5[00:03:20] <dpkg> Debian incorporates security fixes into the
version currently in <stable>. This is non-trivial so it may
take a couple days. Backporting security fixes means that you can
update the package without problems like changed behaviour that
would come with updating to a new version of the software. The
upstream version number doesn't change in the Debian package
when this is done; check the changelog or the <tracker of
doom>.
7[00:04:29] <MrHatter123> I have just done a fresh debian 10
install, with RAID1 mdadm, trying to install grub on the 2nd disk
with : grub-install /dev/sdb, getting this error :
/usr/sbin/grub-install: error: disk `md0' not found.
8[00:04:51] <MrHatter123> This is a process I have been doing
for the last decade
38[00:19:57] <ratrace> on normal Debian installation, you
ABSOLUTELY do not need sudo to run grub install (or anything
requiring root)
39[00:20:01] <MrHatter123> grub-install /dev/sda from the
command line no longer works
40[00:20:05] <ratrace> !buster su
41[00:20:05] <dpkg> In buster, su no longer overrides PATH by
default, requiring that you use "su -" or "su
-l" for login shells (which is not really a new thing at
all...). See
replaced-url
42[00:20:13] <ratrace> so you're confusing PATH with being
root perhaps?
43[00:20:31] <MrHatter123> so if you use the path
/usr/sbin/grub-install /dev/sda : this error happens :
/usr/sbin/grub-install: error: disk `md0' not found.
44[00:20:37] <ratrace> ie. grub-install vs /sbin/grub-install
45[00:20:50] <MrHatter123> if I use the command sudo
grub-install /dev/sda : it works.
46[00:21:10] <ratrace> I see no reason why that should work.
Also, I have dozens of servers with mdadm for /boot and I never had
to invoke sudo
47[00:21:30] <MrHatter123> ratrace, me too, more than dozens.
this has not happened before.
48[00:21:36] <ratrace> I'm pretty sure you're either
not root, or that's not regular debian, but something with
broken permissions or ACLs
49[00:21:51] <MrHatter123> it was a vanilla debian 10 buster
50[00:22:11] <ratrace> is it a chroot?
51[00:22:14] <jhutchins> MrHatter123: Yeah, lots of people run
that, and they aren't having problems
58[00:22:55] <dpkg> To reinstall <GRUB> boot to your
Debian install disk/live CD, switch to the other console (Alt-F2),
mount your root filesystem (mount -t ext4 /dev/whatever /target ;
mount --bind /dev /target/dev ; mount -t proc none /target/proc ;
mount -t sysfs none /target/sys), chroot into it (chroot /target),
run "mount /boot/efi" on EFI and "update-grub
&& grub-install /dev/whatever". See also <rescue
mode>, <dual boot guide>, <supergrub>.
59[00:22:57] <MrHatter123> fresh install, off a dvd, debian 10
buster, not much installed yet, ssh, openvpn, resync
60[00:23:08] <MrHatter123> 'rsync'
61[00:23:42] <ratrace> I cannot imagine why running as root
would not work and running as root + sudo, would. UNLESS that's
not regular debian but something broken
62[00:24:02] <ratrace> the only situations where I saw root
being impotent was in context of a MAC
68[00:25:11] <MrHatter123> something changed with the pathing
since buster, I have uprgaded losts of boxes from 9 to 10 recently
and I now have to invoke the direct path
69[00:25:26] <MrHatter123> however, this time the direct path
produced the error
70[00:25:35] <MrHatter123> then didn't work until sudo was
run
71[00:25:47] <ratrace> hold on, yes, if by "invoke direct
path" you mean you need absolute path, see the !buster su
factoid abovoe
91[00:29:15] <ratrace> yes, the installer does the sane thing
and installs sudo for you if you don't give root pass.
otherwise you'd be dead in the water and would have to
reinstall
96[00:29:53] <MrHatter123> but then, even when su'd to
root, it doesn't work without sudo
97[00:30:06] <ratrace> MrHatter123: did you try su -l or just
su ?
98[00:30:15] <MrHatter123> to root ?
99[00:30:17] <ratrace> yes
100[00:30:28] <MrHatter123> logged in via ssh with normal user,
then su
101[00:30:32] <ratrace> so again:
102[00:30:34] <ratrace> !buster su
103[00:30:34] <dpkg> In buster, su no longer overrides PATH by
default, requiring that you use "su -" or "su
-l" for login shells (which is not really a new thing at
all...). See
replaced-url
105[00:30:51] <ratrace> you need su - (or -l) to become root,
otherwise your PATH is broken.
106[00:31:02] <MrHatter123> ratrace, thank you, I wasn't
aware of this.
107[00:31:12] <MrHatter123> this is new since buster ?
108[00:31:18] <ratrace> so I'm guessing using sudo as root
re-munges the PATH again... but I assure you, running sudo as root
is absolutely not ever needed.
185[01:20:50] <TheSilentLink> Hi I'm trying to install
wireguard from the backports but I get Building initial module for
5.9.0-0.bpo.5-arm64 Error! The dkms.conf for this module includes a
BUILD_EXCLUSIVE directive which does not match this kernel/arch.
This indicates that it should not be built. Skipped.
186[01:21:10] <TheSilentLink> anyone know why? I installed
linux-headers-5.9.0-0.bpo.5-arm64
187[01:21:41] <sponix> TheSilentLink: it is skipped because
kernel 5.9 has wireguard built in, it has no need to do the dkms
build :)
188[01:21:56] <TheSilentLink> oh I'm an idiot!
189[01:22:08] <TheSilentLink> sponix: I don't need
wiregurd-dkms then right?
190[01:22:08] <sponix> So, just install the wireguard-tools or
whatever and you are done
191[01:22:28] <sponix> TheSilentLink: nope, no wireguard-dkms
needed for like kernel 5.6 and above
192[01:22:56] <TheSilentLink> yep I knew that didn't
realise I was on 5.9
193[01:23:32] <TheSilentLink> sponix: what about the wireguard
package? do I just need tools?
208[01:27:20] <JackFrost> wireguard is a meta that depends on
either the dkms module or a kernel that supports wireguard native.
It doesn't actually contain anything.
340[03:56:12] <awal1> i have 2 screens here. why a given
software chooses to apear in the external screen instead of the
main? i'm referring to 'obsession-exit' window pop-up
(always appearing in the exyernal screen, and 'tor
browser', it just did that the first start. (openbox wm)
540[09:03:37] <jaggz> themill, man.. I just wanted to install
avidemux but it's not in official repos
541[09:03:46] <ratrace> jaggz: ah, thanks
542[09:03:49] <n4dir> jaggz: librazik has lots of audio tools,
and it has an irc channel. Wether or not they contain
debian-multimedia.org i wouldn't know
543[09:04:09] <n4dir> it doesn't have avidemux though.
544[09:04:49] <jaggz> oh a whole audio system distribution based
on debian.. that's neat, n4dir
545[09:05:01] <jaggz> it's the ubuntu of audio studios :)
546[09:05:18] <n4dir> jaggz: well, you can install all of it on
top of debian. Or whatever you need only
547[09:05:26] <n4dir> and lots is not in the debian repos.
548[09:05:27] <ratrace> you mean like ubuntu studio? :)
574[09:17:33] <jaggz> n4dir, thanks for checking that out
575[09:18:12] <ratrace> n4dir: "< jaggz> it's
the ubuntu of audio studios" "< ratrace> you mean
like ubuntu studio?"
576[09:18:19] <jaggz> don't we have a "why is package
x not included" search? .. or is that in debian-next or
something
577[09:18:39] <n4dir> jaggz: i seem to recall a guy called
golinux used or uses avidemux a much, from #devuan. You might ask
over there, perhaps he can give you a hand
578[09:19:13] <jaggz> well.. I'd just build the thing from
source if I needed to.. I usually use debian-multimedia.org though
579[09:19:29] <n4dir> ah, then all is good, it sounds
580[09:19:30] <jaggz> too many things it has in it that
aren't in debian (throughout the years anyway.. can't
think of much notable right now)
582[09:19:43] <jaggz> yeah.. thank you for the references though
583[09:20:11] <jaggz> I did enjoy ubuntu's packaging of
stuff on top of testing years ago, but I .. well, to be frank I just
don't trust it
584[09:20:52] <n4dir> imho ubuntustudio does a good job though.
Makes it pretty easy and comfortable
585[09:20:56] <jaggz> so in all these years I just really
don't have a good middle ground between the often
extremely-aged debian stable and the often-disturbing instability of
-testing
586[09:21:15] *** Quits: szorfein (~daggoth@replaced-ip) (Remote host closed the connection)
587[09:21:25] <ratrace> jaggz: not even backports?
627[09:52:52] <backupluis> Hi, Im a bit confused about nvidia
drivers, I understand that debian have nvidia drivers but I
don't know what is the difference with propietary from the
nvidia site. I just installed from the site but I need some i386
libraries that there are no more on nvidia package (propietary).
628[09:53:44] <ratrace> backupluis: it's the same driver,
but packaged in Debian. you should _not_ run the .run installer from
nvidia.com
629[09:53:55] <ksk> backupluis: read the wiki article about
nvdia, its quite useful.
633[09:54:46] <ratrace> because it installs files outside the
package manager, because you need to remember to re-run it after
each kernel upgrade, because it's a mess requiring special
tools to uninstall..... just use the packaged proprietary driver, if
you want the proprietary one.
634[09:55:40] <backupluis> Yes, Im in that page, but because the
page says that The propietary provides blah blah and says nothing
about the driver on Debian I choose to install the nvidia one.
640[09:59:13] <ratrace> (and thus part of package management and
DKMS triggered rebuilds)
641[09:59:41] <Mister00X> ^^ which is what one wants for ease of
use
642[10:00:26] <backupluis> Will move to the debian one, Years
using the propietary because for me the page says its better from
the nvidia site, maybe someone should clarify in the page.
643[10:01:36] <ksk> backupluis: did you run nvidia-detect?
644[10:01:40] <Mister00X> > Multiple precompiled driver
versions are available for Debian 10 "Buster": it says
645[10:02:20] <Mister00X> it also says > This page describes
how to install the NVIDIA proprietary display driver on Debian
systems.
646[10:02:23] *** Quits: tijara (~tijara@replaced-ip) (Disconnected by services)
649[10:02:42] <ratrace> ... and proceeds to do it with apt.
650[10:02:45] <backupluis> I've change my card some weeks
ago, so I just download the latest from nvida, uninstall the old one
(legacy) and put this new one, everything its ok, but I knwow, as
you say that I have to reinstall or rebuild on every system upgrade.
651[10:02:53] <ratrace> nothing on that page seems to suggest to
go and run the .run
652[10:03:10] <backupluis> The proprietary "NVIDIA
Accelerated Linux Graphics Driver" provides optimized hardware
acceleration of Open
653[10:03:26] <backupluis> And says nothing about the Debian is
better
654[10:03:39] <ratrace> the nvidia-driver package IS the
proprietary driver
658[10:05:01] <backupluis> Well, will reboot to change my
driver, Thank Mister and ratrace!!!
659[10:05:07] <ratrace> I mean the whole page is about
installing "the proprietary driver" in Debian, and has
steps involving non-free repo and apt, to install the package.
Except for some CUDA toolkits, it doesn't even link to
nvidia.com to fetch the tarballs
662[10:06:24] <ratrace> I'm curious about something though
..... did they even use the proprietary driver? Nouveau loads by
default and unless it's blacklisted (and it is by the
nvidia-driver-package), it would still load and preclude the
proprietary one, no?
663[10:06:43] <ratrace> (assuming it's not manually
blacklisted)
664[10:06:57] <Mister00X> ratrace: I think they conflict each
other
683[10:20:54] *** Quits: nt1036 (~nt1036@replaced-ip) (Remote host closed the connection)
684[10:23:29] <Mister00X> ratrace: Ok yes this is done by the
post inst. Hopefully nvidia does the same..... not that they care
about it but it would be nice
688[10:25:21] <dpkg> Please do not ask if anyone can help you,
knows 'something' or uses 'some_program'.
Instead, ask your real question. (If the real question _was_
"does anyone use 'some_program'?" ask me about
<popcon> instead.) See <ask> <ask to ask>
<polls> <search> <sicco> <smart questions>.
690[10:26:57] <Lope> :) I dumped my gnome-terminal profile like
this: `dconf dump /org/gnome/terminal/legacy/profiles:/ > pro`
Then I edited the visible-name and the UUID slightly, then loaded it
like this `cat pro | dconf load
/org/gnome/terminal/legacy/profiles:/` if I dump again, it's
there, but doesn't show up in gnome-terminal preferences even
after closing and re-opening gnome-terminal.
691[10:29:11] <Lope> okay, this looks helpful:
replaced-url
694[10:30:55] <Lope> Interestingly the profile that I created
with dconf doesn't show up when I do `gsettings get
org.gnome.Terminal.ProfilesList list` but it does when I do this:
`dconf dump /org/gnome/terminal/legacy/profiles:/` lolz
732[11:13:20] <choice> Hello! Does anybody know if the mate
desktop has a dark mode? It seems Chromium only accepts a dark mode
flag from the os. And I would like Chromium to be in dark mode.
795[12:26:47] <iGadget> Hi all. Question - I have to save my
data from a cloud provider which is terminating services next week.
The only way to get the data is via web gui (not really doable for
250GB+) or webdav. So I installed davfs2, mounted the share and
started the copy job with cp. Hoping out would be finished in the
morning... only to find out now that the process has completely
stalled, but cannot be stopped (not even with kill -9 as
796[12:27:33] <iGadget> Apparently, the process is in
'Uninterruptible Sleep', status D+ according to ps.
797[12:28:23] <iGadget> So... am I cordially screwed and do I
have to reboot my system, or is there anything else I could try?
806[12:33:25] <ksk> I think yes, you will have to reboot.
807[12:34:22] <ksk> there is the lazy umount option, but I dont
think thats something to recommend ever, it will also not get rid of
the dead cp most likely.
808[12:34:54] <iGadget> Darn. Well that's a tough lesson
then - NEVER EVER use davfs on a production server for a large
amount of data.
809[12:34:59] <ksk> If I were you I would maybe checkout some
"dav downloader" tool instead of mounting it.
810[12:35:16] <ksk> I dont think that DAVfs in general is a
problem, I have seen it used in production.
812[12:35:35] <ksk> since webdav is http, it might be wget can
do the job.
813[12:36:04] <iGadget> If not DAVfs, then what else could it
be?
814[12:36:26] <dob1> I don't find anything in
/etc/network/interfaces where is them defined? for sure I
didn't use networkmanager, I setupt it on installation
815[12:36:51] <ksk> I mean it has most likely something to do
with implementation here.
842[12:46:44] <dob1> ratrace, I didn't let me to edit
network manager connection, maybe becasue it is in use?
843[12:46:54] <ratrace> systemd-networkd is always there but
without explicit configuration files and enabled service, it
won't affect your config
844[12:47:10] <ratrace> interfaces(5) is run by
networking.service, by default enabled on Debian
845[12:47:57] <ratrace> NM is by default configured NOT to touch
explicitly defined interfaces(5). so technically you can have all
three of them _present_ , the defautls are somewhat sane
852[12:51:59] <ratrace> dob1: you can also explicitly tell NM
not to touch your NICs, via NetworkManager.conf and no-auto-default=
entry
853[12:52:34] <ratrace> the entry managed=false under [ifupdown]
should ensure that NM doesn't try any of NICs managed by
interfaces(5)
854[12:53:16] <dob1> ratrace, well what I want to be sure while
testing is one of them works to not be cut-off from the pc
855[12:54:38] <ratrace> there's no guarantees, if you
fblarghk something up. ideally you'll have physical access or
an OOB way to rescue/change configuration
866[13:00:10] <dob1> this is my NetworkManager.conf
replaced-url
867[13:00:44] <ratrace> seems reasonable
868[13:02:36] <dob1> ratrace, the problem is that I am connected
via rdp to have xfce running and I can't edit connections (the
option is not available) and using sudo nm-connection-editor from a
terminale in xfce gives me cannot open display: :11.0
871[13:05:47] <ratrace> dob1: if you're feeling lucky, you
could configure interfaces(5) and reboot. this is not a sane advice.
please refer to the above requirement to have physical and/or OOB
access in case you flak things up.
872[13:06:41] <dob1> ratrace, but I cannot edit network manager
configuration ?
888[13:27:07] <meltingwax> i am getting this apt failure bug
(libef1:i386 fails to install, holding up all of apt). are there any
workarounds?
replaced-url
889[13:27:33] <Mister00X> meltingwax: you're on unstable or
testing?
890[13:27:37] <meltingwax> unstable
891[13:28:13] <Mister00X> meltingwax: there is a workaround mv
/usr/share/locale/en@boldquot/LC_MESSAGES/elfutils.mo
/usr/share/locale/en@boldquot/LC_MESSAGES/elfutils.mo-back &6
apt -f install
892[13:28:23] <Mister00X> but I guarantee for nothing
893[13:28:44] <meltingwax> thx will try
894[13:28:57] <Mister00X> meltingwax: also discussions about
unstable are usually in #debian-next on oftc
895[13:29:18] <meltingwax> thanks Mister00X , trying that
896[13:29:26] <Mister00X> meltingwax: note the typo in that
command
897[13:29:27] <meltingwax> re: oftc.. noted
898[13:29:31] <meltingwax> yes i figured that out :)
899[13:30:20] *** Quits: towo^work (~towo@replaced-ip) (Remote host closed the connection)
914[13:48:22] <iGadget> Well... things are escalating here
pretty quickly with the whole davfs debacle - when I now try to cd
into the mounted webdav directory using autocomplete (TAB), the
shell hangs :-(
917[13:50:32] <iGadget> As a wise wizard once said...
"Nothing is ever easy."
918[13:50:47] *** Quits: shingouz (~not@replaced-ip) (Disconnected by services)
919[13:51:55] <ksk> iGadget: thats kind of the "remote does
not answer" behaviour. Did you check you can access it using
wget? I can only guess a short notice "we go out of
business" letter could lead into many people wanting to
donwload stuff ;)
920[13:52:48] <iGadget> True that. Will try the wget method now.
937[14:08:36] <iGadget> ksk: wget method does not seem to work,
it only downloads the exact last part of the url
(.../remote.php/webdav) instead of the files located at that url. So
now I have exactly one file 'webdav', containing just the
word 'OK'.
1004[15:12:09] <jack2019> Hi, does make sense to encrypt root
partition?
1005[15:12:24] <wsky> sure
1006[15:12:48] <n4dir> i could not think of a good reason.
Depends what is installed, i'd guess.
1007[15:12:49] <wsky> full disk encryption is your friend
1008[15:13:22] <n4dir> but a typical desktop installation? hmm.
1009[15:14:18] <jack2019> a server
1010[15:14:26] <iGadget> Absolutely. It prevents the scenario
where (if you would only have /home encrypted) someone would
physically access your system, boot it, install a keylogger and
extract your credentials.
1011[15:14:28] <rue_mohr> wsky, what are you protecting what
from?
1012[15:14:46] <iGadget> For a server, I highly doubt it's
useful.
1013[15:14:50] <wsky> unauthorized offline access to the system
1014[15:15:24] <iGadget> When would the server be offline, at
night?
1015[15:15:43] <jack2019> can I encrypt from preesed.cfg?
1032[15:29:55] <iGadget> n4dir: for a laptop, I would recommend
it. Agree re stationary desktop usefulness, unless you want to
prevent the (for most people pretty unlikely) scenario where someone
would break in to compromise your system.
1033[15:30:13] <iGadget> break in physically, that is
1038[15:31:33] <n4dir> not that unlikely. Did it last summer in
my case, iGadget
1039[15:31:54] <ratrace> theft is not the only reason to encrypt
1040[15:32:49] *** tijara_ is now known as tijara
1041[15:32:55] <iGadget> ratrace: agreed on broken / replaced
disks, but how does this work for unattended servers? Since
you're required to enter the pass @ boot? Storing the pass in a
script on the server defeats the whole purpose, so how would that
work?
1042[15:33:38] <ratrace> iGadget: dropbear in initramfs + ssh
unlocking on boot. yes that means if the server reboots, someone has
to unlock it
1049[15:34:31] <iGadget> n4dir: depends on your risk profile I
suppose? I cant think of anyone doing such a thing just to be able
to access my mums pc ;-)
1059[15:38:04] <ratrace> jack2019: that was tongue in cheek. no,
you can't learn ansible in one hour. please use google to find
ansible documentation, articles on how to use it, and best of all:
start using it. the only way to learn is to actually use it.
1061[15:38:43] <iGadget> ironically, youtube is often a very good
starting point for tutorials as well.
1062[15:38:57] <n4dir> i agree.
1063[15:39:00] <ratrace> why's that irony?
1064[15:39:27] <jelly> jack2019, install it, set up some nodes
and play with it
1065[15:39:31] <gildasio> hi all, is there any recommendation
(preferable from official doc) about disc/partition size?
1066[15:39:34] <iGadget> because youtube, at least in my mind, is
associated to bs entertainment
1067[15:40:16] <iGadget> gildasio: I guess that totally depends
on your use case...
1068[15:40:22] <ratrace> still not irony. irony would be if YT
actively discouraged learning and educational videos, so,
ironically, that's where oyu find most of them.
1069[15:40:39] <ratrace> yes I have an axe to grind with misuse
of "ironically" :)
1070[15:40:43] <iGadget> ratrace: I stand corrected.
1072[15:41:23] <ratrace> (or better example: a YT video on how to
disable tutorials is, ironically, a tutorial) :)
1073[15:41:36] <gildasio> iGadget: I'm thinking to create a
base template for VMs and depending on the case, when instantiate
this template, if necessary, create others disks and mounting where
I need. So my case is only about the /
1074[15:42:33] * ratrace steps out for lunch
1075[15:42:34] <iGadget> gildasio: I dont have enough experience
with that, sorry. Anyone else?
1076[15:43:51] *** Quits: milkt (~debian@replaced-ip) (Remote host closed the connection)
1080[15:45:22] <iGadget> ...but thinking about it, I still think
it depends on the role that the VM will be getting. The more apps
you want to install on it, the larger the rootfs will need to be.
Some filesystems allow for resizing though.
1084[15:48:42] <gildasio> Yeah, I'm thinking about it. I
want to install only the basics packages (with some for debug cases
when need such as dig, tcpdump, htop...)
1085[15:49:35] <gildasio> Maybe I'll use LVM too. And for
hypervisor I use KVM/QEMU/LibVirt, so can resize the disk if needed
1086[15:50:20] *** Quits: kfvn (~weechat@replaced-ip) (Remote host closed the connection)
1094[15:53:12] <iGadget> Re my davfs issue from earlier - If
anyone ever needs to sync a large amount of data from a webdav
server - don't use davfs but use rclone instead. It's
breeze to set up and it's FAST... And I mean a like a speed
difference of > 4 times :-D
1109[16:04:08] <iGadget> Fair enough. Personally, I tried doing
it myself first but quit after a week of fruitless effort. But that
probably says more about my second-prize skillset ;-)
1128[16:18:28] <vanfanel64> UUID=3537-3761 With my udev rule
disbled and with the quotes removed from the UUID="..." in
fstab: /data exfat-fuse
defaults,auto,nofail,x-systemd.device-timeout=2ms 0 0
1129[16:18:38] <vanfanel64> With the quotes removed, it will now
mount to /data if I use 'sudo udisksctl mount -b
/dev/mmcblk3p1' but it no longer tries to mount at all on
insertion.
1130[16:18:56] <vanfanel64> I need this SD card to mount, to
/data, 1) on boot, if inserted already, and 2) if it has been
ejected and then inserted again.
1131[16:19:49] <vanfanel64> I've tried researching about
this but can't seem to find a good answer as to how to get it
to mount automatically.
1149[16:36:42] <karlpinc> vanfanel64: I used to use something
that used udev to mount-on-insert, but I've forgotten what it
is. It might be easier if you just had it automount. You add the
x-systemd.automount option to the line in fstab. See: man
systemd.mount
1162[16:43:50] <karlpinc> vanfanel64: Then it mounts when you try
to access it. You probably don't want boot to hang, so
you'll also need (iirc) noauto, and maybe other mount options.
And you don't want it backed up, or ordered with those that
mount on boot so the last 2 fields are 0.
1163[16:44:20] <karlpinc> vanfanel64: Does not hurt to skim: man
mount (iirc)
1165[16:46:50] <user99> I am not sure my install completed
properly. There was a hang when loading the gui initally (amd
graphics nonfree etc), I have that working properly but I am not
sure the sytem completed everything at setup.
1167[16:47:29] <user99> Some things are broken...like roots Path.
I can export of course but /etc/profile does not seem to be sourced
when I su
1168[16:47:54] <user99> and sudoers doesn't work and visudo
is not installed and not sure what I should do sinc sudoers issues
1169[16:48:05] <vanfanel64> karlpinc, oh, it does mount when I ls
/data, wow, never saw that before tbh
1170[16:48:13] <ectospasm> user99: are you using `su -` instead
of `su`?
1171[16:48:30] <user99> no I should I guess ...sill me
1172[16:48:33] <jelly> !buster su
1173[16:48:33] <dpkg> In buster, su no longer overrides PATH by
default, requiring that you use "su -" or "su
-l" for login shells (which is not really a new thing at
all...). See
replaced-url
1174[16:48:36] <vanfanel64> karlpinc, I thought nofail and
x-systemd.device-timeout=2ms would prevent boot hang when trying to
mount /data
1191[16:55:44] <shtrb> Is there a dvd source for oldstable that
would include non-free packages ? I'm looking at
replaced-url
1192[16:56:31] *** Quits: cfoch (uid153227@replaced-ip) (Quit: Connection closed for inactivity)
1193[16:56:55] <vanfanel64> ectospasm, I like the new ones though
sometimes I have to deal with old scripts that expect the older
tools. I like your nick btw, like if Slimer has siezers
1194[16:57:47] <ectospasm> Those scripts should be updated, since
as you can see netstat isn't installed by default anymore.
1195[16:58:01] *** debhelper sets mode: +l 1185
1196[16:58:25] <ectospasm> I always recommend not using
deprecated tools, you never know when they will no longer be
available in the repos.
1197[16:58:29] <vanfanel64> ectospasm, right, I do update things
when I am allowed to
1198[16:59:00] <ectospasm> Yeah, not having authorization to
update things is always a pain when your a (junior) sysadmin.
1199[16:59:03] <shtrb> ectospasm, I hope that net-tools would not
be removed at least post Debian 15
1200[16:59:37] <ectospasm> shtrb: I don't know when Debian
will drop that package, so you're probably right.
1201[16:59:51] <ectospasm> That it's not installed by
default should tell you something, though.
1202[16:59:53] <shtrb> deprecated or not, if stuff is working
many people would say don't touch it :D
1204[17:00:15] <ectospasm> There's a reason things are
deprecated. Ignore it at your own risk.
1205[17:00:49] <vanfanel64> shtrb, good point, netstat and
ifconfig are really ingrained for a lot of people. I still find my
self trying to use them without thinking
1206[17:00:55] <shtrb> There had been reasons to migrate most of
debian to systemd, not saying the reason are accepted by all or they
are good for all
1210[17:02:58] <vanfanel64> ectospasm, I wouldn't say it is
so much about being against change, it is more when alternatives
aren't well presented before the change occurrs
1212[17:03:15] <shtrb> ectospasm, calling people names without
addressing the actual reasons why someone would not be happy with
it, is little bit not productive
1214[17:05:16] <ectospasm> In many cases that ship has sailed,
though. You're free to replace systemd on your own systems with
whatever you feel like. But many parts of the Linux world have
already adopted it, you should have presented your objections a
while ago.
1219[17:06:08] <dpkg> #debian is primarily a support channel for
Debian users. Please keep the discussions in #debian on-topic and
take longer discussions and non-support questions to
#debian-offtopic. Imagine the chaos if each of the hundreds of
people in the channel felt the need to wander off topic for a few
minutes every day.
1220[17:06:20] *** Quits: xet7 (~xet7@replaced-ip) (Remote host closed the connection)
1222[17:06:57] <shtrb> ectospasm, having a rotten tooth when you
can not afford a replacment , doesn't mean you should not try
to live with it and try to leave around the pain and maintenance of
it.
1227[17:09:09] <foxide> Not really sure what the purpose of this
discussion is. systemd is, objectively, far superior and presents
far fewer issues than what it replaced. I do not believe that a
Debian system with systemd removed is supportable, as that's
the equivalent of a lobotomy. Find another distro if you're
that opposed to systemd, but I would first recommend you learn the
tools now available to you.
1228[17:10:08] <shtrb> sorry, I was venting as I yet again
stumbled on a systemd idiotic bug - that someone else might have
already reported
1229[17:10:48] <foxide> Yeah, bugs happen, and I can appreciate
the frustration.
1230[17:10:53] <shtrb> my issue was not just with systemd, but
with ectospasm claim that who ever complain about systemd is
curmudgeons
1231[17:11:14] <shtrb> so he hit a nerve as to speak :)
1232[17:11:26] <foxide> I can also appreciate that, yeah.
1233[17:11:40] <foxide> Bugs are viable complaints. :P
1234[17:13:00] <shtrb> Was I correct to assume that offical DVD
would not include non-free packages ?
1240[17:14:28] <jelly> including myself when I do it
1241[17:14:36] <n4dir> both, anitx and devuan, don't look
much like a lobotomy to me.
1242[17:15:14] <foxide> n4dir: Well, they are. /shrug
1243[17:15:44] <foxide> Is there a process to request that a
package version in testing be brought into stable-updates? There is
no new functionality, so it doesn't qualify for
stable-backports afaik.
1244[17:15:48] <n4dir> been a bit, but when i ran debian minus
systemd it worked pretty ok too
1257[17:19:24] <shtrb> I however doubt that a new version only
have bugfixes, however a specific patch to fix a specific version
would be much more appriated via an bug report
1258[17:19:43] <jelly> foxide, a newer upstream version in
general won't be eligible, but maybe you can cherry-pick a
patch to be applied to the code already in stable
1259[17:20:13] <shtrb> jelly just said what I said in different
words
1260[17:20:15] <foxide> shtrb: I know the software in question
VERY well, its all bugfixes.
1261[17:20:22] <foxide> jelly: Okay, I'll see what I can
manage. Thanks!
1265[17:21:17] <shtrb> repeat what I say in different words ?
1266[17:22:02] <shtrb> foxide, if you are the dev behind said
software, you can suggest a patch for the specific version already
in stable, then it should be safe (and probably accepted)
1267[17:22:09] <oxek> when I do `pgrep -a host`, I get 28 results
of '$PID host -t soa local.'
1275[17:28:40] <iGadget> gildasio: Sorry, got called away. Re
your question what I ran into setting up my VM server - Well
actually, it was an Ubuntu server (I wanted LXD at the time) and I
wanted ZFS and a bunch of other stuff. For some reason I am
incredibly talented at picking exactly those combinations which lead
to an enormous amount of work. After a week or so it still
didn't work as I wanted. And then I tried Proxmox, which worked
in a
1300[17:36:58] <judd> No package named 'virsh' was
found in amd64.
1301[17:37:01] <jelly> ,v vmomi
1302[17:37:03] <judd> No package named 'vmomi' was
found in amd64.
1303[17:37:05] <shtrb> dpkg, esxi is do you wish to manage VMware
ESXi ? verify it is in read-write mode (and not read only) and then
use virsh (from virt-manager) or vmomi
1304[17:37:05] <oxek> jelly: perhaps, but not entirely sure, the
parent is '/bin/sh
/usr/lib/avahi/avahi-daemon-check-dns.sh'
1305[17:37:06] <dpkg> shtrb: okay
1306[17:37:26] <jelly> oxek, you're getting somewhere, and
what's ITS parent? :-)
1307[17:37:52] <oxek> jelly: so I always run `ps -fp PID` and
pick the PPID from the result?
1340[17:49:13] <jelly> which debian release is this?
1341[17:49:19] <oxek> debian-stable
1342[17:50:04] <oxek> the /etc/network/if-up.d/avahi-daemon file
comes from the avahi-daemon package, which has been automatically
installed, and the file has not been modified
1343[17:50:15] <jelly> hm. It's a bug. My _ubuntu 18.04_
laptop has it fixed.
1357[17:53:13] <oxek> I edit the
/etc/network/if-up.d/avahi-daemon to be the same as on a ubuntu
system?
1358[17:53:16] <jelly> oxek, well. If you don't need avahi
you might uninstall the damn thing
1359[17:53:39] <oxek> jelly: to be fair, I don't really know
what avahi is or what it does. It got automatically installed with
everything else.
1360[17:53:56] <ratrace> setup networking with avahi instead of
dhcp, and DigitalOcean will hire you.
1361[17:53:56] <oxek> and since it's part of default debian
installation, I don't know if I should be removing it
1362[17:54:18] <jelly> oxek, it makes linux and windows machine
names be visible on the local network
1363[17:54:19] <oxek> ratrace: is avahi better than dhcp?
1364[17:54:39] <ratrace> oxek: I thought that sentence oozing
with sarcasm would be obvious :)
1365[17:54:49] <jelly> so you can "ping glimmer" even
if the laptop named glimmer does not have a DNS entry
1366[17:55:09] <oxek> ratrace: I just really don't know
anything about avahi. Sounded vaguely like some iphone thing.
1367[17:55:26] <jelly> linux and macos use it, and apparently
windows as well these days
1368[17:55:45] <jelly> it's adjacent to dhcp.
1369[17:55:50] <ratrace> oxek: it's a "zeroconf"
thingy. things talk in the network and exchange info so avahi based
systems would know where to find printers and other network thingies
1371[17:56:05] <oxek> jelly: I do that through dhcp. but back to
the issue, what's the proper fix for the issue? Modify the
config file to be the same as ubuntu?
1372[17:56:07] *** Quits: Jerrynicki (~niklas@replaced-ip) (Remote host closed the connection)
1373[17:56:09] <ratrace> oxek: think "autodiscovery" of
network services, devices, etc...
1374[17:56:22] <jelly> oxek, it's not even a config file,
it's a proper packaged script
1376[17:56:34] <jelly> you can modify it to fix the bug.
1377[17:56:53] <oxek> ratrace: yeah autodiscovery never really
worked on any of my machines, I always had to give the IP address
directly or the hostname and have it resolved through DNS on the
router
1378[17:57:04] <oxek> and DHCP was giving machines hostnames
1379[17:57:08] <jelly> but until Debian fixes it, any further
upgrades of "avahi-daemon" package will overwrite your fix
1380[17:57:24] <shtrb> jelly, windows using something does not
make that something a good choice :P but if some day windows would
really start using true avahi then we could put some meat in the
stove and celebrate
1381[17:57:25] <oxek> ah, I didn't realize I'd need to
modify the packaged stuff
1382[17:57:52] <shtrb> !pinning
1383[17:57:52] <dpkg> Pinning is a method to choose which version
of a package to install when multiple versions are available from
<sources.list>. Bugs are explained at
replaced-url
1384[17:57:53] <oxek> I'm searching through the debian
bugreports to see if it has been reported already
1426[18:14:14] <jelly> almost no package starting with lib* is
not important on its own. Except if there -bin at the end, or if
it's libc6 which is not easy to remove
1427[18:14:31] <jelly> s/is not important/is important/
1428[18:14:56] <oxek> I also spotted there's
avahi-daemon.service systemd thing running
1429[18:15:25] <oxek> how did I not know there's this mess
on my pc? I only found out by pure chance by searching for processes
that start with host (was looking for hosted).
1430[18:15:56] <user99> bluetooth was enabled on mine default.
not sure why that is. but ez enough to kill
1431[18:16:09] <n4dir> something like that was the reason why i
said "foam at mouth". oxek .
1460[18:30:32] <dpkg> A zombie process is one which has issued an
exit(), but whose parent has not yet issued wait() (or one of its
variants) to retrieve the exit code. This usually means the parent
process is buggy. You can't kill a zombie, because it's
already dead; you have to kill the parent, or just ignore the
zombie.
1510[19:13:40] <dpkg> The 'freeze' is when large
changes to testing version of Debian are stopped for a few months.
During this period, no new package uploads are made except to fix
critical bugs or to provide documentation updates. During the
freeze, bugs are squashed to 'stabilize' the release to
make the new 'stable'.
1515[19:16:55] <greycat> that factoid is out of date
1516[19:19:40] <user99> is bullseye's status
'freeze'?
1517[19:20:02] <greycat> dpkg, no, freeze is <reply>In the
final stages of a release's testing cycle, a series of
increasingly restrictive "freezes" are enforced,
preventing various kinds of changes, in order to focus on getting
the release into shape.
1553[19:59:32] <bifunc2> In what ways is Qubes overkill compared
to Debian?
1554[20:00:28] <shtrb> You use debian within Qubes
1555[20:00:36] <bifunc2> lol
1556[20:00:53] <shtrb> what ?
1557[20:01:26] <shtrb> Your question is ill defined, these are
two distinctive things , which solve different problems
1558[20:01:52] <bifunc2> ok in that case, i kinda would ask,
under what circumstances would it be overkill to use debian within
qubes rather than just debian only.
1559[20:02:27] <bifunc2> sorry about the illformed question btw
1561[20:04:04] <shtrb> You can use similar domain approaches
within debian if you trust that enough, in fact you can think of
each domain as a separate VM (xen or even qemu level), using other
isolation tools works too
1565[20:05:53] <shtrb> One thing I do not like these days is to
chatty dbus, but if you have a truly separation between vms for
isolation , debian which had been secured enough might be more than
enough than having a full blowen QubeOS
1566[20:06:45] <shtrb> bifunc2, ^
1567[20:07:11] <karlpinc> What's a good application for
cross-platform screen sharing, with the debian screen as the one
being shared?
1569[20:07:54] <bifunc2> shtrb what are some ways to isolate, for
instance, the USB port or a particular directory (with sensitive
files) in debian only?
1599[20:25:45] <shtrb> karlpinc, Oh and I had been using in the
past HTML5 to rdp/vnc from guacamole, it was amazing experience, but
not all phones could work with HTML5 nicely
1630[20:41:25] <greycat> removed 3 and a half years ago
1631[20:41:38] <shtrb> time had passed fast
1632[20:41:49] <karlpinc> Guacamole does seem like the right
answer.
1633[20:41:55] *** Quits: TheSilentLink (~TheSilent@replaced-ip) (Quit: Good Bye! My bouncer has probably crashed or lost
connection to the internet...)
1636[20:43:11] <gordonfish> Anyone know why a mounted disk
(sdcard, as f2fs, which I can browse and write to in at it's
mount point) would not show up in the `mount` or `df -a` ? Is this a
namespace thing?
1637[20:43:25] <shtrb> karlpinc, The only minuses in guacamole
that I had seen was that not all phones supported HTML5 properly
,and that file transfer was done by a very quirky flow
1638[20:43:41] <jelly> gordonfish, are you running mount as
normal user or as root
1642[20:46:12] <gordonfish> jelly: I have
'UUID=6895bca2-ce0c-4a8b-8ba0-8e904366f395 /data f2fs
defaults,auto,nofail,x-systemd.automount' in my fstab and
systemd mounts it when I access it
1646[20:50:55] <karlpinc> shtrb: (The other down-side is the
java-ness of it all. Oh well. :-)
1647[20:51:17] <shtrb> Let us keep this channel family friendly
:P
1648[20:51:50] <shtrb> it can work under apache , you don't
have to use tomcat, in fact I was using it with only apache server
and the it's own binaries
1652[20:56:57] <shtrb> Would anyone remind me what extra action I
need to do after putting files directly in /var/cache/apt/archives
(what was not installed) that when I do apt-get update apt would
take them into account and would not try to search them on the DVD
?,
1671[21:11:02] <ryouma> my computer crashed this morning before i
got up, last /var/log/message being a backup and "rsyslogd was
HUPed". can more information be gotten?
1673[21:12:16] <shtrb> /var/log/syslog ? permenant storage from
systemd was configured ? kernel ?
1674[21:13:48] <greycat> usually if there's a hardware
crash, there is no chance for the OS to write anything to disk
1675[21:17:08] <ryouma> shtrb: thanks. ok i have in syslog a
bunch more last one being "Feb 5 09:45:01 napoleon2 CRON[6721]:
(root) CMD (command -v debian-sa1 > /dev/null &&
debian-sa1 1 1)" (which i don't need but sysstat it has
parallel in it) i hav enot configured systemd. idk what you are
asking re kernel; it is oldstable. i'm guessing maybe a
hardware crash.
1677[21:19:31] <ryouma> also, i have /root and /home encrypted.
for some reason i get asked for pass phrase twice (async it seems,
surprisingly to me) and home does not get mounted (async, surprising
to me) until a few minutes after i have already logged in as user
and it failsafes to / because /home/user isn't there yet. is
this probably me setting up something wrong?
1741[22:24:29] <karlpinc> ryouma: The delay is odd, or seems odd
to me. I'd look into that. Start, as always, with the logs. If
you have not already, configure systemd to do boot logs.
1765[22:38:37] <ryouma> karlpinc: thanks. i think i have notes on
configuring systemd to do boot logs someplace, and will do that when
i can. idk what hte second pass pharase is for, root or home (i am
guewssing it is maybe home, but i do have a key for it so there
should be no pass phrase). and then the delay. (messages appear
before the second prompt so it is not visible.)