40[00:44:39] <h-var> I have an issue with printer in debian. I
downloaded the driver from debian's repository that says it is
exactly for my printer model, but when I am trying to print, it
gives an error "filter failed"
41[00:44:59] <h-var> how can I fix this error for the driver I
downloaded for my printer from debian's website?
51[01:02:24] <mrjpaxton[m]> Hey, I can figure out how to
install Debian with either BIOS or UEFI (or with a GPT BIOS
partition), as well as with a GUI or without, but I am still having
trouble with figuring out how to install with software RAID 1 and
LUKS encryption, since I'm not used to it. Can anyone help?
I'll explain more...
52[01:04:07] <mrjpaxton[m]> So I have a USB drive with a folder
with keyfiles, let's say
"/media/keymaster/hdd-keys/pc-unlock.key" which is a
randomly generated 2048-bit key, and that's the ONLY key I want
to use. So when I plug it in, I'd like it to automount (maybe
with crypttab) and unlock the root Btrfs partition, or anything else
specified in crypttab.
53[01:04:46] <mrjpaxton[m]> But then I also want to use RAID 1
with my OS, so I'd have redundancy with the encrypted block
storage.
54[01:05:18] *** Quits: Roedy (Roedy@replaced-ip) (Quit: See you IRL!)
153[03:09:15] <ectospasm> okee: Are you using the UEFI? Or is it
in compatibility mode? I'd recommend not using compatibility
mode, then you don't have to waste space on a BIOS partition.
154[03:09:32] <mrjpaxton[m]> BIOS vs UEFI does have its upsides
and downsides. Sometimes a few things change or don't work the
same way when switching between modes. You can also have a UEFI
system with BIOS "compatibility mode" which will run UEFI
to POST into BIOS, then continue with BIOS. But yes, GPT always has
significant advantages over DOS layout. My favorite feature is
actually partition labeling, separate from FS labelling.
157[03:10:42] <mrjpaxton[m]> Then you can use mount options like
PARTLABEL="foobar", and not deal with the labelling
restrictions that FAT32, ExFAT or the others have.
158[03:11:22] <mrjpaxton[m]> Some people do confuse partition
labels with regular FS labels.
163[03:13:46] <mrjpaxton[m]> Also, if you're working with
GPT, you want to use "gdisk" every time you'd use
"fdisk", but "parted" and "gparted"
can work with both DOS and GPT layout.
172[03:17:48] <okee> ectospam> A message appeared in the
installer indicating a previous file system created in compatibility
mode was present, and I selected the option to "force" EFI
173[03:17:57] *** Quits: trly (~trly@replaced-ip) (Remote host closed the connection)
311[06:20:48] <abff> In a bash script, whats the most robust
debian way to identify a file's mime extension, rename the file
with said extension, and return the new file name to be referenced
later. I have gotten 2/3rds of the way there with exiftool but I
can't figure out how to put the file's new name to stdout.
312[06:22:19] <abff> I've also just used `file --ext |sed
magic` but it always seems to fail for gifs and will likely fail for
other file types
326[06:38:35] *** Quits: Conradish006 (~Conradish@replaced-ip) (Quit: Guess I left :D)
327[06:40:21] <mrjpaxton[m]> I would ask in #bash too. But as
far as I know, a lot of the standard `exiftool` and `file` command
options and utilities will work just about the same in any Linux OS.
I only know that Debian (and Ubuntu) uses the `mime-support` package
which provides "/etc/mime.types". So I am not sure if MIME
magic recognition is radically different between OSes. But I'd
wager they're all pretty close.
351[07:06:03] <mrjpaxton[m]> I should make a correction and say
that once Debian 11 Stable (Bullseye) is released, the package will
transition from `mime-support` to `media-types` apparently... Hmm...
352[07:07:38] <abff> oh well
353[07:07:49] <abff> can't wait for this script to break
364[07:20:59] <okee> I seem to be having trouble getting the
10.7.0 stable release installer to detect the Linux drivers found at
this URL
replaced-url
365[07:22:21] <okee> I tried both of the USB ports in the XPS 15
7590, and get an error indicating it can't detect it. One
thought is that the drivers might be buried in a folder than the
installer can't detect, but I don't know which if any of
the drivers will work with my ethernet card. Dell India support
doesn't know anything about my hardware.
366[07:22:41] <okee> I will check with Dell in the morning. They
really should put the drivers on their website. All they have is
windows drivers.
389[07:57:08] <abff> okee it might be the drivers for your card
are non-free? I usually test a machine with debian-live xfce
non-free variant if I can't figure out what's wrong. Then
once I've pinpointed the exact driver I need I get that
specific one and load it over usb during install
427[09:01:02] <mixfix41> compiling chromium on slackware arm.
see that the directory i had to update for a patch in the SlackBuild
arm-linux-gnu change to arm-linux-gnueabihf in
tmp-chromium/chromium-88.0.4324.190/build/linux/debian_sid_arm_sysroot/usr/lib/arm-linux-gnueabihf/libsqlite3.so.0
hopefully i can compile this !
493[10:25:32] <EugenMayer> Hello. Trying to get an entire buster
upgrade unattended on my stretch fleet. Seems to work out pretty
fine until i get
`/dev/disk/by-id/ata-VBOX_HARDDISK_VB6aeffc8f-e0621d34 does not
exist, so cannot grub-install to it!` when grub-pc is upgraded. The
reason is, i suggest, that i have build all those boxes using
`packer` on a virtual-box
494[10:25:32] <EugenMayer> provisioneer and later shipped for
qemu and others. no somehow grub has a pointer to the old disk
identifier. Can i somehow change this via cli preseed or similar
e.g. setting `d-i grub-installer/bootdev string /dev/sda` or similar
prior runniung ` DEBIAN_FRONTEND=noninteractive apt-get -qy -o
Dpkg::Options::='--force-confdef' -o
509[10:35:00] <MegaCarp> i'm having a problem with user
trying to login into a xrdp session, if done from windows remote
desktop it allows to login, to enter the pw but afterwards it's
just a blue screen. from linux (remmina) it's just a mess of a
screen or just black. here's what i think is the relevant part
from /var/log/xrdp-sesman.log :
replaced-url
510[10:35:13] <MegaCarp> it's something about "X
server for display 62 startup timeout"
526[10:50:37] <dpkg> For GRUB: 1) press 'e' to edit
the kernel setting in the grub command line (add
'init=/bin/sh' to the end of it) 2) 'fsck' your
root file system, 3) 'mount -o remount,rw /', 4)
'passwd root' 5) 'mount -o remount,ro /' 6)
'reboot -d -f' (exec /sbin/init should work); For LILO: 1)
'Linux init=/bin/sh' at the LILO boot prompt (hold Shift
while booting), steps 2-6 are the same; For yaboot: 1) 'Linux
init=/bin/sh' at yaboot prompt.
565[11:57:30] <EugenMayer> Gramcore it can be what ever you
consider is public.
566[11:58:15] <ratrace> also a weird choice of words,
"permanent"
567[11:58:17] <Gramcore> EugenMayer, so not the internal one i
have for the device? like 192.168.1.5
568[11:58:25] <EugenMayer> If it is a box with a WAN interface,
you might set this to your WAN ip, thus public ip. If this box is on
the private lan, you set this to your private ip. How you obtained
your ip does not matter
569[11:58:26] <Gramcore> with netmask 255.255.255.0
575[11:59:39] <EugenMayer> If your box is e.g. 1:1 natted you
might have a private ip but set the public ip here. It really
depends on what you want the system to think your permanent ip is
576[11:59:52] <Gramcore> EugenMayer, i have setup an nginx
server, with a domain name, basically a website, along with an xmpp
server. can i just let that /etc/hosts be as it is? 127.0.1.1
hostname ?
577[11:59:53] <tete_> hi, i try to install debian on a 1blu
server ... but it seems impossible and i have no idea wtf goes on.
this takes me now more than 2 days :E the error is here:
replaced-url
578[11:59:56] <Gramcore> or will it break anything?
579[11:59:58] *** Nigalia is now known as bootdisk
581[12:00:14] <ratrace> it's not about a
"permanent" IP at all. it's the IP you want the name
you put there to resolve to
582[12:00:14] <tete_> anyone an idea what goes on? the .iso i
have is correct, i checked its md5, but of course i dont know whats
on the hoster side. i did not get an error during ftp upload
589[12:02:05] <EugenMayer> ratrace it is not that easy. if that
entry matches your hostname and , the first one in there will become
the reverse-dns entry for your system for your own hostname. so
`1.1.1.1 foo.org foo` with hostname foo will let the system
underestand, that 1.1.1.1 is your rdns entry for your hostname. this
makes a quiet a difference for stuff which wants to
590[12:02:05] <EugenMayer> know that
591[12:02:17] <EugenMayer> be it kerberos, proxmox or alike
592[12:02:43] <Gramcore> ratrace, so i should change 127.0.1.1
to the ip i access the machine from the internet (the one i use for
ssh etc), and hostname like: gramcore.website.com
593[12:02:44] <EugenMayer> ratrace it does only identify it, if
the entry matches the hostname
594[12:03:12] <EugenMayer> ratrace anyway you obviously seem to
know it already - not going to argue
595[12:03:30] <ratrace> if the entry matches the hostname,
there's no such lookup
596[12:04:01] <ratrace> a process wants to know its hostname. it
does a local reverse lookup based on IP it's using, because
there could be more than one.
598[12:04:27] <ratrace> then the /etc/hosts entries come into
play to tell the process what's the hostname associated with
that IP, or else it'd have to do a recursive rdns outbount
599[12:04:30] <ratrace> *outbound
600[12:04:51] <ratrace> so keep in mind you can have multiple
IPs, and the hostname identified depends on the IP for which it was
looked up by the process
602[12:05:35] <Gramcore> we have been using in linux the
loopback 127.0.0.1 or that debian 127.0.1.1 to resolve the hostname.
i don't get why anything else is needed. i saw some examples
that it might make a difference but those seem like edge cases
603[12:05:48] <ratrace> which means if you have public NAT, the
processes can't identify with it because that's not an IP
on the machine
604[12:06:21] <EugenMayer> ratrace AFAIK the first entry, if it
exists, matching the hostname you try to lookup (well depend on
nsswitch if /etc/hosts is used at all) is used as the result for the
response. so the order does really matter in there, that is why it
is up there in the top
605[12:06:21] <ratrace> Gramcore: depends on the context.
606[12:06:50] <ratrace> first entry, for the IP looked up for
607[12:07:06] <Gramcore> ratrace, yeah for my case for example,
i can't find what i should do. the archwiki doesn't
explain either, they link the debian manpage
608[12:07:23] <Gramcore> those seem more like opinions that
anything else
609[12:07:35] <ratrace> Gramcore: what are you even trying to
do? your original question was what that "permanent" IP
was
610[12:08:13] *** Quits: GNUtoo (~GNUtoo@replaced-ip) (Quit: Lost terminal)
611[12:08:28] <EugenMayer> Gramcore you configure an arch based
server with nginx and ask in the debian channel about what you
failed with, while what you fail with is at the very basic of it
all. Why do you use arch in the first place if you actually are a
learner in that field?
612[12:09:07] <MegaCarp> i'm having a problem with user
trying to login into a xrdp session, if done from windows remote
desktop it allows to login, to enter the pw but afterwards it's
just a blue screen. from linux (remmina) it's just a mess of a
screen or just black. here's what i think is the relevant part
from /var/log/xrdp-sesman.log :
replaced-url
613[12:09:14] <MegaCarp> it's something about "X
server for display 62 startup timeout"
615[12:09:47] <ratrace> EugenMayer: I don't see them saying
they're using ArchLinux :) and ArchWiki is _defacto_ mostly
distro agnostic linux documentation these days
616[12:10:02] *** Quits: dselect (~dselect@replaced-ip) (Quit: ouch... that hurt)
617[12:10:16] <Gramcore> this is my /etc/hosts
replaced-url
618[12:10:28] <EugenMayer> ratrace really? Did i missunderstood
him then? Blame me then, sorry
622[12:11:04] <ratrace> Gramcore: eh you actually own
random.com?
623[12:11:11] <Gramcore> EugenMayer, debian 10 buster. i dont
use arch for servers.
624[12:11:19] <tete_> no one who knows what could go wrong at my
debian installation?
625[12:11:21] <Gramcore> ratrace, an example, thats not my
domain
626[12:11:30] <ratrace> Gramcore: and the hosts file is not
okay. you need your public IP there with both fqdn and hostname
entries
627[12:11:44] <ratrace> Gramcore: ah then don't use
existing domains as an exmaple
628[12:12:00] <EugenMayer> ratrace i wondered. I rember me
asking a ubuntu question for my dang idiot ubuntu zimbra
installation back then (i run debian only, only for this box i have
to use broken-debian-buntu) .. got fairly burned in here ...
understandingly :)
642[12:15:18] <Gramcore> ratrace, and to clarify that ip you
used for an example the 123.45.67.8, is the one use use to ssh into
the box right? (ssh root@123.45.67.8)
648[12:15:55] <tete_> i can reproduce that - always
649[12:16:37] <tete_> i have that vps for more than 2 days now
and i am unable to install a debian .iso, thats ridiculous, no idea
what is broken on the hoster side... i never encountered such
problems
650[12:16:48] <tete_> and i work with linux now for more then 10
years
651[12:16:51] <ratrace> Gramcore: it's the (probably only)
public IPv4 of the server, so it's for whatever public access
outside of the 127.0.0.0/8
652[12:17:07] <ratrace> tete_: you should really ask your host
653[12:17:18] <tete_> he says he see's no problem - i
should check everything and create a report
654[12:17:28] <tete_> thats why i thought i ask here, maybe
someone knows whats causing that
655[12:18:02] <tete_> the only thing i could imagine is a broken
.iso
656[12:18:10] <tete_> because of the wierd errors
657[12:18:15] <tete_> or a memory problem
658[12:18:18] <ratrace> tete_: how are you even accessing the
console? is this a VPS with a virtual console?
659[12:18:26] <tete_> yes its a VNC
660[12:18:53] <ratrace> VNC to what? I'm not sure the
debian installer supports VNC connections
661[12:19:14] <tete_> the VNC connects to the hoster, he has
some magic that connects to my machine, i guess its a virtual
console
662[12:19:30] <tete_> thats pretty common (at least for those
hosters i know like netcup, 1blu etc.)
663[12:20:10] <tete_> i am not that familiar with
virtualization, maybe thats already on-board with kvm
664[12:21:31] <Gramcore> ratrace, another question for some
reason on my debian 10 system the ipv6 entries are like i show in
the paste:
replaced-url
665[12:23:03] <ratrace> tete_: accessing your virtual
server's console over VNC is anything but common. As for this
particular problem ... it does look like corrupt installer. Why
don't you use a hosting company with preinstalled debian
images?
666[12:23:26] <tete_> ratrace, because i am student and that
hoster had an offer: 1 EUR / mtl. for a good vps
667[12:23:31] <tete_> so i thought i give that a try
668[12:23:33] <ratrace> Gramcore: no need, those are just
convenience names for IPv6 standards
669[12:23:49] <ratrace> tete_: and this is what you get for
1€
670[12:23:52] <tete_> otherwise i would use my good ol'
hoster netcup, but then i have to pay about 10 EUR / monthly for the
same stuff
671[12:24:07] <ratrace> tete_: or try hetzner they go as low as
2-3€ and are quality, with preinstalled debian images
672[12:24:09] <tete_> ratrace, hm ya... but maybe its just a
small problem
673[12:24:26] <tete_> i wrote them when the problem is not fixed
until monday i abort that order
675[12:24:47] <ratrace> tete_: you're expecting too much
from 1€/mo hosting lol
676[12:24:49] <tete_> its a 8gb 6vcore system... for 1 eur,
thats so sad
677[12:24:56] <tete_> ratrace, thats a special offer
678[12:24:57] <ratrace> just reading that email on their part
costs more than 1€
679[12:25:08] <tete_> normaly that would be about 5 or 6 EUR
680[12:25:14] <ratrace> tete_: 8GB 6 core system is a DAMN LIE
for 1€
681[12:25:21] <tete_> lol its too good eh? ;)
682[12:25:25] <ratrace> it's impossibly good
683[12:25:55] <tete_> hold on, my bad, its
replaced-url
684[12:26:09] <tete_> but still too good - but its a special
offer, only valid for a couple of days hm
685[12:26:20] <ratrace> I mean... I'm not saying you
won't actually get 8GB of RAM and /proc/cpu would list 6 cores
.... but you'd be one of hundreds crammed on the host, in order
to gain any ROI on the hardware...
686[12:26:37] <tete_> i create a report, tell them my
findings... if they dont get it working until monday, i am forced to
switch, because i need a cheap vserver (student) for my project
687[12:26:49] <ratrace> use hetzner
688[12:27:08] <tete_> i write hetzner if they offer special
prices for poor students ;)
692[12:29:12] <tete_> i know what you are going to say ;)
693[12:29:16] <ratrace> there goes my asinine attempt to point
at the source of monies for quality hosting :)
694[12:29:19] <tete_> and i only eat rice with ketchup :P
695[12:29:30] <tete_> and i live below a bridge ;D
696[12:29:37] <tete_> i get that point... i write to hetzner
697[12:29:46] <Gramcore> EugenMayer, i have been using linux for
10 years or more. i use gentoo on my desktop currently, used arch
before etc. never had to mess with the /etc/hosts file. most of
these distros also don't give a lot of information about
/etc/hosts (or didn't at least). the only real information i
found from debian manual
698[12:29:48] <tete_> i still give 1blu the chance to fix that
until monday, then i go to hetzner
699[12:30:01] <ratrace> tete_: just one fair warning about
hetzner
700[12:30:31] <ratrace> their hardware is good, their cloud
offering as well, but otherwise they're zero support. you get
the hardware, and that's it. if you had this problem with them,
you'd have NO support
701[12:30:51] <ratrace> the only support that exists is for
reporting hardware issues and they're quick to fix those
702[12:31:14] <ratrace> that's why they can be cheap AND
have quality hardware. they don't waste time and personnnel,
everything is automated.
703[12:31:26] <tete_> ok, should be no problem, as long as the
hardware and upload process works
704[12:31:42] <ratrace> tete_: upload? of ISO?
705[12:31:44] <tete_> when i upload the .iso to 1blu i always
get a TLS connection aborted - so i am very sure its corrupted data
706[12:31:46] <tete_> jep
707[12:31:56] <ratrace> wait... why. they have preinstalled
debian images
708[12:32:01] <tete_> 1blu not
709[12:32:07] <tete_> not with kvm
710[12:32:20] <tete_> or i did not find that
711[12:32:20] <ratrace> I don't think they allow random OS
ISO uplods. OVH does. with hetzner you can use debootstrap from
their rescue env if you really want a custom installation
712[12:32:26] *** Quits: wilson (~wilson@replaced-ip) (Ping timeout: 264 seconds)
713[12:32:30] <tete_> they do, netcup also does
714[12:32:46] <tete_> and that other cheap hoster also does... i
cant remember its name
715[12:32:52] <ratrace> pretty much every hoster on the planet
offers preinstalled debian images. those that do not ....
shouldn't call themselves hosting companies
716[12:33:13] <tete_> i like that feature tbh
717[12:33:29] <tete_> it doesnt matter for me if a hoster
provide images or not, as long as i can upload whatever i want
718[12:33:41] <ratrace> you can't get that for 1€.
719[12:33:50] <tete_> remember: thats a special offer
720[12:33:56] <tete_> only valid for 4 days or so
721[12:34:01] <tete_> normaly it would be 5 EUR
722[12:34:06] <ratrace> 1€ are super constrained predefined
environments. and you want a generic environment into which you
install any ISO? not gonna happen
723[12:34:09] <tete_> or 6 dunno
724[12:34:52] *** Joins: wilson (~wilson@replaced-ip)
733[12:47:43] <EugenMayer> ratrace but there installimage is so
useful and powerful - i never had a reason to dare to do so with
such a barebone. installimage then chef for everything else
734[12:47:49] *** cdown_ is now known as cdown
735[12:49:22] <MegaCarp> i've inherited a server with xrdp
and that also makes available over remote desktop windows sessions.
how should i go about finding out how are windows sessions
implemented?
755[12:59:57] <ratrace> MegaCarp: that's confusing very
different technologies in the same sentece, so maybe you should
start from the beginning and explain in detail what's the
environemnt, what it does, and what do you want to achieve.
757[13:02:13] <ratrace> MegaCarp: unless you mean ... vnc access
by qemu-kvm to the VM, and the using virsh to start the display? but
that's VNC, not rdp
763[13:07:44] <MegaCarp> ratrace: server's debian 10,
DE's xfce. thinclients connect to the server using xrdp. most
thinclients are debian boxes connecting over Remmina, using RDP. A
few are Windows boxes using WinRD. a thinclient can connect to a
debian session using one server local (or remote, actually) address
or it can connect to a Windows session, again using RDP, over a
different address. I know that both are hosted on a single physical
server. I'm very unsure how
764[13:07:44] <MegaCarp> exactly. I'm not sure where to
start looking. I'm just guessing that Windows is hosted
virtually - but how I check for that? Google says to check with
virsh - virsh isn't installed.
765[13:12:54] <ratrace> virsh is just one tool, libvirt shell.
virtualization on linux can be done in several way. first, determine
on the host if there's a virtualization process running for
VMs.
776[13:15:24] <ratrace> eh no... that's the grep itself
777[13:16:01] <ratrace> you could just dump the entire ps -axu
into a pastebin, we can look it up for clues. but I have to warn
you, that output could be privacy or security sensitive
778[13:16:20] <MegaCarp> i understand, i'll try and
sanitize
809[13:38:55] *** igrtrrt_ is now known as igrtrrt
810[13:39:15] <MegaCarp> ratrace: any obvious way to find
process names for a given application?... or should i just go dig
through processes from root and a few important users? FWIW - no
qemu, no vbox or virtualbox
1053[17:35:28] <vezult> They support all the latest ubuntu,
centos, RHEL, etc
1054[17:35:37] <vezult> debian, they support 6, 7, 8
1055[17:35:43] <vezult> and they seem to be serious about that.
1056[17:35:57] <vezult> the only way I can make it work (that
I've been able to figure out)
1057[17:36:19] <vezult> is to use an ancient 3.16 kernel from 8
setting it to the default kernel
1058[17:36:22] <vezult> upload
1059[17:36:27] <vezult> err, import
1060[17:36:47] <vezult> then switch to a modern kernel for each
instance
1061[17:37:09] <petn-randall> !enter
1062[17:37:10] <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.
1063[17:37:10] <vezult> that works (so far), but is a huge pain.
1064[17:37:31] <petn-randall> vezult: Check out
replaced-url
1065[17:37:41] <petn-randall> I vaguely recall them being there,
too.
1066[17:37:58] <vezult> petn-randall: yes, I know they exist. but
as I said, I build custom ones for our use, because reasons.
1067[17:38:34] <imMute> vezult: can you build yours off the ones
provided by debian?
1068[17:39:02] <vezult> no, not currently.
1069[17:39:59] <petn-randall> vezult: I must admit I never had to
build custom Debian AMIs, what technical things are preventing you
to boot them? Does it fail on import or when spinning them up?
1075[17:43:26] <petn-randall> vezult: TBH I always used the stock
images, and cloud-init to customize them. If you can't go that
way for $reasons, you might want to ask the AWS support why they
prevent you from booting your custom AMIs.
1115[18:12:43] <ddsys> i installed an older version of
kconfi-frontends but the kconfig command is not found. It configured
and make and make install just fine. any ideas?
1116[18:14:51] <jelly> ,file bin/kconfig
1117[18:15:13] <jelly> judd, ping
1118[18:15:14] <judd> pong
1119[18:15:26] <judd> Search for bin/kconfig in buster/amd64:
kconfig-frontends: usr/bin/kconfig
1120[18:15:45] <jelly> ddsys, which debian release and which
version of kconfig-frontends package?
1167[18:45:11] *** Quits: llh (~coke@replaced-ip) (Remote host closed the connection)
1168[18:45:59] <sney> consistency with previous debian versions,
probably. ifupdown has been the standard in debian since well before
the switch to systemd.
1174[18:47:55] <jelly> Gramcore, is there an "eth0"
interface at all, btw?
1175[18:48:37] <sney> if you run 'ifup -a' directly,
there might be more useful error output
1176[18:48:54] <Gramcore> jelly, its a vps so i physically i
don't have access but there are two interfaces (doing ip addr):
lo eth0 and eth1. eth0 has ipv4 and eth1 ipv6
1177[18:49:23] <Gramcore> actually eth0 is the ipv6
1178[18:50:10] *** Quits: dftxbs3e (~dftxbs3e@replaced-ip) (Remote host closed the connection)
1179[18:50:52] *** Quits: szorfein (~daggoth@replaced-ip) (Remote host closed the connection)
1215[19:19:11] <Gramcore> jelly, well this
/etc/network/interfaces was autogenerated by the cloud
infrastructure i use my vps. i uploaded the debian iso and they
generated this file
1216[19:19:32] <jelly> it autogenerated crap.
1217[19:19:35] <Gramcore> should i leave it as is and wait for
the ifupdown bugfix? or edit it you think?
1218[19:19:54] <spacedust> why cant autogenerated crap be good in
2021
1219[19:19:56] <jelly> thta depends on whether eth0 exists always
or not
1220[19:20:26] <Gramcore> jelly, every time i reboot it exists.
it has eth0 as ipv6 and eth1 as ipv4
1221[19:20:32] <jelly> Gramcore, if it always exists, leave auto
lines. If it appears after a while, leave just allow-hotplug.
Don't have both.
1222[19:20:37] <Gramcore> but i don't have other information
on the networking of the vps
1223[19:20:45] <greycat> If it's not correct, fix it.
1230[19:21:54] <Gramcore> so you say only one of those should be
allowed only: auto eth0
1231[19:21:55] <Gramcore> allow-hotplug eth0
1232[19:21:58] <Gramcore> and both is wrong
1233[19:22:05] <greycat> it makes no sense to have both
1234[19:22:08] <jelly> that is miy understanding yes
1235[19:22:40] <jelly> do you ssh over eth0 or eth1? :-)
1236[19:22:48] <Gramcore> jelly, well i login via ssh using ipv4
which is eth1
1237[19:23:28] <jelly> okay, then for the time being don't
touch nything about eth1, just comment out allow-hotplug eth0 and
reboot and see what happens
1238[19:23:31] <Gramcore> also why doesn't ifupdown complain
about eth1 also, that has the same syntax mistake
1239[19:23:47] <jelly> no idea
1240[19:24:41] <jelly> ipv6-only ifaces might be managed in a
different way. Typically a cheap VPS would only have a single
interface and ipv4 and ipv6 both on that.
1270[19:36:27] <greycat> setting up the dot-link files to tie MAC
to iface name seems like a really good idea
1271[19:37:01] <Dagger> jelly: I had my interface names change
when removing a SAS expander from my desktop (and I had no idea what
the interface name was going to be either before or after I removed
it), so... yeah. agreed
1272[19:37:04] <jelly> it doesn't work when net.ifnames=0 is
in effect, I still have to test if they work without those boot
params
1274[19:37:45] <jelly> Dagger, probably the SAS expander had a
tiny PCI bridge inside, and why wouldn't it. Your PCI bus
changed. You lose!
1275[19:38:56] <Dagger> ah, but nope. it did plug into a PCIe
slot but only for power, not for PCIe
1276[19:39:22] <jelly> where did it plug then, just the sas
controller?
1277[19:40:18] <Dagger> it was plugged into one of my x16 slots
for power, which was enough for my motherboard to split the other
x16 slot to x8/x8 even though the expander isn't a PCIe device
1278[19:40:37] <Dagger> and that shuffles the PCIe slot IDs along
by one
1279[19:40:59] <jelly> that's something FDO people did not
realize might easily happen
1281[19:42:37] <jelly> I would have guessed the other way, that
the board detected a just-power device and made one PCIe slot _less_
available, not one _more_
1325[20:10:06] *** Quits: frostschutz (~frostschu@replaced-ip) (Remote host closed the connection)
1326[20:12:06] <Gramcore> sney, i enabled allow-hotplug, and
comment out auto. now it works fine
1327[20:12:19] <Gramcore> what does this mean?
1328[20:12:43] <greycat> ... that the interface springs into
existence later, not at boot time?
1329[20:12:57] *** Quits: dbristow (~dbristow@replaced-ip) (Remote host closed the connection)
1330[20:13:02] <sney> allow-hotplug is supposed to bring up the
interface when the device is detected by udev. on a vm, your
interfaces are virtual, so they should always be there, but
apparently not here
1335[20:13:43] <Gramcore> should i do the same about eth1? right
now it has both auto eth1, and allow-hotplug eth1
1336[20:14:00] *** wyre_ is now known as wyre
1337[20:14:00] <sney> you definitely never need both. that much
is clear from the interfaces(5) man page.
1338[20:14:16] <terps> Would anyone happen to know how I would go
about controlling my gpu fan speed with nvidia driver 460.39? All
the pre-existing guides seem to no longer apply on debian 10 or
I'm missing something
1339[20:14:17] <sney> and yes, I'd say do the same thing
with both
1340[20:15:18] <hendursaga> So, I'm trying to connect to my
RPi via a USB to UART adapter and I can't screen /dev/ttyUSB0
because some getty process is running - how do I disable that?
1341[20:15:41] <greycat> ask your OS channel, probably
1342[20:15:47] <hendursaga> No no no..
1343[20:15:56] <greycat> it's Raspbian, right?
1344[20:15:59] <hendursaga> I'm on Debian Stable, trying to
disable that service on my laptop.
1348[20:16:19] <terps> my issue specifically is there's no
xorg.conf for me to put the coolbits string into I'm not sure
how to proceed
1349[20:16:23] <greycat> well, undo whatever you did to enable
that getty, either in /etc/inittab if you're running sysvinit,
or in some systemd unit if you're running systemd
1350[20:16:33] <Gramcore> commented out auto eth0, and auto eth1.
leaving allow-hotplug for both
1351[20:16:36] <hendursaga> I never enabled it - it came with the
installation?!
1352[20:16:42] <Gramcore> lets see if i will be able to ssh now
:D
1353[20:16:54] <sney> terps: you can put xorg.conf snippets in
individual files in /etc/X11/xorg.conf.d/
1354[20:16:57] <greycat> if it's unique to the ARM versions
of Debian then I don't know anything about it
1356[20:17:26] <terps> sney thanks so much I will give that a go.
1357[20:17:36] <sney> terps: xorg.conf doesn't exist by
default on debian because modern X should autodetect almost
everything. but files in that directory can override little bits of
it.
1358[20:18:30] <hendursaga> greycat: No no no... I'm trying
to connect to a FreeBSD image running on my RPi - the client has
NOTHING to do with Linux, I'm trying to get my host client,
Debian 10, to connect, but the agetty process is hogging the serial
port.
1359[20:18:57] *** Quits: fra (fra@replaced-ip) (Quit: Leaving)
1390[20:28:40] <goose> When going to resize a lvm partition, I
accidentally did 'lvreduce' before 'resize2fs',
and now my superblock for that partition is busted. Any way to fix
this?
1437[21:16:49] <sney> Gramcore: considering the situation with
interfaces, I think you're going to find lots of stuff in that
image that doesn't make sense.
1438[21:16:59] <sney> probably initially configured by someone
inexperienced
1446[21:24:33] <mentor> goose: You might be able to reallocate
the physical extents back to the logical volume, to be able to
recover the data. Assuming they haven't gone anywhere else
1448[21:25:48] *** Quits: raekuul (~pokota@replaced-ip) (Remote host closed the connection)
1449[21:25:48] <jhutchins> Gramcore: I've worked in places
where production server, like high traffic Oracle servers, had
obviouly been built by some Windows admin and had a full GUI desktop
installed and active, complete with mp3 player. There may habe been
a shortage of cluse among those building the VPS image.
1450[21:27:08] <Gramcore> jhutchins, lol
1451[21:28:29] *** Quits: Grldfrdom (uid391113@replaced-ip) (Quit: Connection closed for inactivity)
1452[21:28:45] <Gramcore> did they have the windows xp wallpaper?
the one with the green fields?
1453[21:29:10] <jhutchins> RedHat seems to be pushing this idea.
You don't have to hire those expensive Linux experts, any
Windows admin can manage a RedHat server farm.
1454[21:29:40] <Gramcore> redhat has gui tools. like ansible
tower
1455[21:30:09] <jhutchins> Gramcore: I don't know, they
didn't have remote desktop enabled and the one monitor in the
datacenter was on the other side of the room.
1456[21:31:20] <Gramcore> how do people become windows admins?
are there microsoft workshops?
1457[21:31:36] <Gramcore> never used windows server willingly
1458[21:32:28] <sney> there's a certification program, but
in practice they usually get promoted up from desktop support
1459[21:32:40] <Gramcore> a friend who works at a bank tells me
all their infrastucture is windows server, and the sys admins
can't open up a terminal to do basic things
1477[21:58:50] <jhutchins> When I did it there were $50 books
that would give you a pretty good grounding. I kind of picked it up
along the way, being the guy who would work on PCs when being paid
to tend the printer. The DOS 3.5 manual was really good for
understanding the DOS layer, after which overlaying Windows
wasn't that hard.
1478[22:01:28] <abff> Gramcore: I mean to be fair if I had to do
anything in powershell on windows I'd be 96% useless
1479[22:01:49] <jhutchins> I worked with a variety of systems,
from mainframes to Apple IIs and a Commodore 64 before I really
started doing PCs, and DOS on PCs. I was lucky to have pretty good
knowledge at the point where companies were realizing that no, the
receptionist can't manage the whole office's computers and
the Novel server.
1480[22:07:03] *** Parts: fra (fra@replaced-ip) ("Leaving")
1481[22:07:53] <Gramcore> abff, supposedly you can do things in
powershell, so those windows admins from microsoft should be able to
1482[22:08:00] <Gramcore> but they just use guis only
1493[22:29:18] <jhutchins> I was a pretty deep MS expert, but my
knowledge pretty much quits at NT4. I did a little 98 support for
friends, but I wouldn't want to be responsible for it now.
1494[22:29:55] <dfcnvt> I'm getting fuming and irrational. I
- do - not - like - Files - application. It's very limiting and
I'm starting to hate it. I've tried to see if I can
install nautilus but it seem to claim that it's already
installed...And yet I can't seem to open nor use it.
1495[22:30:26] <jhutchins> dfcnvt: Are you sure that's not
what "Files" is opening?
1496[22:30:58] <sney> customizing gnome is difficult. if you want
a gnome-like desktop environment with any kind of different
behavior, try mate or xfce4 or cinnamon instead
1497[22:31:32] <dfcnvt> The "Files" aren't going
to let you navigate with something like, "/home" or
"/etc" It's unsearchable.
1498[22:32:33] <dfcnvt> Here's an example >>
replaced-url
1499[22:33:01] <oxek> looks like some gnome3 abomination
1500[22:33:35] <jhutchins> dfcnvt: That's expected behavior
on manyh systems - ordinary users don't have access to those,
just $HOME
1501[22:34:05] <greycat> Ordinary users can read *most* of the
files in /etc.
1502[22:34:21] <greycat> And other people's homes may be
opened or not. They're opened by default.
1503[22:34:31] <oxek> if they don't have access, then the
Files application should say so. If it hides errors from users, then
it's a bad app
1504[22:34:52] <jhutchins> dfcnvt: You're not running a GUI
as root, are you?
1505[22:35:05] <dfcnvt> It's smart to curate the dumb user
-- preventing any damage to any critical files......That'd make
sense why such thing like this exist -- but since I'm a
different user...and I'm not liking this type of limitation.
1506[22:35:36] <dfcnvt> jhutchins: No, I'm just a regular
user running with any gui.
1508[22:36:30] <greycat> File permissions already prevent damage
by not letting you *write* to the files in /etc. But most of them
should be readable. There is no reason for a GUI app to hide them.
1509[22:36:35] <dfcnvt> Will I assume it'll have more access
if I run with something like, "gksudo Files" or
"gksudo nautilus"
1510[22:37:43] <Mister00X> dfcnvt: try other devices it
"should" work there
1522[22:42:09] <Mister00X> and there I can read those directories
1523[22:42:44] <dfcnvt> I'll just say it -- Files is
garbage, I'm unsatisified. Is there another alternative File
Explorer that can gives me an ample functionality with no
limitation?
1532[22:44:11] <dfcnvt> Okay, thanks guys. I"ll research and
check it out.
1533[22:44:24] <dfcnvt> (Don't know if I should switch to
different desktop environment...?)
1534[22:44:36] <jhutchins> dpkg: status
1535[22:44:36] <dpkg> Since Fri Feb 26 12:05:07 2021, there have
been 0 modifications, 5 questions, 0 dunnos, 0 morons and 0
commands. I have been awake for 9h 39m 29s this session, and
currently reference 40857 factoids. I'm using about 44468 kB of
memory. With 0 active forks. Process time user/system 3.82/0.32
child 0/0
1546[22:56:03] <dfcnvt> ^^ Do it, you'll thank me later.
1547[22:57:31] <sney> that's the cinnamon one? neat
1548[22:57:34] <jhutchins> dfcnvt: I'm ok with thunar. Can
be a little awkward on some things, but it gives me the power to
totally screw up my hierarchical storage.
1549[22:58:49] * Mister00X uses dolphin, but wont reccomend it for GTK
based DEs