup, raidX, ...)
2589[17:10:57] <GenTooMan> I am using debian Buster, I was attempting to run nrsc5 when I received this mess "Please fix the device permissions, e.g. by installing the udev rules file rtl-sdr.rules", whereas I know where the rules are etc. it's weird they aren't already their as I installed the drivers...
2598[17:13:50] <GenTooMan> petn-randall rtl-sdr/Software defined radio receiver for Realtek RTL2832U (tools)|librtlsdr0/Software defined radio receiver for Realtek RTL2832U (library)|librtlsdr-dev/Software defined radio receiver for Realtek RTL2832U (development)
2605[17:18:18] *** Quits: ich (~ich@replaced-ip) (Quit: Ex-Chat)
2606[17:19:01] <GenTooMan> hmmm I don't remember this problem in stretch well the rules are located in /lib/udev/rules.d I will reread the rules debian documentation, it's weird because I found the RULES (searching) file but it's not installed. I'll have to reinit udev too.. argh.
2607[17:19:38] *** Quits: Immanuel (~Manu@replaced-ip) (Ping timeout: 272 seconds)
2740[18:46:39] <diogenes_> Vanfanel, cobgrats, now we all know your password :)
2741[18:46:46] <diogenes_> cobrats*
2742[18:46:47] <grobi> diogenes_: did you experience the same??
2743[18:46:55] <Vanfanel> diogenes_: yeah, so it seems XD
2744[18:47:08] <diogenes_> grobi, no i don't because i never upgrade.
2745[18:47:18] <grobi> :)
2746[18:47:22] <msdoof> yeah it is great, that also happened to me and it was the same password as my e-mail-account and i did not change after posting it to irc :)
2759[18:51:01] <Vanfanel> I am having a strange problem with a minimal Debian buster install: apt-get says "Unable to acquire the dpkg frontend lock (/var/lib/dpkg/lock-frontend), is another process using it?", but the system just booted and there are no graphical frontends to apt in this system. If I delete /var/lib/dpkg/lock-frontend, apt-get creates it again when I run it, and fails.... any idea?
2760[18:51:10] <diogenes_> grand now: echo $XDG_SESSION_TYPE
2787[18:55:31] <Vanfanel> alkisg: there is not automatic updates or any other service running apt automatically. The lock file is created WHEN I run apt
2815[19:04:13] <diogenes_> grobi, then create a new user, it might be some misconfigs and if it works with the new user, then the problem lies in your /home/user/.
2816[19:04:29] *** Quits: jikz (~k047n0de@replaced-ip) (Remote host closed the connection)
2823[19:06:36] <Vanfanel> ratrace: like EXT4 ACCESS CONTROL LISTS, Security labels, encryption, debugging... no one of these was needed with an Stretch system for APT/DPKG to work
2824[19:08:35] <Vanfanel> alkisg: flock seems to work fine. Anything special to try?
2849[19:25:45] <Vanfanel> jhutchins_wk: no, I tried just in case some scrip/service/whatever was using apt, but no, the lock is created only when I manually run apt
2857[19:28:30] <jhutchins_wk> I don't understand what you said about chroot though, are you building a new system from an old one, or are you chroot'ing to / ?
2874[19:39:26] <useretail> hey guys, how to fix broken apt? i'm getting libgl1 : Depends: libglx0 (= 1.1.0-1) but it is not going to be installed, but it's installed
2875[19:39:44] <useretail> libgl1-mesa-glx : Depends: libglx-mesa0 but it is not going to be installed
2876[19:39:50] <useretail> libkf5coreaddons5 : Breaks: libkf5auth5 (< 5.54) but 5.28.0-2+deb9u1 is to be installed
2877[19:39:56] <useretail> Breaks: libkf5globalaccel-bin (< 5.54) but 5.28.0-1 is to be installed
2878[19:40:01] <useretail> libkf5crash5 : Breaks: libkf5globalaccel-bin (< 5.54) but 5.28.0-1 is to be installed
2879[19:40:07] <useretail> E: Error, pkgProblemResolver::Resolve generated breaks, this may be caused by held packages.
2881[19:40:37] <jmcnaught> useretail: please use replaced-url
2882[19:40:40] <jmcnaught> !bat
2883[19:40:40] <dpkg> In order to troubleshoot your problem with apt-get, apt or aptitude we need ALL OF THE FOLLOWING information: 1. complete output of your apt-get/apt/aptitude run (including the command used) 2. output from "apt-cache policy pkg1 pkg2..." for ALL packages mentioned ANYWHERE in the problem, and 3. "apt-cache policy". Use replaced-url
2899[19:43:41] <weedloser> sheldon cooper from big bong theory is my favourite technology advisor
2900[19:43:58] <Kocane> Anyone using ZFS on Debian who can help me figure out how to ensure that ZFS pool is mounted before a service start? Should [Unit] After = local-fs.target zfs-import.target be sufficient?
2920[19:51:39] <useretail> so it's 9 with 60-70% packages from 10
2921[19:51:57] <jmcnaught> useretail: so you are in the middle of the stretch→buster upgrade? Can you make pastes of the rest of the information requested by dpkg in the !bat factoid?
2941[19:59:15] <jmcnaught> useretail: so are you partway through a Debian 9 to 10 upgrade? What does "apt upgrade" want to do? Also I strongly recommend to use codenames in your sources.list ('buster' not 'stable')
2942[19:59:19] *** Quits: xcm (~xcm@replaced-ip) (Remote host closed the connection)
2980[20:26:48] <Vanfanel> Could recent versions of DPKG depend on some obscure kernel option on which it did not depend before? Related to file locking, since thats what I am having problems wit
2991[20:31:10] <karlpinc> I'm curious why debootstrap does not install the "locales" package. (My chroot was unhappy until I installed locales and did a dpkg-reconfigure.)
2992[20:31:27] *** Quits: useretail (~useretail@replaced-ip) (Remote host closed the connection)
2993[20:32:38] <vlt> karlpinc: debootstrap without --include=... only installs a very basic system.
3038[20:52:15] <alkisg> Vanfanel: with the stock jessie kernel, I'm getting only 8 MB RAM used right after boot, how much smaller do you need it, or is it about space, not RAM?
3091[21:10:47] *** Quits: BadOgre (~BadOgre@replaced-ip) (Remote host closed the connection)
3092[21:11:13] <dvs> What Desktop Environment
3093[21:11:18] <dvs> (DE)
3094[21:12:18] <diogenes_> and there is a possibility, if it worked in the previous version before upgrading then write a usb with the previous version and boot a live session.
3095[21:12:21] <grobi> aha :) i use two things xfce (my default) but mostly i use my i3wm session
3205[22:09:55] *** Quits: bliv (~bliv@replaced-ip) (Remote host closed the connection)
3206[22:10:13] <vishal> I *think* I found everything I was looking for in the configs dir. A pastebin from someone running it on amd64 would be nice for confirmation though.
3270[22:33:46] <n00buser> humpled, sddm starts, but i don't see login screen
3271[22:34:04] <nunllk> one question regarding debian install: I am installing on a clean pc without any OS installed but still the installer says that he found "other os installed in BIOS mode", why?
3272[22:35:02] *** Quits: Vanfanel (~quassel@replaced-ip) (Remote host closed the connection)
3273[22:35:27] <diogenes_> n00buser, /ect/default/grub remove the resume=xxx line and then sudo update-grub
3276[22:35:50] <nunllk> im using the graphical installer
3277[22:36:00] <nunllk> what do i do to achieve what u are suggesting?
3278[22:36:01] <Bushmills> nunllk: a remainder of a previously installed OS - it's partition may still be there, and Debian installer concludes from its presence that there must be another system
3279[22:36:16] <nunllk> i disconnected all other drives
3280[22:36:21] <nunllk> only the clean drive is connected
3281[22:36:52] <Bushmills> any partitions on that clean drive?
3282[22:36:56] <n00buser> diogenes_, well, there's no such line
3283[22:36:58] <diogenes_> nunllk, because bios still keeps it's record on that drive i suppose.
3284[22:37:06] <nunllk> no partitoons on the drive
3285[22:37:09] <nunllk> fully wiped
3286[22:37:44] *** Quits: xcm (~xcm@replaced-ip) (Remote host closed the connection)
3287[22:37:54] <n00buser> nunllk, you may have other drives (just guessing)
3376[22:55:55] <mtn> n00buser: I install inxi and run: inxi -Gx
3377[22:56:00] <n00buser> diogenes_, no changes
3378[22:57:29] <n00buser> mtn, it says driver: nvidia v:340.107
3379[22:57:59] <mtn> n00buser: I still think the driver needs to be reinstalled. why not try it?
3380[22:59:00] <n00buser> mtn, well, it's proprietary + obsolete and is not available from debian repos anymore
3381[22:59:05] <nunllk> fixed!
3382[22:59:12] <nunllk> i made usb stick with dd
3383[22:59:20] <nunllk> and gave the clean disk gpt partition scheme
3384[22:59:26] <nunllk> ty boyz xD
3385[22:59:29] <diogenes_> n00buser, then only 3 things that come to mind are: 1) look in /etc/X11/ and /etc/X11/xorg.conf.d for any suspicious configs 2) rename ~/.config folder 3) sudo apt purge nvidia*
3386[22:59:42] <mtn> n00buser: did you try booting with nomodeset in grub?
3387[22:59:45] <greycat> I see lots of packages with "nvidia" and "340" in their names.
3388[22:59:53] <nunllk> btw
3389[22:59:56] <Vanfanel> Habbie: Do you happen to know what service allows CTRL+ALT+Fn TTY changing??
3390[22:59:59] <nunllk> do you recommend using LVM?
3391[23:00:07] *** Quits: mirabela (~mukesh@replaced-ip) (Remote host closed the connection)
3392[23:00:24] <greycat> Vanfanel: That's the Linux kernel itself, but it's configured by some voodoo deep inside systemd...
3395[23:01:02] <Vanfanel> greycat: thing is, I have disabled every uneeded service, and now I can NOT change to different TTYs
3396[23:01:10] <greycat> E.g. Ctrl-Alt-F8 takes me to tty8 but there is nothing *running* on tty8, so I just get the black screen with cursor.
3397[23:01:31] <greycat> So the actual switching-of-virtual-consoles is done by Linux itself, and the lack of anything running there is (not) done by systemd units.
3398[23:02:17] <greycat> And there is probably some way to tell Linux "only support 6 virtual consoles, not 24" or whatever. I do not happen to know what that might be.
3399[23:02:35] *** Quits: enoq (~enoq@replaced-ip) (Remote host closed the connection)
3412[23:07:27] <Vanfanel> greycat: I have fixed the problem. I dont use static TTYs, so I only have those enabled as services by me. And I had forgot to enable the tty2 :D
3413[23:07:29] <Vanfanel> thanks!!
3414[23:07:37] <greycat> which may or may not control the number of gettys that get spawned, who the heck knows
3415[23:08:20] <greycat> I also know that getty@.service is involved somehow.
3418[23:10:08] <Habbie> Vanfanel, as was basically said - it's a kernel feature, but it relies on those ttys actually having something running, which these days is indeed managed by systemd
3419[23:10:13] <Vanfanel> greycat: if the static ttys are disabled, well, only those launched by getty@.service are spawned
3476[23:36:56] <dpkg> rufus is a tool that can be used to make bootable USB devices under Windows. It is not recommended for use with Debian CD/DVD images, as it mangles the installer in cruel and unusual ways, resulting in hard to debug problems. Ask me about <hybrid images>, <usb install>, <win32diskimager>.
3477[23:37:14] <greycat> !yumi
3478[23:37:14] <dpkg> yumi is probably a tool that can make bootable usb device. It is not recommended for use with Debian CD/DVD images, as it mangles the installer in cruel and unusual ways, resulting in hard to debug problems. Ask me about <hybrid images>, <usb install>, <win32diskimager>
3483[23:40:59] <nunllk> firmware-amd-graphics, firmware-linux-nonfree, firmware-misc-nonfree firmware-realtek and nvidia-driver included all drivers i needed