17[00:10:54] <dpkg> Some packages intended for Buster (Debian 10) but recompiled for use with "Stretch" (9.x) can be found in the "stretch-backports" repository. See replaced-url
18[00:11:02] <sloshy> i looked online and saw a lot of different things and i don't know what is safest
19[00:11:07] <somiaj> sloshy: at this junction, I would suggest just upgrading/installing buster
37[00:12:52] <sloshy> i prefer to just back up my files and do a clean install
38[00:13:07] <somiaj> this is more like a oldstable->stable upgrade (there are still a few things to iron out, but it is near enough to the release things should be mostly ready)
40[00:13:20] <sloshy> i didnt realize buster was going to be the new stable so soon
41[00:13:29] <somiaj> well if you are gonna backup, might as well just try an upgrade. I did one a few days ago, took like 30mins, no problems, running buster
43[00:14:03] <somiaj> It has been frozen since feburary, so there has been a good amount of time of testing the packages and upgrade path
44[00:14:23] <somiaj> (note there are still a few issues left, but it is mostly ready and you shouldn't experience to many problems)
45[00:14:33] <dvs> 76 bugs left!
46[00:14:48] <somiaj> since you are wanting the full mesa stack, I think this will be better, and just give you a head start on running buster before the offical release in a week and half.
47[00:15:13] <somiaj> in 20,000+ packages, not bad
48[00:16:02] <dvs> only? ;-)
49[00:17:29] <dvs> Wasn't it just a little while back when a release had just 1000 packages?
50[00:17:51] <dvs> according to the release notes?
104[00:37:43] <somiaj> sloshy: gnome just needs to practice being more modular, so if they want to introduce something a bit more core (to improve the desktop expereince) not push it on non-gnome users.
105[00:38:04] <somiaj> I think there is room for gnome in the linux world, they just need to also play nicely
106[00:38:08] <sloshy> well the gnome developers are total douchebags
107[00:38:17] <sloshy> in their mind gnome is the only desktop that matters
108[00:38:39] <sloshy> and everyone using DEs other than gnome can go screw themselves
109[00:39:06] <sloshy> ive seen gnome devs on reddit acting like total dicks, arguing with people who criticize any of their decisions
110[00:39:18] <EdePopede> wasn't one of the fundamental changes from 2 to 3 to take away a lot of choices from the user?
111[00:39:29] <sloshy> and basically saying "it's my way or the highway"
112[00:39:43] <sloshy> yeah
113[00:39:48] <sloshy> they dont even like theming
114[00:39:51] <sloshy> because it "weakens the brand"
115[00:39:54] <sloshy> the brand
116[00:40:02] <sloshy> like theyre apple or something
117[00:40:09] <somiaj> we should of course try to be a little more civil and try to keep on debian support here. (:
126[00:43:14] <EdePopede> could still become relevant for debian if they decide that it isn't gnome anymore when it gets themed. sounds familiar somehow...
135[00:48:52] <sloshy> is there any benefit to using a swap partition over using a swap file on the root partition
136[00:49:42] <somiaj> I think hibernation works better, I think it is also faster (since you don't have the overhead of writing to a file inside a file systme)
138[00:50:27] <somiaj> the swap partition can never be fragmented, while a swap file could, you can put the partition on a part of the disk that is faster to read
139[00:51:00] <somiaj> hmm, seem (according to that link) since linux 2.6, provided the swap file is not fragmented there isn't a prerforamnce differnce.
199[01:08:41] <_Vi> Why during dist-update it sometimes asks me whether to update a config file in regular text, but sometimes on blue-background curses interface? Both variants can alternate during the same upgrade.
239[01:38:53] *** SuperKaramba is now known as BenderRodriguez
240[01:40:05] <trackanddirt> I am using strech with keepalived, I have it triggering scripts but although they show under service status as executed, they are not, is this a known issue?
241[01:40:10] *** Quits: dvs (~hibbard@replaced-ip) (Remote host closed the connection)
348[02:08:05] <donofrio> how do I tell....I was here yessterday evening about the powerpc install that needed the archive_2019.key applied but I didn't have the mirror setup, now I foudn the .deb and I'm like almost there I thought
349[02:08:28] <themill> I have no idea what you are asking
472[02:57:18] <jmcnaught> teksimian: the new version of the debian-installer for buster asks if you want to leave free space on the volume group, but it's something you have to do manually in the stretch installer
474[02:57:31] <teksimian> jmcnaught: and to rollback you use the merge command, but as i understand it, it's essentially a live rolling backup... so how does it know when in time to rollback to ?
476[02:58:10] <teksimian> jmcnaught: is buster testing or unstable at this point?
477[02:59:20] <teksimian> wiki says testing :)
478[03:00:01] <jmcnaught> teksimian: buster is testing, it's scheduled to be released on July 6th. To merge you do "lvconvert --merge vg/lv" and it sets the origin logical volume back to the state it was at the time of the snapshot
546[03:54:14] <donofrio> iulian_, I'm working on an sid/buster netinstall that has no gpg on the mirrors so I cannot update or something like that....I'm booted and sshed into the client from my workstation
547[03:55:16] *** Quits: Tom01 (~tom@replaced-ip) (Remote host closed the connection)
569[04:08:44] <donofrio> themill, I'll use whatever sources you good folks say, can you paste.ubuntu.com over your sample of what I should do cause last time I switched sources.list I was unable to get back to working or general confusion
571[04:09:41] <themill> I have no idea what you need, I just know that powerpc is not a supported architecture any more
572[04:10:08] <somiaj> I know that #debianppc on irc.oftc.net is fairly dead, but that is still your best bet at getting someone with experience with this.
573[04:10:09] <themill> The last supported powerpc release was jessie fwiw
916[08:47:33] <mrtnt> If one is serving (mostly) static web pages with Apache on Debian, then is there a still point to set up caching? One would think that operating system caching is enough..
917[08:47:45] *** Quits: Kevlar_Noir (~manjaro-u@replaced-ip) (Remote host closed the connection)
1038[09:41:34] <CrazyTux> Habbie: but that package requires at least 250 GBs of disk space.
1039[09:41:34] <Habbie> CrazyTux, ok - that's not really a debian question, but if you can mention the packages, maybe somebody will know anyway
1040[09:41:45] <finn0> themill: Isn't it good idea to automatically subscribe users who has participated in discussion like other bug tracking system (for ex. bugzilla, gitlab) etc?
1041[09:41:47] <CrazyTux> Habbie: SAP IDES.
1042[09:42:02] <Habbie> CrazyTux, those don't sound like they demand working 3D or other 'hardware' stuff
1043[09:42:08] <Habbie> CrazyTux, so my guess would be that it would work
1044[09:42:34] <CrazyTux> Habbie: it requires at least 250 GBs of disk space.
1045[09:42:40] <Habbie> CrazyTux, so?
1046[09:42:53] <CrazyTux> Habbie: how can I allocate that much disk space?
1047[09:42:58] <ayekat> assign it a 250+ GB virtual disk
1050[09:43:33] <CrazyTux> I tried dual booting windows with linux distro. But, encountered some problems.
1051[09:43:53] <CrazyTux> I thought installing it in VBox would be a better idea.
1052[09:44:06] <Habbie> what problems did you have?
1053[09:44:25] <CrazyTux> I also need to install and use MS Office. Can I install it inside VBox?
1054[09:44:41] <CrazyTux> Habbie: one of the OSes not getting detected.
1055[09:45:09] <Habbie> lots of people use ms office in vbox
1056[09:45:19] <ayekat> CrazyTux: given the nature of the questions so far, though... sanity question: do you have enough space to store a 250+ GB disk?
1057[09:45:39] <ayekat> because "how can I allocate that much space" makes me think that you don't have the space on your disk
1099[09:55:50] <CrazyTux> Habbie: would it be cumbersome to install SAP IDES inside VBox? would it be as easy as installing it on Windows that is installed on HDD?
1100[09:56:14] <Habbie> CrazyTux, it should be exactly as painful as on windows without vbox, unless there is something special about SAP IDES
1149[10:07:54] <uio> Hello, doing a Debian stable install on an Asus eeepc with the encryption option provided in the installer. Now I have rebooted the machine, it displays the unlock encryption prompt, so I enter the password, it says it's fine, and then shortly after, black screen. I should perhaps mention that during the install, I got through the 'write random information' just before the key selection, and then had to interupt install. So I started
1150[10:07:54] <uio> installation again after, but all seemed to go well using the non GUI installer. What is going on?
1181[10:17:45] *** Quits: CaptainN (zelda@replaced-ip) (Quit: I have to pee!)
1182[10:18:32] <Marbug> Good morning! I seem to have an issue with python, and more specifically pip packages. I got multipel packages that are throwing a stack trace after pip is mentioning Cleaning up.. and the package doesn't seem to be installed. That is with pip v9 from apt, whenever I upgrade pip through pip to v19, it quite immediately throws a stack trace about so
1183[10:18:32] <Marbug> mething in main. pip2 isn't really an issue, it install everything, although I need the pip3 packages, and that is crashing constantly. Is there a way to reinstall all pip and pip3 packages? Or even the system packages?
1184[10:18:51] <uio> Would adding nomodeset help me?
1187[10:19:06] *** jbond42|away is now known as jbond42
1188[10:19:27] <CrazyTux> Win 10 or Win 8.1 in VirtualBox?
1189[10:19:42] <blackflow> Marbug: just reinstall the virtualenv? Hopefully you have a file with list of packages (and versions) you can easily pip install -r with
1190[10:19:54] <blackflow> and first thing in the new virtualenv, upgrade pip
1195[10:21:50] <patterson> VB is broken i windows 1903
1196[10:21:53] <patterson> Pro
1197[10:22:34] <patterson> Here,. Try a bare mwetal hypervisor
1198[10:22:36] <Marbug> blackflow it's not a virtualenv, well it's an lxd container, but then going further, when I wanted to delete it, it mentioned that the dataset was busy as I used lxd from snap potencially, and there might be a bug :(
1199[10:22:57] <blackflow> Marbug: wait, you didn't install packages as root did you? even if lxd
1202[10:23:15] <Marbug> I know the best option is to start from scratch, I already have a temporary solution where I installed a new lxd env with the requirements
1203[10:23:23] <Marbug> blackflow I did install the packages as root :)
1204[10:23:52] <blackflow> Marbug: yeah no, you shouldn't do that. Even if it's a container, root'd pip will clobber system python packages. don't do that. use a virtualenv
1205[10:23:53] <Marbug> as I don't want to install it as the user, I use the same user as on my host inside the lxd enviroments
1206[10:24:06] <blackflow> and also build a list of packages you can use with -r for pip
1207[10:24:13] <Marbug> that I have :)
1208[10:24:30] <blackflow> you can run as root, but use a virtualenv for pip installed stuff
1209[10:24:33] <Marbug> so virtualenv would be a solution, even when using lxd ?
1210[10:24:50] <blackflow> yes because even an lxd has a base OS system installation you can easily break
1211[10:25:15] <blackflow> containers are ONLY (and nothing more than) namespaces. inside namespaces you need a bare minimum OS userland (so, sans teh kernel and sans systemd)
1212[10:25:16] <uio> I tried following the advice replaced-url
1220[10:26:25] <Marbug> well anyway, I'll try to get it fixed using a virtualenv :)
1221[10:26:33] <blackflow> Marbug: wel thats okay. my point was that you don't need an init (any init) inside a container, but sure you can treat the container as a "VM" and have full OS sans the kernel
1222[10:26:33] <Marbug> thanks for the feedback blackflow!
1223[10:26:45] <finn0> I found a torbrowser-launcher package which is referred to by other package(s) but not package is not available.
1224[10:27:21] <Marbug> yep that is what I want, as we install multiple packages, and I wantto have the flexebility of a system and not as a container like docker blackflow :)
1225[10:28:19] <blackflow> Marbug: that's fine. which still means you should use pip only inside a virtualenv, as it _will_ break system python
1226[10:28:28] <blackflow> (otherwise will)
1227[10:29:37] <Marbug> okido, and I thought that containing it within an lxd would be fine, well I already did it for more than a year as I want to avoid virtualenv, up untill now when I got a brand new laptop :)
1228[10:29:46] <blackflow> uio: did you remove "quite" from the kernel cmdline in grub? do you have any output after passphrase?
1229[10:30:17] <blackflow> Marbug: it is fine wrt the _host_ but it is not fine wrt the OS inside the contianer. you just have another OS in there with very same problems as the host.
1230[10:31:16] <blackflow> so, you won't clobber host's filesystem, only guest's (if we can use the term "guest" for a container)
1263[10:41:20] <uio> Is it a free / non-free thing!,
1264[10:41:23] <uio> *?
1265[10:41:39] <blackflow> uio: you will know when you try the text mode and read the logs.
1266[10:41:46] <uio> okay!
1267[10:42:09] <blackflow> btw afaik lubuntu doesn't pre-install proprietary drivers, so it's using the very same kernel drivers, albeit different version and different ubuntu patches
1268[10:42:45] <uio> This is tricky to enter the right chars with a keyboard mismatch...
1365[11:39:01] <blackflow> mms_: sorry, I had to step out for a moment. yeah, says no carrier for enp0s3, so that'd mean you misconfigured it on the host side.
1373[11:41:49] <blackflow> mms_: that's virtualbox and/or windows specific issue, not debian
1374[11:42:10] <mms_> I tried adding allow-hotplug enp0s3 and then iface enp0s3 inet dhcp but not good
1375[11:42:21] <mms_> blackflow: ok
1376[11:43:22] <blackflow> mms_: otoh, see if virtualbox supports "virtio" network driver, and use that instead. that'll change the NIC name in the guest though. but use virtio wherever possible.
1377[11:45:15] <tachikomas> mms_: are you sure your network is bridged ?
1378[11:45:29] <tachikomas> if you want some easy stuff you can always use Network Manager
1393[11:51:31] <blackflow> mms_: virtio is paravirtualized and if the VB on windows supports it, then by all means use it.
1394[11:52:03] <blackflow> mms_: "need"? no, but virtio is performing better and is supportd by the kernel better, than some badly emulated intel nic
1395[11:52:58] <blackflow> mms_: this setting is on the host side. "Adapter type" under Advanced, but I don't know if there's virtio on windows, haven't used it in many years.
1398[11:54:10] <mms_> blackflow: my laptop is work laptop and have admin rights for limited time to do things but if I add things that will conflict with their standard drivers it will be problem
1399[11:54:35] <blackflow> mms_: you don't have to add anything on the host side, you only need to select virtio-net from the drop down, if it's there.
1425[12:01:29] <blackflow> mms_: I'm gonna hazard a guess and say yes, you need to plug in the virtual cable for the virtual nic to work the virtual network. :)
1471[12:28:13] <uio> Hi, did a fresh Debian 9 install. After entering disk key without the quiet setting I see lines of text fly by like normal until it passes 'Welcome to Debian' and some more lines. Then right when the screen text usually resizes to smaller text on a working boot, the screen goes black. How can i solve this?
1503[12:38:50] <blackflow> if that doesn't fix it, I have no idea what could be wrong. the installer USB worked, right? perhaps you can reboot into it, mount your installed filesystem and inspect its logs.
1504[12:39:06] <blackflow> there should be a rescue mode in the installer
1510[12:42:45] *** Quits: mayurvpatil (~mayurvpat@replaced-ip) (Quit: This computer has gone to sleep)
1511[12:42:56] <finn0> I was installing some collection packages using apt but later decided not install and press ctrl+c to kill process when package downloading was a 50%. Now, I want remove all downloaded packages. How can I delete all downloaded packages which are not installed by apt?
1516[12:45:14] <blackflow> uio: more likely it's wtf linux kernel.
1517[12:45:59] <humpled> that's the point where the framebuffer is started right?
1518[12:46:00] <blackflow> uio: you are however trying a rather old, 32-bit hardware. even though debian itself will continue supporting i386 packages, the kernel is in rather bad shape with 32-bit support.
1519[12:46:02] <finn0> colo-work: I already tried it but nothing happens (means no output to terminal)
1520[12:46:06] <uio> blackflow, I don't think so, because there was no wtf Lubuntu, so I think this is Debian's issue.
1525[12:46:35] <jelly> blackflow: where did you get that about bad 32bit support
1526[12:46:38] <uio> blackflow, I see your point, but if it's offered, in my opinion it should work.
1527[12:46:56] <uio> Is it that X is not starting??
1528[12:47:06] <blackflow> jelly: from various mailing lists, GKH himself spoke often about 32-bit in the kernel being like that, and there no being mitigations for meltre
1530[12:47:48] <blackflow> uio: (l)ubuntu will frequently blackscreen on boot after minor updates, though. it really is alla bout the kernel, debian isn't doing anything special to it.
1531[12:48:02] <jelly> noone paid to get 32bit mitigations
1532[12:48:04] <blackflow> anyway, it would be awesome if you could mount your installed root and inspect the logs, could be something trivial to fix.
1533[12:48:20] <jelly> that's quite a different issue from video-intel / i915fb bugs
1534[12:48:44] *** Quits: Kali_noob (~Kali_noob@replaced-ip) (Remote host closed the connection)
1535[12:48:45] <blackflow> jelly: it's not tested enough, that's what the main problem is.
1536[12:48:49] <jelly> uio: which cpu is there?
1537[12:48:49] <colo-work> finn0, that is expected
1562[12:53:54] <uio> blackflow, It looked to me like rescue mode was trying to do an install again, but I'll take another look.
1563[12:54:24] <jelly> blackflow: or even unlock it blindly after framebuffer messes the console up
1564[12:54:39] *** Joins: muhammed (uid375807@replaced-ip)
1565[12:55:10] <jelly> N2600 seems to support 64bit. The question is whether BIOS disables that intentionally
1566[12:55:16] <blackflow> jelly: that's already done, uio said this happens after unlocking root, after systemd diagnostic output and right about when it should modeset
1567[12:55:29] *** Quits: Soo_Slow (Soo_Slow@replaced-ip) (Remote host closed the connection)
1610[13:06:06] <blackflow> uio: it helps because you can mount your installed root and a) inspect the logs, b) chroot and install the firmware as jelly suggested
1611[13:06:14] <uio> Get this: hooking it up to an external monitor gives the correct display!?
1622[13:11:26] <m1dnight_> Hey all. I wanted to install intellij, and figured snap would be the easiest way, because otherwise you have to move it to /opt/ manually and create symlinks all over the place. But now that I installed it through snap, I still have to manually execute /snap/intellij/152/bin/run.sh
1623[13:11:33] <m1dnight_> Did I do something wrong, or is this normal behaviour?
1624[13:11:41] <blackflow> m1dnight_: pycharm?
1625[13:11:59] <m1dnight_> yeah pycharm/intellij both the same concept applies really
1627[13:12:09] <m1dnight_> intellij idea in myh case to be specific
1628[13:12:22] <blackflow> oh raw intellij. k, yeah, you need to create a .desktop for it. on debian, snaps aren't as integrated as they are on ubuntu, last time I checked.
1629[13:12:35] <blackflow> also consider putting /snap/bin in your path
1631[13:12:55] <m1dnight_> Ah! That's something I can workw ith. I was trying to google for an introductory article explaining where snap puts its bin files and whatnot
1632[13:12:59] <m1dnight_> but this is good too. thanks!
1633[13:13:02] <blackflow> m1dnight_: actually, you iirc you don't need to create the .desktop link, just copy it off from /snap/... whereever it installed
1635[13:13:21] <blackflow> put it in ~/.local/share/applications/
1636[13:14:05] <jelly> uio: "lspci -nn" and "lsmod" and dmesg and /var/log/Xorg.0.log will be useful in determining that
1637[13:14:31] <blackflow> m1dnight_: I opted to unpacking to /opt/ myself. I don't like how the snap autoupdates in the middle of me working, and I don't like all the namespace/mount pollution that snapd does.
1654[13:19:35] <blackflow> m1dnight_: not simpler but different -- containerized application delivery platform
1655[13:19:43] <m1dnight_> Hrm, this is an interesting thing. I was playing around with symlinking /snap/bin/intellij to /usr/bin/intellij. /snap/bin/intellij works to start intellij. but then /usr/bin/intellij just shows me this: replaced-url
1658[13:20:33] <blackflow> m1dnight_: yeah don't symlink it. /snap/bin are already symlinks to snap which then resolves the path it was called from, to figure out which snap is called.
1659[13:20:54] <blackflow> bad design but eh... canonical, cant' expect better.
1660[13:21:35] <uio> Okay, I don't get this, I'm trying to install sudo (to install hexchat) and # su apt instal sudo gives 'Media change: please insert the disk labeled 'Debian....' shit
1661[13:22:49] <blackflow> uio: comment out the cdrom deb line in /etc/apt/sources.list, re-run apt update
1745[13:50:10] <blackflow> uio1: okay, try rebooting with this in kernel command line: video=VGA-1:d that should disable the external monitor and likely force the use of built in screen
1746[13:50:47] *** Quits: encod3 (~encod3@replaced-ip) (Remote host closed the connection)
1747[13:51:01] <uio> blackflow, In theory though, should not booting without an external monitor attached also force the use of the laptop screen?
1777[14:02:10] <blackflow> uio: so anyway, there's a number of options for you to try: xrandr, xorg.conf and kernel command line. see different approaches by googling for "xorg disable external vga output". the suggestions I gave above are a summary of some of those approaches. not sure how else I can help with it.
1779[14:03:11] <uio> blackflow, Okay thanks, but in my opinion the issue is not disabling the VGA output, but rather actually getting the laptop display to ... display.
1780[14:03:22] *** Quits: disposable2 (~Thunderbi@replaced-ip) (Remote host closed the connection)
1807[14:10:27] <uio> I dunno, I'm starting to think that Debian is just too niche for people who can't / won't spend weekends to get a useable machine.
1844[14:24:34] <blackflow> DammitJim: ifconfig/ip are commands/tools for configuring the network. ifupdown is a whole framework for persistent (across reboot) network configuration. I think it's actually using ip in the background
1845[14:25:01] <blackflow> (and other things for routing, dns, bonding, bridging, ...)
1859[14:28:48] <ayekat> also, wicd is a network management framework (like NetworkManager, ifupdown, systemd-networkd), while buster is a debian release name
1860[14:28:59] <blackflow> that too :)
1861[14:29:08] <DammitJim> I was just saying that I'm having the issue in buster
1862[14:29:19] <DammitJim> (didn't have the issue with stretch)
1952[14:52:22] <blackflow> DammitJim: I'm not very familiar with wicd but that config doesn't look complete to me, ie. there's no definition of the bond, so perhaps that's why you have the issue you have. question is, why would you use wicd for wired? just set up /etc/network/interfaces that you obviously already have, that obviously works, and all is done.
1953[14:52:22] <ayekat> CrazyTux: what is your goal, actually? I see you've been asking question about "which DE/distro/setup is better" since at least january this year
1954[14:52:40] <Psi-Jack> ayekat: He's been doing this for 3~5 years now.
1964[14:53:39] <blackflow> DammitJim: you mean by "wired_interface = bond0"?
1965[14:54:00] <DammitJim> but then when I come to the office where there is wired and wireless, I'd like to use wired and not wireless and since wicd doesn't get an IP for wired, it puts me in the wireless network
1966[14:54:00] <blackflow> DammitJim: you can use wicd for wireless, doesn't conflict with eth0 being managed by ifupdown
1967[14:54:20] <blackflow> or in your case bond0
1968[14:54:33] <DammitJim> blackflow, bond0 doesn't come up unless I manually run ifdown bond0 && ifup bond0
1969[14:55:00] <DammitJim> btw, I appreciate the help... the wicd channel is almost dead
1970[14:55:07] <blackflow> wicd is (almost) dead.
1971[14:55:10] <DammitJim> and I know this might not be an actual debian issue
1984[14:58:53] <blackflow> DammitJim: then what's the bond for? but what I was aiming at is that perhaps you'd have to be more explicit on which one is master, in active-backup mode. see examples here: replaced-url
2008[15:07:16] <muAdmDev> my default gateway is bound to eth1, if this gateway is not reachable, i'd like to switch to another gateway on eth0. any recommendations, maybe using post-up/down? interfaces currently configured as "auto eth0" and "auto eth1", if allow-hotplug would be better
2039[15:26:22] <trackanddirt> I've read the docs and googled the heck out of this but I am running into issues with keepalived in strech. It will show that it calls on a script but it never executes it. I can run the scripts manually so I know they work. Any Ideas? I have tried changing permissions and still no go. any ideas?
2040[15:27:28] *** Quits: akem (~akem@replaced-ip) (Remote host closed the connection)
2087[16:01:00] <greycat> He clearly didn't come here to ask for help. He came here to announce that he has given up. If he actually wanted help, he would have provided enough *fundamental* background information to even identify what the problem is.
2108[16:05:44] <trackanddirt> scroll up, I posted right before uio. But I'll repeat since the answer is obviously no
2109[16:06:08] <greycat> Imagine you're not an IRC newbie. What you do first is type "/lastlog thatperson" to see all of the things thatperson has said.
2110[16:06:23] <greycat> I showed the result of /lastlog trackanddirt
2115[16:07:33] <trackanddirt> I am using keepalived and although service status is showing that it runs my scripts during state changes, it does not seem to actually execute them. I know they work since I can run them manually. And I am a newbie, so thanks for that shortcut.
2121[16:10:48] <greycat> It only shows things I've actually *received* since I got here. Not stuff that happened 7 hours before I joined.
2122[16:11:28] <trackanddirt> greycat: either way I would apprechiate any help.
2123[16:12:14] <greycat> I know zero about "LVS" which is what the description for keepalived talks about, and I've never heard of LVS or keepalived before.
2125[16:12:40] <greycat> That's why you don't pick a random person in the channel and ask that one person for help. You ask the whole channel instead.
2126[16:12:52] <iulian_> uio, in /etc/default/grub comment these 2 lines replaced-url
2127[16:13:05] <uio> greycat, Oh, I see. You don't have the info. Okay. I did put it on for users like jelly who were here, but can add info again. That would indeed be useful. Okay - did a fresh Debian 9 install on an Asus eeepc Flare series. After disk unlock (LUKS) when the text resizes the screen goes black and is dead. But the external monitor works pefectly.
2128[16:13:17] *** debsan_ is now known as debsan
2129[16:13:53] <trackanddirt> Fair enough
2130[16:14:09] <trackanddirt> If anyone could give me input that would be nice
2138[16:16:42] <greycat> Also make sure you install from a non-free-firmware installer. The fact that you had to add some firmware package(s) manually means you didn't.
2139[16:17:03] <Tenkawa> wow this machine still surprises me at how fast it is
2140[16:17:24] <greycat> With a laptop, you almost always need some non-free firmware. With a laptop line that is *notorious* for being a problem like the eee pc, I can't imagine even bothering to try without it.
2141[16:17:55] <uio> iulian_, GRUB_GFXMODE=yours native resolution is already commented. GRUB_GFXPAYLOAD_LINUX=keep isn't even there.
2179[16:30:03] <somiaj> uio: do we know what hardware you have? Have you tried busters install image?
2180[16:30:15] <greycat> dpkg, no, firmware images is <reply>Unofficial <netinst> images containing non-free Debian <firmware> packages are available from <replaced-url
2186[16:31:50] <greycat> If you can't even get stretch to boot into the text console, I would try the buster NON-FREE rc2 installer (just came out yesterday), and if that doesn't work, either give up, or give MUCH more details when asking for help.
2204[16:37:43] <greycat> I have no idea. I'm just giving general advice that *I* would start with if I had to work on the world most notoriously finicky and unpredictable laptop line.
2209[16:38:06] *** Quits: encod3 (~encod3@replaced-ip) (Remote host closed the connection)
2210[16:38:12] *** Quits: vutral (vutral@replaced-ip) (Remote host closed the connection)
2211[16:38:36] <greycat> If you actually want help, you start with statements like "I'm using an eeepc model number ___ and here's where I've uploaded the lspci -nn output ____"
2245[16:49:55] <dpkg> If you have a question, just ask! For example: "I have a problem with ___; I'm running Debian version ___. When I try to do ___ I get the following output ___. I expected it to do ___." Don't ask if you can ask, if anyone uses it, or pick one person to ask. We're all volunteers; make it easy for us to help you. If you don't get an answer try a few hours later or on debian-user@lists.debian.org. See <smart questions><errors>.
2264[16:55:49] <edgardoj> I am between django-cms and mezzanine
2265[16:56:51] <edgardoj> I just want to start out with a news scraper and go from there, have some django/react boilerplates but read somewhere they are not good for SEO
2266[16:56:55] <jeddi> django's pretty heavy duty - if you know python then look at flask or pyramid first.
2281[17:00:45] <somiaj> edgardoj: nice thing about debian is it is a frozen system, so if you develop a webapp on it, it should be easy to deploy (and one should often develop webapps on the system they plan to host them on). You may also want to consider how often you need to update your webapp. Debian stable has about 3 years of support (and up to 5 once you include LTS support).
2282[17:00:55] <jeddi> there are ready-to-use cms's out there already, aren't there?
2309[17:10:25] <blackflow> I've been using busterino since soft freeze
2310[17:10:33] <greycat> Logg: that's one possible answer
2311[17:11:35] <greycat> for me, it's simply the last nail in su's coffin, and reason enough to drop it entirely and use sudo for everything (sudo -s if you want a shell, sudo -i if you want a login shell, sudo vi foo if you don't want a shell at all)
2316[17:13:33] <Logg> yeah, I've been doing "sudo -i" usually, but, in a new debian install with standard user, you don't get your user in the sudoers file, so you have to "su" at least once, unless you can toggle a box to put yourself in /etc/sudoers now
2317[17:14:01] <Logg> good for systems without sudo
2342[17:24:11] <lolcat-007> what is the actual stable version of debian
2343[17:24:12] <lolcat-007> ???
2344[17:24:26] <Logg> It's still stretch until July 6th-ish
2345[17:25:03] <lolcat-007> ok and what kernel version does debian stretch use
2346[17:25:04] <lolcat-007> ??
2347[17:25:17] <Logg> 4.9, but you can get 4.19 from backports
2348[17:25:19] <jelly> lolcat-007: 4.9.x and there's 4.19.x available if you need it
2349[17:25:36] <jelly> !enter
2350[17:25:36] <dpkg> The enter key is not a substitute for punctuation. Hitting enter unnecessarily makes it difficult to follow what you are saying. Consider using ',', '. ', ';', '...', '---', or ':' instead. If you hit enter too often, you will be autokicked by debhelper for flooding the channel.
2353[17:26:13] <jelly> the question marks can be put at the end of each question, they don't have to be on their own lines and one is usually enough!
2354[17:26:25] <edgardoj> Problem with cli is that you need to log everything if you want to know what are you doing. Nowadays is ctrc+C ctrl+V on notepad for all of the website and cheat sheets.
2355[17:26:46] <Logg> pastebinit is a cool package edgardoj
2356[17:26:58] <lolcat-007> jelly: so after july 6th what will be the latest stable version of debian
2367[17:27:49] <greycat> horribleprogram: because they are different operating systems. Welcome to the diverse and often confusing world of unix.
2368[17:27:55] <horribleprogram> greycat: yup
2369[17:28:12] <edgardoj> I am thinking about pastebinning my processes like install debian > then install openbox > then install x, y and z. I know there is plenty of 'perfect desktops' out there but it's good to keep track of what you do
2370[17:28:12] <horribleprogram> greycat: learning two sets of syntaxes for top and ps for my macOS and Debian
2371[17:28:39] <jelly> horribleprogram: which version of Debian/kFreeBSD do you have?
2372[17:28:39] <blackflow> horribleprogram: use only Debian so you'll only have to learn one.
2373[17:28:56] <BCMM> horribleprogram: they are separate implementations. many of the common `ps` options aren't actually specified by POSIX
2374[17:28:58] <jelly> or did you mean Debian vs. a BSD
2376[17:29:28] <BCMM> i'm not sure of top is specified by posix at all, so it might as well be considered an entirely separate program that happens to have the same name and purpose
2377[17:29:35] <horribleprogram> jelly: Stretch, and I meant Debian and BSD have different implementations of those two commands
2379[17:29:45] <lolcat-007> BCMM: ok i want to know if my dell inspiron 5566 will have support on debian 10 i mean bluetooth support and everything else
2380[17:29:46] <lolcat-007> ??
2381[17:29:49] *** Quits: edgardoj (~edgardoj@replaced-ip) (Remote host closed the connection)
2382[17:29:50] <jelly> horribleprogram: Linux ps inherits and can mix incompatible syntax from both System V UNIX ps and BSD ps
2383[17:30:04] <horribleprogram> jelly: yup, very confusing
2384[17:30:06] <jelly> horribleprogram: I bet "ps aux" works on both OSX and Debian
2385[17:30:09] <BCMM> lolcat-007: if it's working on stretch, it'll work on buster, basically
2386[17:30:10] *** Quits: encod3 (~encod3@replaced-ip) (Remote host closed the connection)
2387[17:30:12] <horribleprogram> it does
2388[17:30:24] <BCMM> lolcat-007: what are you using at the moment?
2390[17:31:04] <jelly> horribleprogram: many Linux tools are bastardized reimplementations of their predecessors from old, dead unices
2391[17:31:05] <BCMM> horribleprogram: most simple shell commands have separate implementations on *bsd and debian (or any linux distro, for that matter)
2396[17:31:32] <greycat> stretch to buster seems like it's not going to be a disruptive upgrade for most people, other than the changes to the su command
2397[17:31:33] <BCMM> bsd doesn't want those for licensing reasons
2399[17:31:48] <horribleprogram> BCMM: well you can install them
2400[17:31:49] <BCMM> (and at the time they were first developed, bsd wasn't as liberally licensed as it is now)
2401[17:32:07] <lolcat-007> BCMM: right now im using ubuntu 18.04 bionic because i tried debian stretch like a years ago and i didnt have bluetooth support on my dell that is why i asking now
2402[17:32:12] <BCMM> horribleprogram: from ports, sure. but they don't want them as a core part of the system, due to the licensing.
2403[17:32:23] <horribleprogram> yup
2404[17:32:38] <BCMM> lolcat-007: if bluetooth worked in ubuntu but not in debian, that's probably just because your bluetooth adaptor requires proprietary firmware
2406[17:33:01] <BCMM> lolcat-007: and debian won't install any proprietary software unless you explicitly tell it to, by enabling non-free
2407[17:33:10] <BCMM> !non-free
2408[17:33:10] <dpkg> [non-free] a component which contains software that does not comply with the <DFSG>. To add non-free packages to your packages index, ask me about <non-free sources>. To see which non-free packages are installed ask me about <non-free list>.
2428[17:37:46] *** Quits: horribleprogram (~horriblep@replaced-ip) (Quit: Where I came from the Great Wild 'n shit, where you can get shot if you crack smiles and shit...)
2430[17:37:51] <BCMM> lolcat-007: "firmware" is the software that runs on the microcontroller inside your bluetooth adaptor. it is distinct from the driver, which runs on your CPU and communicates with the firmware
2431[17:38:19] <lolcat-007> BCMM: ok i get it so the proprietary firmware are also free to use or i have to paid i belive gnu/linux is free to use right???
2432[17:38:44] <uio> lolcat-007, Generally speaking, Debian is not best for newer users. It might work, but it will likely take much more time.
2433[17:38:59] <BCMM> lolcat-007: they are available free-of-charge. you just don't necessarily have the right to modify them, and there may be restrictions on how you are allowed to copy them, etc.
2434[17:39:01] *** debhelper sets mode: +l 1524
2435[17:39:03] <jelly> lolcat-007: yes it's free to use. It just has different licensing.
2437[17:39:18] <BCMM> lolcat-007: the firmware comes in the form of a file provided by the manufacturer, which has to be sent to the bluetooth device every time it powers up. on windows, the firmware would be bundled with the driver.
2438[17:40:45] <BCMM> lolcat-007: but the manufacturer doesn't provide any details on how they actually made that file, so if you ever wanted to modify it, you're out luck. it isn't allowed to be part of the main debian archive because debian developers and users aren't allowed to fix it if it's broken.
2439[17:41:03] <uio> BCMM, Have you ever wanted to modify firmware?
2440[17:41:10] *** Quits: factor (~factor@replaced-ip) (Remote host closed the connection)
2441[17:41:40] <BCMM> uio: not personally, no. but if there was a security problem with a wireless chipset, for example, we would just be stuck hoping that the manufacturer fixes it
2459[17:43:34] <BCMM> of course, the really frightening thing is all the firmware that *isn't* visible, because it's stored in a ROM inside your computer, not shipped as a binary blob with the driver software...
2460[17:43:44] <lolcat-007> BCMM: ok now i understand so what i have to do in order to make the bluetooth work on debian is remove those bluetooth default driver and install the proprietary firmware that is right???
2461[17:43:46] *** Quits: xcm (~xcm@replaced-ip) (Remote host closed the connection)
2462[17:43:53] <BCMM> like vulnerabilities in the actual USB controller, for god's sake...
2464[17:44:01] <BCMM> lolcat-007: you don't need to remove any driver
2465[17:44:17] <BCMM> lolcat-007: the *driver* is probably Free software. you just need the firmware file.
2466[17:44:17] <uio> I'm kinda sad, but I must admit that I'm in the process of installing Lubuntu on the eeePC. I just couldn't afford to spend another day trying to get Debian to work. Alas. I still have Debian on my Thinkpas X61.
2467[17:44:29] <jelly> lolcat-007: negative. You use default driver, AND the non-free firmware.
2468[17:44:48] <BCMM> lolcat-007: (if you want to check this, you can look at the output of `dmesg`. you will probably see the driver load, complain that it can't find any firmware, and then exit.)
2470[17:44:53] <jelly> uio: that's just fine, use what works.
2471[17:45:13] <uio> jelly, Yeah, I mean at least it's based on Debian I guess.
2472[17:45:17] <blackflow> uio: and don't forget taht's teh last lubuntu you'll be putting there. they've dropped 32bit installers.
2473[17:45:21] <BCMM> lolcat-007: just add non-free to your sources, do `apt update`, do `apt install firmware-linux-nonfree`, and reboot
2474[17:45:23] *** Quits: vutral (vutral@replaced-ip) (Remote host closed the connection)
2475[17:45:25] <jelly> cousin twice removed
2476[17:45:34] <uio> jelly, lol.
2477[17:45:48] <uio> blackflow, What!!!
2478[17:45:49] <lolcat-007> BCMM: ok like i said before i new in linux so that will be kind of advance users but i will give it a try and maybe you guys can help me with that right??
2479[17:45:54] <blackflow> so they can have sex? :) lubunu linux debian edition
2480[17:46:01] <jelly> blackflow: N2600 has 64bit support
2481[17:46:09] <blackflow> oh cool then
2482[17:46:14] <BCMM> lolcat-007: well, i'm logging off now. but i'm sure people here can help you enable non-free
2483[17:46:21] <greycat> !firmware images
2484[17:46:22] <dpkg> Unofficial <netinst> images containing non-free Debian <firmware> packages are available from <replaced-url
2485[17:46:27] <greycat> If you have a laptop, just install with one of those.
2487[17:46:35] <jelly> no idea if the BIOS disables x86_64 support, tho
2488[17:46:52] <BCMM> lolcat-007: (they might help more willingly if you can limit the number of question marks at the end of any single question somewhat, though)
2489[17:47:03] <blackflow> worth checking out, because if it doesn't, then lubuntu 18.04 is the last one supported.
2490[17:47:09] <friendofafriend> I'm running Buster, I see no /etc/rc.local. I want a command to run once. I'd like to avoid getting Poettering'd as much as I can. What should I do?
2491[17:47:19] <BCMM> greycat: do those images automatically add non-free to the installed system's sources.list, by the way?
2492[17:47:20] <greycat> !rc.local
2493[17:47:20] <dpkg> /etc/rc.local may be used to run simple commands at boot time. It exists by default in jessie or older; in stretch you need to create it. Don't forget the <shebang> and be sure to chmod 755 it. rc.local is considered a hack, a stopgap, or a temporary band-aid; see <systemd>
2494[17:47:22] <jelly> blackflow: supported until 2023
2495[17:47:26] <uio> blackflow, For another two years.
2496[17:47:31] <blackflow> yeah
2497[17:47:34] <BCMM> greycat: or do they just provide firmware for use during the actual installation?
2498[17:47:38] <friendofafriend> !thanks greycat
2499[17:47:38] <dpkg> sure thing, friendofafriend
2500[17:47:55] <greycat> BCMM: not sure. I don't laptop.
2509[17:49:09] <uio> greycat, Or, a conference, or a meeting or some such? Classes?
2510[17:49:16] <somiaj> uio: sometimes seeing why something works on another distro can help get debian to work, you could also do a debootstrap install if it is just the installer causing problems.
2511[17:50:27] <uio> somiaj, The distro worked beautifully... just the screen wouldn't work. Had to use an external monitor. But that issue could'nt be solved without some other non free testing other cd stuff. So I just decided that life is short and downloaded Lubuntu.
2519[17:52:34] <jelly> iulian_: they spent two days on this
2520[17:53:15] <uio> lol, but maybe Lubuntu will have the same issue...
2521[17:53:18] <iulian_> uio and after editing it don't forget to run sudo update-grub
2522[17:53:25] <jelly> not saying we're already sick of hearing about it but...
2523[17:53:37] <uio> jelly, lol
2524[17:53:52] <uio> jelly, What's the longest lasting problem?
2525[17:54:03] <uio> Sorry, off-topic...
2526[17:54:32] <jelly> there are bugs in Debian's bug tracking system opened more than a decade ago
2527[17:54:45] <uio> lol
2528[17:54:55] <uio> slow and steady.
2529[17:55:31] <lolcat-007> greycat: ok can yoou give me a little help choosing the right image of debian 10 cause there are a lot of directory there i choose firmware-buster-DI-alpha1-amd64-dvd-1.iso that one is ok for me???
2530[17:55:44] <greycat> you want rc2
2531[17:55:56] <blackflow> probably means the problem is not big enough for someone to shell out enough resources (of temporal or financial persuasion) to fix it.
2532[17:56:01] <jelly> or latest weekly if rc2 is more than a month old
2533[17:56:14] <lolcat-007> greycat: rc2?? what is that??
2534[17:56:28] <greycat> release candidate 2 of the buster installer
2535[17:56:34] <greycat> it's on the first page
2536[17:56:39] <greycat> !firmware images
2537[17:56:40] <dpkg> Unofficial <netinst> images containing non-free Debian <firmware> packages are available from <replaced-url
2538[17:57:09] <greycat> subdirectory named buster_di_rc2+nonfree/
2539[17:57:14] <friendofafriend> I think that's replaced-url
2558[18:02:02] <jelly> huh, so apparently that dvd-1 has all of: task-desktop_3.53_all.deb task-cinnamon-desktop_3.53_all.deb task-gnome-desktop_3.53_all.deb task-kde-desktop_3.53_all.deb task-lxde-desktop_3.53_all.deb task-lxqt-desktop_3.53_all.deb task-mate-desktop_3.53_all.deb task-xfce-desktop_3.53_all.deb
2559[18:02:31] <lolcat-007> alright good, so i can start downloading debian 10 buster and installing on my system right now and i dont have to wait until july 6th??
2560[18:02:47] <greycat> Correct.
2561[18:03:03] <uio> buster will be stable soon?
2562[18:03:14] <jelly> lolcat-007: you do not have to wait, but the system you install right now and the installer itself might have more bugs than the final release
2579[18:07:04] *** Quits: noosanon (~noosanon@replaced-ip) (Remote host closed the connection)
2580[18:07:23] <uio> Does this mean I'll have to leave stretch soon?
2581[18:07:42] <lolcat-007> ok
2582[18:07:44] <lolcat-007> thank guys
2583[18:08:08] <jelly> uio: a release keeps getting security updates about a year after the next one is released
2584[18:08:17] <jelly> for* about a year
2585[18:09:18] <greycat> Stretch will receive full security support for at least 1 year (sometimes more), but reduced support which is called "LTS" for a longer period.
2610[18:16:23] <jelly> lolcat-007: everything from Debian 7 onwards has had some form of LTS support, but that is a separate subproject of Debian and depends on donations
2612[18:17:01] <somiaj> lolcat-007: LTS is a team of volunteers that get funding from various sources who want server side support for longer than debian provides it.
2613[18:17:07] <jelly> if donations suddenty dry out _nothing_ gets LTS
2614[18:17:11] <jelly> !lts
2615[18:17:11] <dpkg> Debian Long Term Support (LTS) is a project to extend the lifetime of all Debian stable releases to (at least) 5 years. Debian LTS is not handled by the Debian security team, but by a separate group of volunteers and companies. Ask me about <jessie-lts> and see replaced-url
2616[18:17:15] <lolcat-007> ok but i can be sure i will have a least two years of debian 10 support
2617[18:17:17] <lolcat-007> ??
2618[18:17:47] <jelly> lolcat-007: you can be sure, if you or some other company keeps donating enough :-)
2619[18:18:07] <somiaj> and if you use debian for desktop use, that is not included in LTS (and sometimes debian stops supporting things during that year of old stable, for example jessie lost support for firefox about 6-9 months after stretches release)
2620[18:18:34] <jelly> somiaj: LTS support is not strictly limited to "server side"
2621[18:18:37] <somiaj> lolcat-007: debian will offically suport stretch for about 3 years (as both stable and one year as oldstable)
2647[18:29:03] <pyfgcr> Primer: it will work in debian, sooner or later, but not necessarily in debian 10 after the first release
2648[18:29:14] *** Quits: obs2 (58736e9d@replaced-ip) (Remote host closed the connection)
2649[18:29:32] <Primer> please. Use the correct terminology
2650[18:29:51] <Primer> It won't be 'available' in Debian, but it'll surely work if you put it there
2651[18:30:03] <Primer> getting it there is another story
2652[18:30:19] *** subopt is now known as _subopt_in_repos
2653[18:30:20] *** Quits: _subopt_in_repos (~subopt@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
2654[18:30:31] <jelly> Primer: that's not really correct; Canonical does tests on more hardware than Debian, and they easily enable non-free drivers where needed
2655[18:30:41] <pyfgcr> Primer: not sure I get the difference
2656[18:30:59] <pyfgcr> * I'm not
2657[18:31:21] <jelly> variations in versions, build options and default settings also affect OOTB experience
2658[18:31:32] <Primer> You make it sounds as if some things are mutually exclusive between debian and ubuntu. To suggest one software works in one but not the other, is simply false
2659[18:31:56] <jelly> it's hardware support we're discussing
2660[18:32:12] <Primer> yeah, and that's even less of a distinction
2661[18:32:21] <Tenkawa> Primer: what????
2662[18:32:28] <yvyz> What the fuck
2663[18:32:35] <yvyz> Excuse my language. Apologies
2666[18:32:49] <Primer> getting hardware working is the easiest thing
2667[18:32:57] <yvyz> How much of Debian do you think is in Ubuntu?
2668[18:33:00] <jelly> Primer: okay, stay here for a couple weeks and see how many people show up with something that works on Ubuntu but fails on Debian and vice-versa
2669[18:33:02] <Tenkawa> hardware is heavily dependent on library calls and kernel code
2670[18:33:16] <Primer> jelly: been here, done that
2715[18:38:48] <pyfgcr> Primer: when you say that hardware is heavily kernel depentent => (works in ubuntu => works in debian), you are suggesting it's normal practice for debian users to install ubuntu kernel patches?
2716[18:38:57] <yvyz> I guess it just confusing when people out right suggest that Ubuntu upstreams 100% of Debian. Then assumes that all that works on Debian shall work on Ubuntu and vica versa
2719[18:39:29] <yvyz> Which is a strictly incorrect statement, and displays a serious lack of knowledge of the person who said it.
2720[18:39:30] <pyfgcr> so can you please clarify to me?
2721[18:39:33] <jmcnaught> I would say that if a piece of hardware works in Ubuntu it's extremely likely it will also work in Debian. Some things might require a little effort or waiting for a backported kernel. In some cases the hardware might need a library (like mesa or xorg stuff) that might not get backported. But hardware support is the same across distros is the rule not the exception.
2723[18:39:42] <somiaj> also debian's stock kernel is often older (espcially near the end of a release cycle) and newer hardware is just not supported on it causing extra work to get it to work.
2724[18:39:52] <Primer> To say that hardware relies on the system libraries is false.
2738[18:40:54] <petn-randall> Primer: It really depends. X11 has user space drivers for the graphics cards (xserver-xorg-video-* packages) that also need to support the graphics card for it to work sufficiently.
2739[18:40:55] <pyfgcr> jmcnaught: I mainly agree, but we were specifically tolking about tochscreens in the beginning, and there the matter is different
2740[18:40:56] <yvyz> I actually hand fed my linux.
2741[18:41:01] <Primer> jmcnaught: and what was that?
2745[18:41:33] <jelly> jmcnaught: and Linux _still_ does not have power/thermal sorted out completely for SKL and newer
2746[18:41:47] <somiaj> the xserver-xorg-modesetting driver is trying to help on this front, but not all kernel modules are fully kms and compadabile with it.
2747[18:41:48] <Primer> petn-randall: all of that is ultimately driven by the kernel. You don't drive anything in userspace without a kernel.
2748[18:41:58] <Primer> somiaj: mesa gets you features, it's not driving any hardware
2749[18:42:09] <jelly> Primer: this channel covers stable mostly, and versions of software in stable will often make new hardware problematic
2750[18:42:21] <petn-randall> Primer: Sure, but even if your kernel module supports your GPU, you still need the corresponding user space drivers for X11.
2751[18:42:38] <Primer> jelly: sure, but to say that hardware is 'not supported' has to be taken with a grain of salt
2752[18:42:39] <yvyz> Also, the nvidia user space driver helps drive the kernel module and for things like how Wayland use currenty battling a lack of EGLStreams support, and xwayland has been co-opted to handle the kernel parameters... you would be compeltely incorrect in all of your statements so far.
2753[18:42:43] <jelly> Primer: just that alone makes "if it works in Ubuntu it'll work on Debian" problematic
2754[18:42:57] <Primer> it's really up to the person and what they're willing to do. for example, nvidia 430 is not "supported" on testing
2755[18:42:58] <jelly> Primer: "can be made to work" I'd accept
2756[18:43:02] <Primer> yet, I have it installed
2757[18:43:12] <petn-randall> I thought we were talking out of the box hardware support?
2766[18:44:41] <somiaj> but in years here, I've seen various issues with getting a bleeding edge gpu to work in debian, more than just say kernel+firmware from backports.
2820[18:54:21] <jmcnaught> I have a touch screen on my laptop that I disabled with a udev ATTR{authorized}="0" rule. If I wanted to be able to toggle it on/off, are there any side effects from running "udevadm control -R && udevadm trigger" that I should know about, or is there another way I could do it?
2821[18:54:31] <jelly> Primer: if it takes, say, a whole newer stack of kernel/drm/mesa/xorg to drive their stuff, then picking a distro that already has it is saner
2822[18:54:45] <Primer> jelly: indeed
2823[18:54:52] <yvyz> Well, its a pretty great way of pointing out how an entire architecture bug fails to allow Jessie and really a load of linux kernels to install. petn-randall and it is clear that I would be arguing this point to someone who is crass enough to believe that linux will install on any hardware.
2839[18:58:43] <Primer> It was a precious point that needed to be made
2840[18:59:28] <hmuller> I recently upgraded my BIOS with no chance of reverting. Nothing horrible has happened, but I no longer get sound output over HDMI. pavucontrol no longer shows HDMI in the configuration tab. where do I start looking?
2860[19:02:52] <hmuller> jelly: been using testing, buster I think. which is 4.19
2861[19:03:12] <rant> you may also need amd firmware if that wasnt already covered, I have a machine with an AMD SoC and it requires firmware for audio over dp/hdmi
2862[19:03:33] *** Quits: Zppix (uid182351@replaced-ip) (Quit: Connection closed for inactivity)
2863[19:04:00] <Primer> he said it was working before a bios upgrade though
2889[19:09:12] <Primer> jelly: I mention it because it seems that ACPI is a factor in a lot of things in modern hardware
2890[19:09:20] *** Quits: vutral (~vutral@replaced-ip) (Remote host closed the connection)
2891[19:09:24] <Primer> I've had to specify apci boot params to get certain hardware working
2892[19:09:55] <jelly> yes, and sometimes the firmware only enables things if you make the OS pretend to be a specific Windows generation
2893[19:10:00] *** Parts: andre20021 (~andre2002@replaced-ip) ("Once you know what it is you want to be true, instinct is a very useful device for enabling you to know that it is")
2894[19:10:11] <jelly> or provides different interfaces
2895[19:10:12] <Primer> yup, apci="windows 2009"
2896[19:10:22] *** Quits: rany (~rany@replaced-ip) (Remote host closed the connection)
2897[19:10:23] <Primer> is what I've been having to use. My touchpad still has issues
2898[19:10:32] <jelly> that might have been acpi.osi="Windows 2009"
2949[19:19:08] <Primer> It's a concept that it's fully grasped in many conversations I have about Linux. This gets asked all the time in #ardour: Will _so and so USB sound card_ work in Linux?
2954[19:19:57] <Primer> The only 'driver' you'll need is snd_usb_audio
2955[19:20:18] <Primer> You will never need a driver that is specific to that one model of sound card. This is my point.
2956[19:20:19] <yvyz> You'd be great in #freebsd
2957[19:20:49] <Primer> I'm sure you are
2958[19:21:13] *** Quits: vutral (vutral@replaced-ip) (Remote host closed the connection)
2959[19:21:20] <mutante> i guess as long as stuff that isn't following the specs also works on Windows people are not going to care about specs.. they will just ask why it doesn't work here when it can "just work" there
2970[19:23:27] *** Quits: vutral (~vutral@replaced-ip) (Remote host closed the connection)
2971[19:23:58] <Primer> Then imagine if such a spec exists, and hardware implements it. You plug said hardware in, and yes, it Just Works™ (this is a concept. This already exists in USB)
2977[19:25:32] <jelly> Primer: you really, really don't want to look at the zillion quirks to make usb, pci, pcie hw actually work, in linux kernel source
2978[19:25:51] <Primer> The main reason this crap doesn't exists now is because hardare manufacturers want to compete and make money. ubiquiti is not conducive to such business models.
2981[19:26:51] <Primer> jelly: heh all I said was there is a spec and that the spec is implemented in all modern OSes. For a piece of hardware to be supported in any of these OSes, it simply needs to follow the spec. What it took to get that spec to work in each OS...not relevant
2982[19:27:20] <yvyz> man, this channel.
2983[19:27:27] <Primer> Sure, if your device has a quirk...it likely has that quirk in every OS
2984[19:27:34] <jelly> what you said discounted the thing we call reality
2985[19:27:38] <Primer> yvyz: you're learning new things from it?
2986[19:27:50] <Primer> jelly: how so?
2987[19:28:48] <GenTooMan> I always liked this quote "reality like it or leave it!"
2988[19:28:49] <Primer> I'm being 100% serious. I'm talking about USB audio devices, specifically, something I'm dealing with on an almost daily basis when using ardour, the free and open source digital audio workstation
2989[19:28:57] <jelly> Primer: drivers follow actual hardware, not the other way round
2990[19:29:14] <Primer> jelly: I'm talking specifically about the usb audio spec
2991[19:29:26] <yvyz> Not at all. Come over to oftc and join the mm and knewbies and learn how linux is developed. You are just not really a nice person. And this channel has gone extremely offtopic with some sensational fear of telling each other to get back to just chilling > debating the nuances of your intellect.
2992[19:29:29] <jelly> Primer: I'm talking about almost everything else.
2993[19:29:30] <Primer> this spec being the "universal" _driver_ for such things
3013[19:33:04] *** Quits: Zvmdyv (~Zvmdyv@replaced-ip) (Remote host closed the connection)
3014[19:33:19] <Primer> the USB spec for audio, block storage, video...were a HUGE leap in making computer hardware more open. Hardware manufacturers don't like openness.
3018[19:33:59] <Primer> So yeah, go read up on these specs, and consider what I wrote about it in the context of this conversation, and get back to me. I'll be here all day.
3075[19:52:52] *** Quits: dtux (~dmtucker@replaced-ip) (Remote host closed the connection)
3076[19:53:18] <jelly> Dragone2: could you please disable that away nick setting? irc has a native away flag, you can use that and most clients show it.
3077[19:53:46] <jelly> Dragone2: it just adds to noise in large channels like this
3160[20:37:24] <bernyrd> hello, very serious issue. I can log in. I know I use right password, am not told password wrong. But it looks like I log in, but then I do not log in, am back at prompt again.
3161[20:37:29] <bernyrd> This also I think affect x2go session
3162[20:37:32] *** Quits: traveltissues (~traveltis@replaced-ip) (Remote host closed the connection)
3163[20:37:37] <bernyrd> got around it with xvfb-run for a while
3164[20:37:38] <bernyrd> bu no
3165[20:38:29] *** Quits: vutral (vutral@replaced-ip) (Remote host closed the connection)
3217[20:57:20] <Logg> i did it, i made it to debian buster. It was not smooth lol. lightdm broke and had to be reinstalled. cryptsetup broke in initramfs so i had to chroot the system from livecd and set it up again,... kaffeine & k4dirstat and all the kde stuff I just removed because there was some conflict. My font rendering looks bad now, so I have to figure that one out. AwesomeWM always breaks their stuff on purpose, I think because they’re bored, so
3218[20:57:20] <Logg> I had to diff between their rc.lua and my rc.lua, awesome 4.1 to 4.3. update-initramfs complained that amdgpu vega20 firmware was missing, so I copied them from the linux-firmware git. But now I’m on debian buster so that’s pretty neat
3223[20:58:32] *** Quits: vutral (vutral@replaced-ip) (Remote host closed the connection)
3224[20:58:51] <Logg> I did actually find a bug report for the cryptsetup thing, but the dude closed it as not reproducible. I'm always weary filing bug reports that aren't detailed enough, or when it could be from something dumb I did
3225[20:59:27] <Logg> like if i wanted to report the lightdm thing i'd say "lightdm broke" right? like, that doesn't do much for them
3226[21:00:17] <Logg> I think it was because I said "go ahead and install maintainers version" and then before apt was finished installing it, i went back into the config and put my startup script line in
3227[21:00:25] <Logg> may have broken it
3228[21:01:14] <greycat> Be more detailed. Did it break during the package upgrade? If so, copy the error messages out of your script(1) log of the session. Does it fail to start when you boot, but startx works fine? Does it run, but you can't login? And so on. Be specific. Be detailed.
3229[21:02:07] <areyouloco> Hi I am trying to get remote desktop of debian via vnc. I followed this: replaced-url
3230[21:02:14] <areyouloco> i know it has something to do with proper $HOME/.vnc/xstartup but I cannot make gnome to work
3232[21:02:25] <greycat> If you think it's because of a customized configuration file, that's also useful information. Explain what customizations you made, and note the error messages in ~/.xsession-errors or wherever lightdm logs stuff. I know very little about DMs.
3233[21:02:38] <karlpinc> Yes, filing bug reports is a pain. But in some sense it's the price you pay for using FOSS.
3237[21:04:11] <karlpinc> (You often get the satisfaction of having your bugs actually fixed, assuming you file an adequate bug report. Unlike the usual case in the proprietary world.)
3238[21:05:50] <NetTerminalGene> hi guys. i encountered a bug on both gnome 3.28 and 3.30 on debian. if you copy and paste multiple files, after the first file paste, indicator freezes, but it copies in the background anyways. can you confirm?
3239[21:06:33] <areyouloco> nevermind I got this working now
3240[21:07:20] *** Joins: wilson (~wilson@replaced-ip)
3273[21:14:12] <somiaj> NetTerminalGene: you are welcome to report a bug, espically if you have a way to ensure the devs can reporduce it, also may want to look at upstream for this (debian may not try to fix this indvidual issue, but the matainer may forward the problem to upstream if it can be reproduced)
3274[21:14:19] *** Alina-malina_ is now known as alina-malina
3275[21:14:24] <Primer> Lady_Aleena: You will forever be known to be as such :)
3317[21:27:39] <Primer> all unknown commands should look scary
3318[21:27:41] <debianuser> NetTerminalGene: Are there any hyperthreading-specific vulnerabilities? All "vulnerabilities" I know are usually the result of cpu cache existence. Hyperthreading may make the exploit easier, but it's not the cause. The cause is the cpu arch, you can't fix that easily.
3319[21:28:21] <NetTerminalGene> debianuser, i think openbsd disabled it by default
3321[21:28:38] <Primer> I thought they were the speculative execution issues, not cache existence
3322[21:29:21] <debianuser> NetTerminalGene: Luckily, usually you don't have to, as those "vulnerabilities" usually only allow one running process to guess partially what the other running process is doing, and that's usually not the issue, unless you're hosting VMs with a finance-critical encryption in one VM and a scary untrusted virus in another one. :)
3353[21:35:11] <Primer> What you heard was probably something like "If this gets any worse, we'll simply have to disable HT". There are still currently ways of mitigating the issues, so there's no need to go there yet.
3354[21:35:41] <Primer> Read the changelogs that have been linked.
3399[21:49:38] <altker128> Hey guys. Kind of a dumb UNIX folder/file permissions question here. I have a folder (i.e. ROOT_FOLDER) and inside there's a directory which I want to grant permission to a specific user (i.e /ROOT_FOLDER/SUB_FOLDER) ; I have permissions set on SUB_FOLDER but I guess because the user doesn't have execute permission on ROOT_FOLDER then, it won't have access to SUB_FOLDER?
3400[21:50:08] <greycat> you need +x permissions for that user on ROOT_FOLDER, either "other", or a group that the user is in
3401[21:51:03] <donofrio> what do I need to do to get my .deb locally copied to be read like a local mirror? replaced-url
3419[21:57:36] <leeward> I have a product I want to ship that runs Debian, and I have some specific configuration that I want to keep on all instances of the product. Is there a preferred way to do that? A .deb package that overrides other packages' config files seems...weird.
3427[21:59:53] <leeward> karlpinc: I can set them at install time, but that doesn't give me a good way to manage them in the future (and in the field).
3447[22:06:04] <jim> debianuser, hi... I screwed up my cadence install, I think by doing an autoremove of something needed by cadence but missing from the depend list of the package
3448[22:06:20] *** Quits: vutral (~vutral@replaced-ip) (Remote host closed the connection)
3449[22:06:37] <altker128> Is there anything that lets you automatically diff what you use to configure something and helps create the ansible scripts?
3467[22:15:03] <bernyrd> starting from nothing and creating comprehensive set of changes is only way I have seen people 1) catch every chagne impacting the sw and 2) understand what they have actually set up
3473[22:17:42] <mutante> altker128: build a .deb that applies the patches and then let ansible or puppet install a package?
3474[22:17:49] <donofrio> I'm stuck till I can figure out how to apply the arcive_2019.key or something to get this ports mirror to work on my powerpc
3475[22:18:04] <leeward> Hmm, ansible looks like it might work, but seems like a centralized push type system, where I want something more similar to apt with distributed nodes that can pull configuration at will.
3476[22:18:15] <mutante> puppet?
3477[22:18:36] <leeward> I probably don't want to maintain a central inventory of all these things.
3484[22:20:01] <blackflow> SaltStack is closer to ansible in concepts and syntax, might wanna take a look at that before Puppet
3485[22:20:02] <debianuser> jim: `apt-get install -f` should probably get you all the missing dependencies...
3486[22:20:12] <altker128> mutante: I don't know what the benefit of having a '.deb' that applies the patches vs. using Ansible / Puppet to do that, if you're going the Ansible/Puppet route
3487[22:20:14] <karlpinc> leeward: And see if you can pull with ansible. (I've no clue, but it always seemed the most flexible of those sorts of systems.)
3488[22:20:27] <altker128> mutante: Or, just keep the patch and run a script that applies it once via SSH
3489[22:20:41] <blackflow> leeward: sorry, didn't hilight you, check out my comment about saltstack just above
3490[22:21:19] *** Quits: Shahnaz_ (~Shahnaz@replaced-ip) (Remote host closed the connection)
3492[22:22:13] <LinuxGuy2020> Hello I am trying to combine 3x 120GB SSD with RAID 0. I created a 2GB EFI partition, a 2GB /boot partition, 16GB swap at the end of the drive , the rest is for / partition. The only partition that is created on a RAID device is the / partition. It will boot just fine. The problem is when I try and do a dist-upgrade, it winds up breaking GRUB on next reboot. Is this normal or is this simply a current bug in unstable?
3496[22:22:55] <mutante> altker128: oh, it seemed to me that creating all the patches and dumping them into a package would be easier than having to tell ansible about a bunch of individual files and their contents and that somehow it would be "cleaner" to have the config management just install packages .. but whatever works
3498[22:23:01] <donofrio> anyone know how to get archive_2019.key reconised on my system?
3499[22:23:04] <altker128> LinuxGuy2020: It might be cleaner to make your boot/OS volumes onto a non-RAID drive
3500[22:23:32] <greycat> donofrio: apt-key add? I have no idea what this file is, so I'm guessing it's an APT archive signing key or whatever they're called.
3501[22:23:43] <mutante> altker128: the "patch via ssh" part seems nice until you have more than 1 server and 1 distro version
3502[22:24:11] <altker128> mutante: You might be right. I still struggle to see what Ansible/Puppet "adds" if you're installing basic software packages, and then changing the config files (which is pretty much what you have to do to configure some software)
3503[22:24:46] <mutante> altker128: it becomes more obvious if you have a larger number of servers and more than 1 server per "role"
3504[22:24:47] <altker128> mutante: How does Ansible/Puppet help in that regard? Basically as a user you need to "tokenize" configuration parameters, and then pass them in from some top-level
3505[22:25:22] <mutante> basically it's that you only do things once and don't repeat yourselves when doing it on 1000 boxes
3506[22:25:26] <altker128> I know that if you wanted to say "mass apt update / apt upgrade" that Ansible/Puppet helps with that
3507[22:25:36] <mutante> and that you have all the history and state in a repo
3508[22:26:00] <mutante> which is much nicer than "our admin guy did magic and now there's the new guy"
3509[22:26:07] <altker128> I get how they help with maintaing systems, but I am still not sure about configuration
3510[22:26:45] <mutante> they dont help with configuring services.. it's the same work to edit that config file once .. i would agree
3511[22:26:59] <mutante> either you edit it directly or a template for it.. but at least you do it only once
3512[22:27:00] <karlpinc> altker128: It helps a lot to know which, of the big pile, configuration files are frobbed on any given system. And what was changed.
3513[22:27:24] <tds> altker128: as an example, I've got 3 identically configured reverse proxies at the edge of my network, I have ansible take in just a list of hostnames then generate the config from that, and deploy it to all of them
3514[22:27:40] <tds> and of course when you scale it up to "real" infrastructure rather than just personal stuff, it makes even more sense :)
3518[22:28:00] <karlpinc> altker128: Or, I want to rebuild this box. What config files do I need to touch?
3519[22:28:13] <altker128> karlpinc: What that's my very question -- how do you resolve that?
3520[22:28:34] <altker128> tds: Did you end up manually scripting the part that takes in list of hostnames ---> modify configuration files?
3521[22:29:19] <tds> altker128: it's a jinja2 template that gets rendered out to a full config
3522[22:29:28] <karlpinc> altker128: I don't know. I just hear that people use ansible to do this. I'd imagine it has to do with what configs get automaticaly pushed to what VMs (etc.).
3523[22:29:37] <bernyrd> mutante: actually I like that, using .deb
3524[22:29:56] <leeward> thanks karlpinc and blackflow, will look into those more
3525[22:30:02] <tds> as an example:
3526[22:30:02] <tds> {% for site in sites.sites.keys() %}
3531[22:30:30] <altker128> tds: Ah, so you tokenized/templatized a given configuration file's syntax, and then modify it from a script
3532[22:30:39] <leeward> Incidentally, Puppet's really bad at explaining how it works. It's got a section on its web site entitled "How it works" but it just talks about outcomes not process.
3533[22:30:52] <bernyrd> mutante: but, I found agentless to be better, you just need way to properly manage roles
3534[22:30:52] <tds> altker128: yeah
3535[22:31:01] <karlpinc> tds: I'd use m4. Just saying. :)
3536[22:31:14] <tds> m4?
3537[22:31:49] <tds> oh, heh
3538[22:31:49] <altker128> Oh man
3539[22:31:51] <altker128> m4
3540[22:31:51] <karlpinc> tds: Instead of jinga2. It's a macro substitution system.
3541[22:31:53] <bernyrd> macro language, use by people with long flowing beards.
3589[22:43:26] <altker128> So, for example, if deploying an application needed A) postgres B) nginx C) sendmail D) dovecot , you're saying Puppet doesn't help there?
3590[22:44:01] <jpw> it can, it's nowhere near as graceful as some other tools
3595[22:45:54] <jpw> that's the combo we use (we are mode 1). puppet to look after the OS and ensure it's monitored and integrated with essential services. then deploy the application on top with ansible.
3596[22:46:37] <altker128> jpw: So, Ansible to install/configure packages, Puppet to keep the systems up to date and monitor them?
3599[22:47:38] <zoredache> altker128: or just ansible for everything.
3600[22:48:08] <jpw> not quite. take a kubernetes cluster for example. each server must have its time synchronised, integrated with active directory for login and rbac, snmp server installed for baseline monitoring. this is handled with puppet. kubernetes is then deployed on top using ansible.
3601[22:49:18] <jpw> zoredache: in an ideal world yea. unfortunately through experience individual teams have not taken requests to change things like community strings seriously so we needed another way to manage them
3602[22:49:20] *** Quits: subopt (~subopt@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
3638[23:12:54] <donofrio> or anyone else that can help me get my local .deb packages installed inthe correct order to get apt updates to work from debian-ports mirror
3645[23:14:44] <sloshy> and make the minecraft deb package not work anymore
3646[23:14:44] <donofrio> what paste do you prefer cause I'm just trying to get info to you good folks....its buster iminimal install from install cdrom that had no mirrors to build because arcive_2019.key is not importe or something like taht
3657[23:18:47] <donofrio> do you know what I need to to get my local .deb files installed in correct order with dependancy resolution aka, how do I mount "sources" directory that is a copy of the buster netinstall cdrom to get reconized so this all "just works"
3658[23:19:00] <donofrio> sloshy, ot openjdk11
3659[23:19:04] <BCMM> donofrio: i don't think the order of installation is the problem here...
3660[23:19:24] <donofrio> ok I'm here to type what is needed to get apt-get update to work
3661[23:19:42] <BCMM> where did you get that gpgv .deb from?
3662[23:19:55] <mutante> move the .deb files to /var/cache/apt/archives/ and then use apt-get install ?
3694[23:24:11] <donofrio> welll how do I show you what I have without installing anything new
3695[23:24:12] <BCMM> donofrio: well, which is it? buster and sid are not the same...
3696[23:24:17] <donofrio> buster
3697[23:24:23] <mutante> lsb_release -a
3698[23:24:28] <donofrio> one min
3699[23:24:33] <BCMM> donofrio: so, how long is it since you ran `apt upgrade` or an equivalent?
3700[23:24:40] <BCMM> donofrio: and what happens if you try now?
3701[23:24:50] <somiaj> also why are you just not using the unoffical powerpc port, that is really all debian has to offer, what is this older iso you are trying to use?
3702[23:25:03] <mutante> kali ?:)
3703[23:25:12] <donofrio> lsb_release not found on this system
3704[23:25:18] <somiaj> the netinstall iso is not going to have that many packages, what is your goal here?
3705[23:25:43] <somiaj> BCMM: they have been having trouble trying to get the old powerpc port to work, of which no one really knows much about it anymore, and it is all unoffical and on ports.debian.org
3706[23:25:45] <mutante> that's odd, i usually have lsb_release on any ubuntu or debian i have recently seen
3707[23:25:48] <donofrio> I'll use whatever I can...aka I'm flexable send whatever iso and sources.list needed to make this work
3708[23:25:58] <BCMM> donofrio: honestly, there's a lot of your pastebin that i'm failing to see the relevance of, but the error at the end is, quite simply, the result of trying to install a new package on a very out-of-date system
3711[23:26:20] <donofrio> the very out of date system was downloaded and installed yesterday
3712[23:26:26] <BCMM> somiaj: what does "old" powerpc port mean in this context? is it discontinued?
3713[23:26:29] <donofrio> it has no mirrors and is only minial install
3714[23:26:50] <donofrio> I'm trying to inport arcvhive_2019.key so it wourld work with the powerpc arch type on th edebian-ports mirros
3715[23:26:51] <BCMM> donofrio: and you're trying to bring it up to date using a different, newer install cd, rather than a mirror?
3716[23:27:07] <BCMM> somiaj: is it gone from the mirrors now? is that why donofrio is trying to update it from cd?
3717[23:27:14] <somiaj> BCMM: powerpc has not been an offical port since jessie, support for it was dropped in stetch, and the packages have been moved over to ports.debian.org. It only has sid, and it isn't that well supported.
3719[23:27:36] <donofrio> local deb files all I have till I can get sources.list that will work with this....asking and waiting not looking for quick answress but I'm here to do whta you recommend
3720[23:27:37] <somiaj> they have been trying to do alot, but it could be that the cd and the install they have work are just not compadable, I don't think any powerpc cds are made anymore
3727[23:28:47] <bernyrd> hi I have a disk I install on, but too small
3728[23:28:48] <bernyrd> can I replace with bigger disk? it use lvm, can I dd the smaller one to the larger, and run a lvm command to see bigger space?
3733[23:29:54] <somiaj> and powerpc is just one of those now obsecure ports that not many people use regurally so as I've menetioned multple times, is not gonna be well supported
3736[23:30:29] <somiaj> BCMM: I've pointed them there multiple times, but yea each time they come here, we have to point out to the person offering help they are using powerpc
3737[23:30:38] <vlt> bernyrd: Yes, you can change the size even of physical volumes later.
3738[23:30:44] <BCMM> somiaj: and the package donofrio is failing to install, which apparently was extracted from the iso, wants sid's current libc6 version (which he hasn't got)
3739[23:30:58] <somiaj> BCMM: ahh, maybe they are still making iso's, problem is finding someone who has actual experience (other than reading the docs) with powerpc to provide good support.
3740[23:31:01] <donofrio> tried the one line in my sources.list and got this - replaced-url
3741[23:31:23] <somiaj> donofrio: did you use the old jessie installer to get where you are at?
3742[23:32:18] <somiaj> donofrio: so you need to get the current key which you have also been pointed to, replaced-url
3744[23:32:55] <somiaj> you could also just disable the key, and get packages without apt-secure, this is just a config option to move beyond tha tpoint
3745[23:33:00] *** Quits: Logg (~Logg@replaced-ip) (Remote host closed the connection)
3746[23:33:00] <donofrio> no I got it from replaced-url
3747[23:33:07] <BCMM> donofrio: ah, i think i see the problem now... the system is out-of-date; you can't bring it up to date from the mirror because a key isn't installed, and you're trying to manually install gpgv so you can update the key. is that right?
3748[23:33:17] *** Quits: JohnML (~john1@replaced-ip) (Remote host closed the connection)
3761[23:35:36] <BCMM> so there are two different install ISOs involved, right? the one you installed from, and the newer one that you're now trying to update from?
3762[23:35:54] <donofrio> no only one iso from that link replaced-url
3769[23:36:52] <somiaj> maybe it would be best if they just disabled secure apt, and updated the keys that way, then moved on. Or use wget to get the key and use apt-key add to add it
3770[23:37:01] <donofrio> from the .iso file itself I expanded it and scp'ed it to the sources directory locally in my homedir
3771[23:37:03] <BCMM> but surely the iso doesn't have packages that use more than one different libc version on it?
3780[23:37:38] <somiaj> BCMM: yea, seems the current state of ports is hard to find good docs, even ports.debian.org didn't update the curernt key from 2018 to 2019 on the webpae, we had to find the new key elsewhere.
3781[23:37:39] <BCMM> somiaj: i *was* starting to wonder about that... you don't actually need gpgv to install a debian key, do you?
3782[23:37:56] <BCMM> huh, i thought curl was in the default install
3783[23:38:02] <BCMM> maybe just libcurl without the client
3784[23:38:13] <somiaj> I'm not quite sure on the details, its always just worked for me using the debian package, I think you can manually use gpgv, but apt-key is a front end to manage this in some sense.
3785[23:38:35] <donofrio> ok so here I am still wondering what to do next
3786[23:38:40] <somiaj> but they can just disable secure apt and get around the invalid key warning, you do have to trust someone doens't try to inject something, but I think they are safe.
3788[23:39:15] <donofrio> I have the files cally, if I could import the archive_2019.key instal the live install session then it would see the ports mirror?
3801[23:41:50] <BCMM> i'm also a bit confused about how they got the outdated files installed in the first place
3802[23:42:13] <BCMM> the only .iso on the page is a netinst iso - i thought netinst didn't actually install any packages from the cd
3803[23:42:17] <donofrio> who 'they' the mirror maintainers or me?
3804[23:42:20] <somiaj> yea, I think that is the main problem, donofrio needs some help, but no one around to help is actually familar with debian-ports and powerpc
3805[23:42:37] <somiaj> netinstall contains a minimial base system that doens't need a network to install
3806[23:42:40] <donofrio> BCMM, your close that sounds irght
3807[23:42:50] <somiaj> so you can do a minimial install from a netinstall without internet.
3808[23:42:52] <donofrio> it didn't install much of anything lucky to have bash and ssh
3809[23:43:05] <donofrio> right but how do you get it to updat eonce the internet is there?
3810[23:43:10] <somiaj> it installs a little more than debootstrap, a little less than the standard task
3811[23:43:10] <Primer> if you have ssh, you can have curl
3815[23:43:43] <somiaj> donofrio: have you read the manpage to apt-key, it says 'apt key filename' add that 2019.key file you downloaded
3816[23:43:43] <BCMM> somiaj: i mostly just use debootstrap, so i'm a bit confused here. does netinst give you the choice between a *very* minimal system from the CD, or whatever you want, up to date, from the net?
3817[23:43:46] <donofrio> I had to apt install sshclient from dvd
3818[23:43:50] <donofrio> to get sh to work
3819[23:43:51] *** Quits: pringau (~pringau@replaced-ip) (Remote host closed the connection)
3820[23:44:10] <BCMM> talking of debootstrap... is that perhaps the answer here?
3821[23:44:12] <Primer> ssh works over a network. If you're networked, use said networks, and apt uddate; apt install curl;
3822[23:44:21] <mutante> BCMM: yea, but it is just whether you select stuff in the software install sectin of the installer.. if you click "desktop system" or not etc
3824[23:44:36] <BCMM> just nuke it and get a new system that's current with debootstrap maybe?
3825[23:44:47] <BCMM> Primer: network works, apt doesn't. keyring is out of date.
3826[23:44:48] <somiaj> BCMM: Basically you can skip the 'configure apt' part and just 'finlize install', in that case you get what is on the cd. If you go as far as configuring apt and setting up a network/mirror, the netinstall will update all the packages and give you the option (via tasksel) to install additional software.
3827[23:45:00] <BCMM> Primer: and the system may or may not be too minimal to bring it up to date...
3828[23:45:11] <Primer> then download a static curl binary? Surely these exist
3829[23:45:17] <Primer> oh powerpc, right...
3830[23:45:19] <mutante> i would never use tasksel..there is always time to install stuff later.. and only the ones you really want, heh
3832[23:45:40] <somiaj> donofrio: and the minimial install didn't include wget, I thought it included that
3833[23:45:58] <BCMM> why is wget the problem, though?
3834[23:46:18] <somiaj> mutante: yea, I like the minimial install, and I like that you don't need network for netinstall, just incase you needed non-free firmware or something, skip the network part, get a minimial system, and build up from there.
3836[23:46:39] <somiaj> BCMM: so they can get the key, ,replaced-url
3837[23:46:40] <BCMM> somiaj: i know there's a bunch of irrelevant stuff in the pastebin, but have a look at lines 13/14 replaced-url
3838[23:46:53] *** Quits: BlueByte (~walther@replaced-ip) (Quit: This computer has gone to sleep)
3839[23:46:55] <BCMM> looks like they have the key, and gpg really is the problem
3840[23:47:15] <Primer> Can't the key just be dropped in /etc/apt/somewhere?
3841[23:47:38] <donofrio> yah I'll do that but it will still erro I believe
3842[23:47:39] <Primer> I thought apt-key was just a convenience
3843[23:47:51] <Primer> apt update after dropping the key in the dir?
3844[23:47:59] <somiaj> ahh, then they just need to disable the key alltogether at this point
3845[23:48:13] <Primer> you can always force things in apt itself
3846[23:48:17] <BCMM> Primer: i think it gets added to a single keyring file
3847[23:48:18] <Primer> or just use dpkg
3848[23:48:25] <BCMM> /etc/apt/trusted.gpg
3849[23:48:36] <mutante> somiaj: actually i do need non-free firmware each time.. and it's always wifi ..iwlwifi..because i cant be bothered to get the ethernet adapter (sigh that this is a thing) and cable and plug in somewhere .. the problem is usually i copy the firmware files to a USB drive and then the installer scans it but doesnt find anything :p
3850[23:49:01] *** debhelper sets mode: +l 1500
3851[23:49:03] <donofrio> it's not seeing it in /etc/apt
3852[23:49:14] <BCMM> mutante: there are iso images available with nonfree firmware included...
3853[23:49:19] <donofrio> should the renameed arcive_2019.key.gpg file go somehwre else
3854[23:49:43] <BCMM> no, it has to be integrated in to trusted.gpg using gpg
3855[23:49:48] <BCMM> using gnupg
3856[23:49:53] <BCMM> (i think)
3857[23:50:03] <somiaj> mutante: I was commenting how you can skip that stage in the installer, then deal with it from a minimial install (copy firmware over, configure network, move on)
3858[23:50:44] <mutante> BCMM: good point, i might look next time
3859[23:50:48] <somiaj> I just am not familar enough with apt-secure, but at this point I just think disabling it would probably be the best way to move forward.
3860[23:50:51] <mutante> somiaj: oh. right! that's another way, sure
3861[23:50:57] <BCMM> somiaj: do debports have their own signing key, or do they use the main debian one?
3862[23:51:08] <somiaj> BCMM: their own, I've linked it a few times,
3863[23:51:26] <somiaj> actually donofrio has the key, they just can't seem to add it due to gpg missing
3864[23:51:30] <BCMM> somiaj: right, so donofrio can't just scp a trusted.gpg from another, working debian machine...
3865[23:51:58] *** Quits: woenx (~quassel@replaced-ip) (Remote host closed the connection)
3866[23:51:59] <donofrio> yep, and the .deb files are in the /home/donofrio/sources directory but dependencies kicking in ;(
3873[23:53:01] <donofrio> yah I was dpkg --install attemted
3874[23:53:11] *** Plagman_ is now known as Plagman
3875[23:53:24] <somiaj> I still think adding a correct config to /etc/apt/conf.d/ to disable the key checking is probably their best bet at this juncture
3876[23:53:25] <mutante> donofrio: if you'd use apt-get that would resolve dependencies.. and you can use apt-get to install from local files..if you copy them to /var/cache/apt/archives/
3893[23:59:01] <Rojola> What has the most recent version?
3894[23:59:01] <somiaj> BCMM: yea, it seems to just disable it for packages, but not for update, which is why I think we also needed Ignore gpg-pubkey
3895[23:59:04] <Rojola> a) Debian 9.7
3896[23:59:09] <Rojola> b) Ubuntu 18.04 LTS
3897[23:59:17] <BCMM> donofrio: only do this if you're reasonably sure your internet connection is safe, and be sure to put it back after you've got the key updated
3898[23:59:23] <somiaj> Rojola: debian 9.9 is the current point release, but debian buster (10) will be released in a week
3899[23:59:26] <donofrio> like this? deb "trusted=yes" replaced-url
3900[23:59:32] <Rojola> I must pick an OS on digital ocean for a droplet