30[00:06:47] <jolt> No idea why you wouldn't allow it, but
it's possible to override for groups or users with Match foo
Subsystem sftp /bin/false like in this example:
replaced-url
57[00:10:14] <mentor> cws: I appreciate your existential crisis
here
58[00:10:33] <Chop> can you explain how to stop
59[00:10:44] <jolt> Chop: you know they can still use scp
right?
60[00:10:51] <Chop> yes
61[00:10:57] <cws> Not until you explain what you think
you're actually going to achieve. You aren't going to stop
people from doing sftp-like things by stopping sftp.
62[00:10:58] <jolt> Chop: I just did above
63[00:11:26] <cws> Chop: scp still works. I don't think
you have any idea what you're doing, to be truthful.
64[00:11:35] <Chop> cws, can you explain how to stop it, or no
66[00:11:55] <cws> Chop: You seem to not understand even basic
concepts. You're trying to do something that serves no purpose
without any explainable reason. So no, I'm not going to help
you.
67[00:12:16] *** Parts: Chop (~email@replaced-ip) ()
81[00:20:22] <H-var> eyes. I remember eyes. Pictures came with
eyes. A painter in my dream. Tell me what you see. A tourist in a
dream. A visitor it seems. A half forgotten story. Where do I
belong? Tell me what you s saw? Tell me what you felt? I know
I'm just a clone. Do you know you're too?
114[00:50:58] *** Quits: deb (~deb@replaced-ip) (Ping timeout: 276 seconds)
115[00:52:57] <Gramcor> can i use nested placeholders in my .j2
templates? for example {{ email['{{ hostname }}'] }}
instead of {{ email['gramcor'] }} , where hostname:
gramcor
123[00:58:44] <Randolf> How can I set up Debian to be a Wifi
server? When I search Google, all I keep finding is instructions
about setting up Debian as a client. Thanks.
138[01:14:13] <Randolf> nolan-___: I want to set up Debian to
provide a wireless network that user can login to and get internet
traffic routed from. That's step one. Then, after I get a basic
configuration working there (so that Debian is behaving like a
typical consumer wifi router that people buy in the stores, like a
LinkSys home router), I want to work on setting up a custom login
web page like some restaurants, shopping malls, and other businesses
have set up for their
139[01:14:13] <Randolf> clients to login to.
140[01:14:25] <Randolf> tomman: Yes.
141[01:15:13] <Randolf> I've heard of people doing this
sort of thing with a Raspberry Pi. I have some PCs here with
wireless NICs that I'd like to experiment with.
142[01:16:08] <Randolf> (These PCs run Debian without any
trouble, and have Intel or LinkSys NICs that Debian can recognize
and use.)
143[01:16:37] <tomreyn> the second thing you described is called
a captive portal
144[01:17:06] <Randolf> Thanks to tomman I just discovered
"hostapd."
145[01:17:21] <Randolf> I will research into Captive Portals on
Debian then, thanks tomreyn. :)
146[01:18:07] <Randolf> Ah, for Captive Portal setup, this seems
to be the proper starting point:
replaced-url
204[02:31:03] <Kobaz> having an issue.. getting an email every
night from cron: /etc/cron.daily/apt-compat: Exception:
'NoneType' object has no attribute
'dependencies' run-parts: /etc/cron.daily/logrotate exited
with return code 1
269[04:05:09] <jmcnaught> hiptobecubic: if you have
allow-hotplug interfaces in /etc/network/interfaces then when they
become available it triggers a udev rule that starts the
ifup@.service template for that interface.
270[04:08:26] <hiptobecubic> that is totally believable...
i'll check thanks
294[05:03:03] <charking_> Hello, how do you troubleshoot a
graphics problem? Every time I boot I get white garbled screen that
cuts on and off for several minutes until I get an X login and then
everything works fine. dmesg has messages about GPU lockup and soft
resets. I'm using correct drivers (xserver-xorg-video-radeon).
I tried with proprietary firmware (firmware-amd-graphics).
295[05:03:13] *** Quits: catman370 (~catman@replaced-ip) (Quit: See you later..)
296[05:04:12] <charking_> It's weird that it just works
fine after a few minutes of not working and cutting on and off.
314[05:26:59] <jim> hi, anyone aware of a bug in okular, where
it works fine until you print something, then the user interface
freezes, and it can't quit without being killed?
335[05:48:16] <Henry151> howdy folks. debian 10, i just ran apt
upgrade and it hung while upgrading grub-pc. I killed the dpkg
process, but now it's saying i need to run dpkg --configure -a,
and when i do so, it hangs on "installing grup-pc for i386-pc
platform"
336[05:48:49] <Henry151> any guidance? much appreciated in
advance
337[05:48:57] <rue_bed> try apt-get remove grub-pc first ?
338[05:49:12] <rue_bed> just dont reboot it in between
339[05:49:31] <rue_bed> even if you do tho, I think its still
all in the boot sector
348[05:52:52] <rue_bed> grub was for all the cryers who said
they wanted to install windows and linux on the same computer, and
they never booted linux anyhow, so it was totally moot
349[05:52:54] <Henry151> one of my earliest memories of a
painful linux-related experience involved screwing up grub and
losing access to my system entirely... every time i see the word
grub i get a little ptsd flashback
365[06:05:18] <graytron> where does debian buster get its
resolution for the system console? i have an Aspeed AST2500 graphics
adapter with vga output and I can't get the text on system
console to fit my monitor. the text is drawn outside of the visible
area
377[06:18:36] <Urk> jcmnaught> I would like to explore
deleting a kernel for 4.19, and wondering if this is a good idea? I
recall you mentioning during the install that I needed a kernel
greater than 5.0 in order to install Debian Linux on my laptop.
378[06:19:06] <Urk> mcmnaught: I discovered the other kernel
when I booted into recovery mode.
381[06:25:18] <Urk> jmcnaught> jcmnaught> I would like to
explore deleting a kernel for 4.19, and wondering if this is a good
idea? I recall you mentioning during the install that I needed a
kernel greater than 5.0 in order to install Debian Linux on my
laptop. Sorry for being redundant. Just realized I typed your name
wrong.
382[06:26:15] <jmcnaught> Urk: I would just leave it, it is not
hurting anything.
383[06:26:56] <Urk> Am I correct that kernel 4.19 can't be
used for this installation because my laptop is too new?
384[06:27:13] <jmcnaught> Urk: one thing I would do though since
you installed the kernel from backports manually is "apt -t
buster-backports install linux-image-amd64" to make sure
it's tracking the kernel from backports for updates.
385[06:27:47] <jmcnaught> Urk: from what I recall only the wifi
card did not work under kernel 4.19.
386[06:28:53] <Urk> So you recommend installing
linux-image-amd64?
387[06:29:18] <jmcnaught> Urk: I recommend installing the
linux-image-amd64 metapackage from buster-backports with that
command.
388[06:29:21] <Urk> I misread, youi want "apt-t
buster-backports install linux-image-amd64"
390[06:30:26] <Urk> ok. Earlier today I had to fix a problem
with xfce. I kept getting an error message indicating I am in kiosk
mode. I looked up a solution online, and it involved deleting cache
files from two folders. XFCE seems to be working now.
391[06:32:53] <Urk> jmcnaught> I get a message indicating --
E: The value 'buster-backports' is invalid for
APT::Default-Release as such a release is not available in sources
392[06:33:11] <Urk> Doesn't look like it is tracking it.
394[06:33:45] <Urk> my mistake. I mistyped the command by one
letter.
395[06:34:42] <Urk> This is the result after I corrected the
command
replaced-url
396[06:35:16] <jmcnaught> That's good
397[06:35:29] <Urk> I have some missing firmware messages at
bootup. Am I correct that dmesg should be able to see the messages
in a file?
398[06:36:37] <Urk> Is that /var/log?
399[06:37:16] <jmcnaught> Urk: there is a dmesg command you can
run as root. You can also look at /var/log/syslog. Or you can run
the journalctl command.
400[06:38:25] <jmcnaught> Urk: "journalctl -b | grep -i
firmware" should show you all log entries containing the word
'firmware' from the current boot
401[06:40:57] <Urk> Where are the journal messages stored at? I
don't see anything in /var/log, but maybe I didn't check
far enough.
402[06:42:04] <jmcnaught> Urk: journal files are stored under
/var/log/journal/ but they're not text files so you read them
with the journalctl command.
403[06:42:51] <Urk> I am in /var/log and don't see anything
named journal. Ok, I will try the command you mention.
404[06:43:39] <jmcnaught> Urk: oh yeah, on buster you need to
create /var/log/journal if you want a persistent journal, otherwise
the journal is kept in memory. See
/usr/share/doc/systemd/README.Debian
405[06:44:32] <Urk> Any harm with creating a persistent journal?
406[06:44:48] <jmcnaught> No, I like a persistent journal.
408[06:44:59] <Urk> so do I. I like to know what is going on.
409[06:47:40] <MegaCarp> debian 10 xfce (still have gnome
installed) with lightDM and arctic greeter (still have gdm
installed) - can't seem to get user-specific locale\language
selection to get to work - only system-wide by using
dpkg-reconfigure locales. i've appended to ~/.bashrc
"export LANG=en_US.utf8" but it didn't help.
thoughts?
410[06:49:20] <Urk> Something is wrong with my file system. I
open up my home folder and there is nothing there -- no desktop, no
downloads, no documents, no nothing.
411[06:53:49] <Urk> Well, its getting late. I will continue this
tomorrow.
542[09:39:41] <dpkg> In systemd, "systemctl set-default
multi-user.target", or remove the DM package(s) with
"aptitude remove gdm3 kdm lightdm lxdm nodm sddm slim wdm
xdm". "echo false
>/etc/X11/default-display-manager" will also disable the DM,
or just hit ctrl-alt-fN to get to a console. nodm is the name of a
minimal/automatic display manager (replaced-url
629[10:53:53] <Waxhead> A quick question about our friend
/proc/loadavg , I am on kernel 5.10 (debian testing so apologize if
this is slightly off topic) and if I run stress -c 1024 -t 2 on a 8
cpu system I sometimes get a big jolt in the loadavg (1min) and
sometimes not. I would not expect this to be as high with only a
quick burst of 1024 processes when the rest of the minute the
machine is idle. I know it is called load average, but should it not
drop about
630[10:53:53] <Waxhead> as quick as it rise with such a minor
burst?
632[10:55:01] <ratrace> pasiz: in real world numbers are very
very very rarely pure powers of 2, so any algorithm relying on them
will be very disappointed
633[10:55:24] <pasiz> ratrace: in binary world, everytime...
642[10:57:30] <ratrace> the core of cryptography is prime
numbers so pray tell... explain to me how powers of two fit that
paradigm.
643[10:57:58] <ratrace> yes, you bit shift and you divide or
multiply BY 2. what about BY 2.34, or BY 3, or BY 7, or BY 5, ......
644[10:58:02] *** debhelper sets mode: +l 1046
645[10:58:16] <ratrace> so you see, relying on power of 2 in
operations is very, very rare. and as Waxhead said, you _might_
approximate, but not always.
673[11:07:22] <pasiz> ratrace: so you could find many references
to kernel_fpu_begin() in crypto side?
674[11:07:31] <Waxhead> but joling aside , I was expecting
loadavg to drop about as quickly as it rise since the number of
processes run is ONLY a quick burst. Is it really that
"bad" ?
681[11:10:46] <ratrace> that's wrong. the load average does
not change with number of cores. it represents the number of
"tasks" the kernel is scheduling at any moment, and the
loadavg represents the moving average of that
683[11:11:39] <ratrace> the only difference that number of cores
make is length of execution of a particular job. scheduled on single
core they'll serialize and take longer than scheduled on 8
cores where they'd last, assuming linearity of scheduling,
1/8th of that time
684[11:12:05] *** Quits: szorfein (~daggoth@replaced-ip) (Remote host closed the connection)
688[11:13:18] <ml9l> hey i am trying to replace nouveau driver
with nvidia through apt install nvidia-drive, it is installed but
after restart i get blackscreen, how do you i go about fixing this
problem? any hints?
690[11:13:41] <ratrace> pasiz: yes. unpack the kernel.org
tarbal, grep for kernel_fpu_begin and you'll see where they
are. avx, avx2, mmx, sse, sha*, poly*, morus*, crc, * .....
691[11:13:44] <Waxhead> ml9l: sarch for : blacklist nouveau
696[11:15:14] <ratrace> ml9l: remove the "quiet" from
grub's kernel command line and see where it hangs, whether it
spits out errors on the console.
697[11:15:33] <ml9l> i restarted and typed startxfce4 (that is
another story but this is the only way it is working right now) and
i get: (EE) no screens found
698[11:15:44] <ratrace> ml9l: alternatively you'd need to
boot into text mode and do some analysis of the kernel ringbuffer,
Xorg.0.log, modprobing and friends
699[11:15:50] <ml9l> to my knowledge that is a problem with the
Xorg config file, how should i generate it for nvidia?
700[11:16:03] <Waxhead> ml9l: have you tried to mv your xorg
conf away ?
701[11:16:09] <ratrace> ml9l: in vast majority of cases you do
not need any xorg.conf
702[11:16:16] <Waxhead> ml9l: you usually do not need a
xorg.conf anymore
703[11:16:24] <ratrace> (in fact it can only be detrimental to
normal operation)
704[11:16:27] <ml9l> so why is there "no screens
found"?
705[11:16:32] *** Quits: bamdad (~bamdad@replaced-ip) (Remote host closed the connection)
706[11:16:39] <ratrace> but if you really need one, nvidia has
xorg-config something something tool
707[11:16:58] <ratrace> nvidia-xconfig methinks
708[11:17:02] <ml9l> oh i checked the log: failed to open drm
device for pci
718[11:19:29] <ratrace> so debug step by step. check dmesg, has
the module loaded properly, any errors mentioned there. then
Xorg.0.log. and check permissions on relevant /dev devices
728[11:26:10] <ml9l> current dmesg output has nothing about
nvidia, nothing. So i went a head and created a new Xorg file just
to check its side-effects with "X -configure", ran
startxfce4 and here is the result of the X log file:
replaced-url
729[11:26:44] <ml9l> there is an explicit error saying
"failed to open drm device for pci"
730[11:27:35] *** Joins: conta (Thunderbir@replaced-ip)
735[11:29:48] <ratrace> ml9l: that looks like nouveau still
loading. which package have you actually installed for nvidia? is it
actually installed? it's a DKMS package, did it actually run?
check find /lib/modules/ -name '*nvidia*'
736[11:30:13] <ratrace> if the module exists, and you have
nvidia gpu, it _must_ have a mention in dmesg, even if about failed
load. did you grep case insensitive?
737[11:30:20] <ml9l> tesla 450 drivers were suggested, lemme
check
738[11:30:33] <ratrace> "suggested" by what or whom?
739[11:30:59] <n4dir> is there a way to make gvfsd not start, or
just stop it, if uninstalling it is not an option?
740[11:31:24] <ratrace> n4dir: I think gnome very much hard
depends on it
741[11:31:25] <n4dir> applet.py is a similar problem
742[11:31:26] <ml9l> nvidia-detect
743[11:31:38] <n4dir> ratrace: yeah, uninstalling it is not an
option, due to dependencies
744[11:31:38] <ratrace> ml9l: which GPU model it actually is?
771[11:43:25] <ratrace> well duh. we said several times you
should remove xorg.conf. a) it's not needed, b) it's
likely causing issues
772[11:44:07] <ratrace> the only reason to have xorg.conf is to
override some settings the driver can't autoconfigure, which is
rare, and it's then better done via xorg.conf.d snippets
783[11:52:06] <ml9l> i removed the xorg file and now i am able
to login through the console with startx, i have no GUI but there
are with very low resolution, i guess the nvidia driver is still not
working, current Xorg log: termbin.com/1dz6
813[12:06:30] <ratrace> ml9l: so that's bullseye? you
should query for help at
814[12:06:33] <ratrace> !debian-next
815[12:06:33] <dpkg> #debian-next is the channel for
testing/unstable support on the OFTC network (irc.oftc.net), *not*
on freenode. If you get "Cannot join #debian-next (Channel is
invite only)." it means you did not read it's on
irc.oftc.net. See also
replaced-url
816[12:06:45] <ratrace> but anyway, try turning secure boot off
and see how that goes
882[12:50:44] <tosted> how to controling default sound device
output to HDMI ? every reboot changing to headphones? every
disconnection of headphone do not go back to HDMI.
912[13:07:23] <EdePopede> the opposite problem here sometimes.
never had anything digital, seems there's a digital output on
the sound, that's all. right now it's fine, Line Out
(plugged in) and Headphones (unplugged). sometimes i unplug the
sound system, then it switches to digital. and doesn't return
to Line Out. so i have to play the old plug/unplug game, sometimes a
few times.
913[13:08:13] <EdePopede> can't say now if it just
doesn't recognize the hardware or some software trying to be
"smart".
915[13:08:19] <shtrb> we may try via pulseaudio config
(/etc/pulseaudio/default.pa) by setting "set-default-sink
XXXXX" where XXXXX is the name for your HDMI connection
916[13:09:50] <tosted> i ll try.
917[13:10:10] <EdePopede> shtrb, it's in pulse/ here, did
it change for stable?
937[13:23:27] <tosted> "set-default-sink
"outputXXXXX" where outputXXXXX is the name( to list the
sinks: $ pactl list short sinks) " for your HDMI connection
938[13:23:59] <shtrb> tosted, did it solve your problem ?
959[13:30:36] <EdePopede> connecting everything needed on an
analogue synthesizer is far easier than understanding linux sound
960[13:30:58] <tosted> EdePopede: $ pactl list sinks
961[13:31:07] <shtrb> Anyone with joycon peered with debian say
if the IR interface is exposed somehow and can share where I should
be looking for it ?
976[13:34:04] <dpkg> Vi IMproved (vim) is an enhanced <vi>
editor. Extremely popular clone with syntax highlighting and
graphical X interface available. The vim-tiny package no longer
installs vim <alternatives> as of version 2:7.2.049-1 (Debian
bug #529977). See also <vim refcard>, <cream>,
<colored pager>, <vim syntax highlighting>.
replaced-url
977[13:34:05] <EdePopede> well i have no sound *card* at all
anyway
978[13:34:20] <shtrb> EdePopede, pa don't see it or you
realy don't have it ?
979[13:34:30] <EdePopede> the only *card* i have is the one for
the images. connected to my monitor.
980[13:34:43] <shtrb> a chip on your motherboard can act as a
"card"
981[13:34:48] <EdePopede> all the cables come out from some
connector on the board
982[13:35:07] <EdePopede> well then they should correctly say
sound*chip* and not sound*card*
986[13:36:26] <EdePopede> *shrug* as i said i don't
understand this mambo jumbo and never did. some day i bought a PC
with "AC97" printed on the carton next to a lot of other
cryptic names and i think every machine i had since has had one
988[13:36:52] <EdePopede> unluckily days are over when they
delivered complete plans like with radios and TVs in the past
989[13:37:04] <EdePopede> so all a customer gets is some
advertising bs
990[13:37:50] <EdePopede> i know 8080 was a cpu made by intel, i
know 68k was made by motorola. AC97? no idea. and where it came from
and what it can do. something sound. that's all.
992[13:38:28] <EdePopede> i couldn't even show you where it
is on my PC's board
993[13:38:46] <EdePopede> or how it is connected to whatever it
is connected to
994[13:39:22] <ratrace> ml9l: "tainted" is not a
problem, every out of tree kernel module will "taint" the
kernel. the problem was signature validation ... so if you booted in
legacy mode... you have nvidia working fully now?
996[13:39:57] <EdePopede> and that pulse must be somewhere
between player and sound chip. alsa? no idea, i thought i had
replaced it years ago with pulse when microsoft decided that
it's not worth supporting two different sound systems.
997[13:40:13] <ml9l> ratrace: it is still trying to load nouveau
how to avoid that?
998[13:40:23] *** Quits: dez (uid92154@replaced-ip) (Quit: Connection closed for inactivity)
999[13:40:38] <EdePopede> so basically player → pulse
→ sound card is all i get so far. still may be wrong.
1000[13:40:58] <ratrace> ml9l: nvidia-driver package
should've installed a file in /etc/modprobe.d/ , I forgot its
name, that basically does blacklist nouveau . chack for that please
1009[13:44:51] <EdePopede> tosted: huh? right now everything is
fine, it isn't predictable when it decides to break. but since
i had similar issues with my PS2 mouse on my 1st PC i don't
really trust consumer hardware quality.
1010[13:44:53] <ml9l> oh damn sorry
1011[13:44:57] <ratrace> ml9l: but I think that last Xorg.0.log
you pasted showed that VESA was being used, which would explain low
resolution. in that nouveau is blacklisted and nvidia-driver
ain't loading due to bad signature
1013[13:46:04] <ml9l> the bad signature shouldnt be relevent
because i am in legacy boot mode
1014[13:46:53] <ratrace> ml9l: I think the kernel (in bullseye)
enforces signed modules now, regardless of secure boot; the two are
related but EFI's not required for signed modules ... AFAIK!
1015[13:47:00] <nifker> why cant ls show me the creation time of
a file?
1016[13:47:36] <ratrace> nifker: I think that's added to
coreutils in bullseye
1019[13:50:32] <ml9l> ratrace: does the x server load the driver_
1020[13:50:37] <ml9l> or it is loaded before?
1021[13:51:45] *** Quits: noosanon (uid489102@replaced-ip) (Quit: Connection closed for inactivity)
1022[13:51:51] <ratrace> ml9l: xserver loads _xorg_ specific
"drivers" which interact with "kernel module"
drivers which are loaded by the kernel/udev on boot when the
hardware is probed and gpu detected
1023[13:52:11] <ml9l> ok
1024[13:52:13] <ratrace> so the traditional "drivers"
here, hardware drivers, are the kernel modules loaded before xorg is
even touched
1028[13:52:51] <ml9l> i looked at journalctl -xe: "failed to
query NVIDIA devices. Please ensure that the nvidia device files
exist and the user has read write permissions blablabla
1029[13:53:22] <ratrace> yes, there's no /dev devices
because the module is not loaded lsmod | grep nvidia shows nothing,
amirite?
1030[13:53:27] <tosted> EdePopede: good!
1031[13:53:36] <ml9l> and most recent Xlog file: "matched nv
as autocinfigured driver 1"
1093[14:13:38] <ml9l> i mean up to signing modules myself i dont
bother
1094[14:14:03] <ml9l> fixing things, but online solutions says i
have to pass self-generated keys to the nvidia installer and even
the nvidia installer without apt doesnt work
1095[14:14:30] <ml9l> the question is if modules installed
through apt can be signed with mokutil after the installation
1096[14:14:36] <ratrace> is that now a new requirement on
bullseye for out of tree modules?
1097[14:15:05] <ml9l> Some folks reported it might be the issue
behind my nvidia module problem
1098[14:15:25] <ml9l> I am not sure if this is the reason why the
module doesnt work, so i gave up :*
1099[14:15:27] <ml9l> :(
1100[14:15:30] *** Quits: xet7 (~xet7@replaced-ip) (Remote host closed the connection)
1104[14:16:31] <ratrace> ml9l: point being, bullseye is in soft
freeze. if a package ain't working normally by mere act of apt
installing it, it's a major bug
1105[14:16:35] <ml9l> i went to ask in #debian-next and got
pretty unfriendly answers
1107[14:17:26] <ml9l> ratrace: i got you, but my question still
stands: have things failed in my case *because* of an unsigned
module?
1108[14:18:15] <ratrace> ml9l: I would say yes, based on your
paste that shows that (signature failure) as last entry related to
"nvidia" module, and it not loading at all
1109[14:18:28] <ratrace> incompatible ABI or other issues would
have specific, different error messages
1110[14:19:33] <ratrace> ml9l: you never confirmed if
there's a proper dkms kernel module in /lib/modules/ , or by
using modinfo as suggested by someone (but I think that works only
for loaded modules? not sure)
1117[14:22:05] <ml9l> all i can say there were no nvidia loaded
modules and /lib/modules hadnt any either but if the "kms"
arguemtn would have made difference?
1118[14:22:47] <ratrace> ml9l: you were looking for
"*nvidia*" names. maybe the .ko are differently named, I
dunno. hence that find -type d -name dkms would list all paths
inside dkms dir, which is where nvidia would reside
1119[14:23:13] <ratrace> I can't check myself either because
I purged nvidia modules and don't have past kernel versions in
/lib/modules/ I clean them out
1147[14:38:22] <ml9l> ratrace: nouveau is working and now i see
lightdm asking for user/pwd i enter the right credintials but it
resets itself. otherwise if i go to a tty and enter startx it works
!
1234[15:32:09] <ratrace> last time I did OCR 'twas many
years ago. Some professional software on Windows. Still had to go
through text and do manual corrections. With paid software. in
commercial/professional environment.
1235[15:32:45] <queip> ratrace: yeah but I specially generate
this text to be ideal for OCR. if OCR's options to select
character set would work as documented
1249[15:45:21] <Kobaz> having an issue.. getting an email every
night from cron: /etc/cron.daily/apt-compat: Exception:
'NoneType' object has no attribute
'dependencies' run-parts: /etc/cron.daily/logrotate exited
with return code 1
1250[15:48:19] <ml9l> ratrace: have a good day and thank you for
the help - really appreciated
1271[16:14:53] *** Joins: conta (Thunderbir@replaced-ip)
1272[16:15:40] <BenW> Hi! I have a weird issue where CPU
frequencies drop under load – temperatures are low, governor
is "performance". I'm sure it's just a user
error, so I don't think a bugreport is the right thing to do
just yet. Can you tell me what to try next?
1273[16:16:00] <BenW> (I have lots of details, but I don't
want to flood the channel.)
1274[16:17:17] *** Quits: JohnML (~john1@replaced-ip) (Remote host closed the connection)
1282[16:22:33] <jhutchins> BenW: Does it affect performance?
1283[16:23:40] <avu> BenW: with most modern CPUs, it's
normal that the maximum frequency can only be reached if not all
cores have work currently, so the frequency goes down as work shifts
from one/few to many/all cores. Sure that's not what
you're seeing?
1287[16:24:32] <BenW> I don't have a direct comparison,
because being unable to try it with a steady 4.3 GHz is the very
problem in the first place.
1288[16:24:54] <avu> BenW: what's the exact model?
1289[16:25:14] *** Quits: Night-Shade (~TimF@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
1290[16:25:35] <BenW> avu: Intel Core i5 10400 6x 2.90GHz, which
claims to run up to 4.3 GHz in "boost mode", which I do
see when only 1 or two threads are active.
1291[16:25:54] <BenW> In any case, it shouldn't drop *that*
far below 2.9 GHz, even if I misunderstood that completely.
1323[16:32:02] <BenW> jhutchins: Would that also affect the
actual CPU frequency? "linux-cpupower frequency-info"
shows low frequencies, as well as the cpu-frequency xfce4-panel
plugin.
1324[16:33:03] <avu> BenW: yes, if there is no more work to do
for your CPU because it's waiting on IO, it might lower its
frequency
1336[16:45:19] *** Quits: bogus (~bogus@replaced-ip) (Quit: Es ist mir Wurst!)
1337[16:47:00] <avu> BenW: there might be something you can do
about it now that you have a better idea where the bottleneck really
is, jhutchins suggested some tools for you to find out
1341[16:51:12] <BenW> I'm not sure how "top" can
help me, and atop … hmm, that's a lot of numbers.
1342[16:52:11] <BenW> As expected, it shows 92-98%
"user" for each of the 12 cores. How can I distinguish
between "actually computing" time and "waiting on
memory bus" time?
1363[17:18:35] <shtrb> Anyone with joycon peered with debian can
say if the IR interface is exposed somehow and can share where I
should be looking for it ? I tried looking under input but I
couldn't find anythning new
1565[18:42:05] <shtrb> rue_bed, I just tried xhost + , sudo -u
newuser kate and hate kate opened on my screen (the user which run X
for me is not newuser)
1694[20:18:13] <koollman> alex11: packages aren't kept, they
are extracted and installed. They are temporarily stored in
/var/cache/apt/archives. Unless you mean where they are installed,
then yes, /usr, /bin, /lib, and others, as defined in the package
1695[20:19:02] *** Quits: Vizva (~Vizva@replaced-ip) (Remote host closed the connection)
1711[20:32:51] <mi11k1> Does anybody know a better way of
managing your Google account other than the web interface? Like is
there a way to do multiple edits Etc
1712[20:33:11] <mi11k1> or do i need to spend the next 48 hrs on
this?
1721[20:37:36] <SponiX> mi11k1: there is some Gnome desktop stuff
that allows you to login to cloud type accounts ( including google
). BUT, I have never used it, so not sure exactly what all it is
capable of
1722[20:37:36] *** Joins: kini (~kini@replaced-ip)
1723[20:38:20] *** Quits: igrtrrt (~igrtrrt@replaced-ip) (Remote host closed the connection)
1773[21:46:35] <Gramcor> does /var/mail autoclean periodically?
like logrotate?
1774[21:46:39] <Gramcor> or i have to set it up
1775[21:49:47] <Kobaz> it shouldn't
1776[21:50:03] <Kobaz> mail is one of those things there's
no possible way to set a default cleanup that everyone would like
1777[21:50:38] <cybercrypto> Gramcor: what are you storing in
var/mail? standard sendmail 'user mails'? If yes, it will
not rotate automatically.
1778[21:51:21] <Gramcor> cybercrypto, i have setup postfix to
send mails to root user and also forward to gmail etc
1779[21:51:39] <Gramcor> would like a mechanism to compress
rotate and delete mails in /var/mail/root
1780[21:51:42] <cybercrypto> Gramcor: you can setup filesystem
quota to avoid 'filling up'. There are many ways to
achieve that. You should define exactly what are you trying to
achieve.
1797[21:55:35] <cybercrypto> Gramcor: I use former pine and mailx
to read mails.
1798[21:55:47] <Gramcor> i have installed postfix and mailutils
1799[21:56:10] <Gramcor> by the way how can i read root mails
with mailutils? i did "less /var/mail/root"
1800[21:56:10] <cybercrypto> Gramcor: if you want 'rotating
contents of the mail file' older than 90 days... than you need
to parse it.
1801[21:56:41] <cybercrypto> Gramcor: if you copy your mail file
everyday, into another place... you will know that is '1 day of
your mail'.
1802[21:57:36] <cybercrypto> Gramcor: than you can remove older
than 90. it is the easiest... otherwise you will need to parse it...
or try the solutions you are investigating
1803[21:58:18] <cybercrypto> Gramcor: type mailx (it is BSD
originals)
1804[21:58:35] <Gramcor> i have gnu-mailutils
1805[21:58:41] <Gramcor> probably the mail command?
1811[22:00:10] <cybercrypto> Gramcor: have you tried mailx
command? install and test. once you read using mailx (or any other
cli tool) it will automatically remove the mail from /var/mail and
place into your ENV (like home)
1812[22:00:21] <cybercrypto> Gramcor: yes
1813[22:00:28] *** Quits: kini (~kini@replaced-ip) (Remote host closed the connection)
1818[22:01:49] *** Joins: kini (~kini@replaced-ip)
1819[22:01:52] <Urk> Any idea on how to filter errors with the
word "firmware" at bootup from a persistent journal log
that I created? I ran jounralctl -b -1, and got an error Data from
the specified boot (-1) is not available: No such boot ID in journal
1820[22:02:58] *** Quits: filohuhum (~filohuhum@replaced-ip) (Remote host closed the connection)
1821[22:03:02] <Urk> I then ran just journalctl, but get a very
long list, and wondering if I can use the grep command to filter
errors with to a file? I'm looking for startup errors that say
firmware. I'm on Debian Buster, and created a persistent
journal log.
1825[22:06:38] <Urk> cybercrypto> Yesterday I heard from JMC
that Buster stores logs in memory so I elected to create a
persistent log feeling it would be easier to find what I am looking
for. I was told the /var/log is where the persistent logs would be,
and that I could find startup errors in that.
1827[22:07:54] <cybercrypto> Urk: what debian version are you
using?
1828[22:08:16] <Urk> What is the difference between journalctl -b
-1 and journalctl -b 1
1829[22:08:53] <cybercrypto> Urk: to persist journald logs, i
recommend using the recommended docs: set to Storage:persistent in
/etc/systemd/journald.conf
1830[22:09:09] <Urk> cybercrypto> Debian Stable 10.8.9
(Buster). However, Buster isn't officially ready for the Dell
XPS 15 7590 so I got a bit of help installing firmware to get it up
and running. Getting rid of the firware errors at bootup is two
remaining issues that have to be fixed.
1831[22:09:17] <cybercrypto> Urk: only after that, journald will
persist logs across boots.
1832[22:09:41] <cybercrypto> Urk: I am afraid you can only see -b
0 (which is current boot, not past ones)
1833[22:11:02] <cybercrypto> Urk: double check that config
(journald.conf) Check journald.conf manual and you will find lots of
features and parameters that may help you
1874[22:40:43] <oxek> isn't nano installed by default on
debian?
1875[22:40:48] <mi11k1> yepp
1876[22:40:52] <mi11k1> same as vi i think
1877[22:41:11] <oxek> nano is probably easier to figure out for
newcomers than vi(m)
1878[22:41:25] <mi11k1> yepp, thats why i dont klnowhow to use vi
1879[22:41:28] <EdePopede> though shalt have no other editor but
ed for it is the only real one.
1880[22:41:37] <mi11k1> i wish i would have no formed the habit
1881[22:41:48] <mi11k1> vi seems to be more efficient
1882[22:41:57] <mi11k1> i cant say why, i just know it
1883[22:42:01] <jhutchins> The nice thing about knowing vi is
that you will nearly never find a system that doesn't have it,
including systems that are partly broken.
1884[22:42:06] <oxek> whenever I open up nano for some reason, I
end up with a whole lot of jjjjjjjjkkkkkjh
1887[22:42:44] <mi11k1> oxek, my mouse caused that
1888[22:42:47] <mi11k1> or ir
1889[22:42:54] <mi11k1> or broken keboard
1890[22:43:41] <mi11k1> oxek, what termoinal? i use terminator
1891[22:44:03] <cybercrypto> EdePopede: I agree. the one and only
ed. period.
1892[22:44:05] <mi11k1> actually does anybody know why sometimes
the text disappears in this term?
1893[22:44:29] <mi11k1> usually after ss session into somewhere
1894[22:44:30] <oxek> it's a bit of a joke, I'm so used
to navigating in vim, where you press 'j' and
'k' to move line down & up, that when nano is opened I
instinctively press those keys but they are instead written into the
file
1895[22:44:41] <mi11k1> lo;l
1896[22:44:46] <mi11k1> omg
1897[22:45:14] <mi11k1> kinda like how i do ctrl alt dlt in vm
and reboot thre host
1898[22:45:38] <oxek> a basic knowledge on how to operate vim is
probably a good test of something
1912[22:49:31] <mi11k1> when i have systemd issues its usually
cause im not booted systemd
1913[22:49:35] <mi11k1> golly
1914[22:49:37] *** Quits: fflori (~fflori@replaced-ip) (Remote host closed the connection)
1915[22:49:44] <Urk> millk1: Trying to get rid of some error
messages at bootup. There are other problems that have to be fixed
as well, but I'm trying to do one at a time. Other problems
include folder series missing from the user: no home, no downloads,
no pictures, no documents, etc, etc.
1916[22:50:30] <oxek> that doesn't sound like systemd
1917[22:50:31] <mi11k1> Urk, haha, live with it. some errors are
more warnings
1918[22:50:49] <mi11k1> if it hangs on boot, i then worry
1919[22:51:16] <mi11k1> usuallt google will get u straighjt to
the answer with a copy paste
1920[22:51:45] <Urk> millk1> There are other performance
issues such as XFCE displaying over below the bottom panel, etc. So
I don't want to second guess the error is something that has to
be addressed.
1923[22:51:57] <judd> Package xdg-user-dirs (utils, optional) in
buster/amd64: tool to manage well known user directories. Version:
0.17-2; Size: 52.5k; Installed: 375k; Homepage:
replaced-url
1924[22:52:03] <oxek> that's one package you might need
1926[22:52:39] <mi11k1> oxek, i thought it was headless
1927[22:52:46] <mi11k1> what desktop u use?
1928[22:52:54] <oxek> xfce
1929[22:53:02] <oxek> but xdg-user-dirs work on headless systems
too
1930[22:53:07] <mi11k1> wanna know a trick?
1931[22:53:09] <oxek> it just sets up the known dirs
1932[22:53:13] <mi11k1> download MX
1933[22:53:39] <oxek> MX?
1934[22:53:45] <mi11k1> its a modified debian
1935[22:53:57] <mi11k1> i use it on anything that has a monitor
1936[22:54:20] <oxek> !mxlinux
1937[22:54:21] <dpkg> MX Linux is a popular distribution
<based on debian>. It is not supported in #debian. Support is
available on their forum:
replaced-url
1938[22:54:27] <oxek> I wanna stay in this channel though :)
1973[22:58:56] <oxek> not sure why you're singling me out
1974[22:59:07] <nickgaw> Hi, Is it possible to have a setup where
you can run Debian both natively on a dual boot system and from with
in a virtual machine so you can just switch between setups?
1982[23:00:37] <nickgaw> yes but I want to be able to run the
same system from both natively and in a virtual machine on windows
not reinstalling it into a virtual machine is that possible?
1983[23:01:02] <mi11k1> nickgaw, make an image of the debian
system?
1984[23:01:04] <oxek> nickgaw: that's the thing, you
don't reinstall it, you just create a new virtual machine, and
point the storage to the actual partition
1985[23:01:12] <mi11k1> like make a backup and restore it to a vm
1986[23:01:28] <mi11k1> or sumthin like that
1987[23:01:52] <nickgaw> Is doing this safe where I just use the
existing partition?
1988[23:02:00] <oxek> nickgaw: I can either boot debian directly
on raw hardware, or boot the other OS on that disk and then run that
same debian installation in qemu/virtualbox/vmware
2000[23:03:48] <oxek> mi11k1: I get the feeling you're
trolling a bit, aren't you
2001[23:03:51] <HelloShitty> Hello. I have qemu buster version
which is 3.1 but I would like to try the version from
debian-backports. Can I just install it or is it better to remove
the oler version first? I'm on Debia Buster, btw
2002[23:03:54] <Rer> hallo need help
2003[23:04:16] <oxek> HelloShitty: you can just install it
2005[23:04:26] <jmcnaught> HelloShitty: when you install the
newer package from buster-backports it will upgrade the existing
package, you don't need to remove anything.
2006[23:04:41] <Urk> jmcnaught> Get a message indicating that
firmware-misc-nonfree is already the newest version.
2007[23:04:42] <nickgaw> So if I pointed my Debian installation
using a virtualization system in windows 10 pro 64 bits on this dual
boot system I could then run Debian natively or from with in windows
in a virtual machine could I install the guest tools so when running
virtually do file sharing between the guest and the host?
2008[23:04:52] <mi11k1> oxek, no, still recovering from drone in
house issues.
2009[23:05:00] <Urk> I will check the backports, but it looks
like it isn't in there -- even though it is a required driver.
2010[23:05:08] <jmcnaught> Urk: what does "apt policy
firmware-misc-nonfree" say?
2030[23:08:32] <mi11k1> it will install backports version
2031[23:08:36] <jmcnaught> HelloShitty: yes if you "apt -t
buster-backports install foobar" and foobar is already
installed it gets upgraded to the buster-backports version.
2032[23:08:45] <HelloShitty> ok, thank you
2033[23:09:20] <fireba11> hi, my first partition still started at
sector 63. that caused issues with the latest grub update. i deleted
the raid partition, recreated and and now grub2 is fine. however the
new partition is listed as md127 in mdadm, and renaming (and
updateing initramfs) doesn't help. any ideas? for now i edited
my fstab but ...
2034[23:09:35] <Urk> jmcnaught: My sources.list appears to have
backports already
replaced-url
2036[23:10:36] <jmcnaught> Urk: you are missing contrib and
non-free for the buster-backports lines 16-17
2037[23:11:03] <oxek> Urk: it does have backports, yes. I'd
like to point out that you're using 'stable' keyword
isntead of 'buster', is that intentional?
2039[23:12:09] <jmcnaught> Urk: oxek is right, using
'stable' instead of the release codeword
'buster' could lead to a surprise upgrade when bullseye
gets released as stable. Recommended to use 'buster' in
sources.list, not 'stable'.
2040[23:12:50] <oxek> plus the combination of 'stable'
and 'buster-backports' strikes me as possibly problematic
too, when 'stable' starts pointing to bullseye
2043[23:13:24] <jmcnaught> Urk: your /etc/apt/sources.list file
should look like the second example, with the additional two lines
for backports:
replaced-url
2049[23:14:40] <jmcnaught> Rer: to get any help you will probably
need to show more information, such as the complete output of
"systemctl status dovecot" and "journalctl -b -u
dovecot" (in a paste on
replaced-url
2050[23:14:40] <Rer> [cant start dovecot
2051[23:14:44] *** Quits: Freneticks (~Frenetick@replaced-ip) (Remote host closed the connection)
2059[23:16:30] <jmcnaught> Rer: this error is truncated but you
should look into it: doveconf: Fatal: Error in configuration file
/etc/dovecot/conf.d/90-plesk-sieve.conf line 30: Unknown setting: ma
2060[23:16:49] <Rer> what must i do ?
2061[23:17:10] <jmcnaught> Rer: correct the error in that file,
which does not look to be part of a standard Debian installation.
2062[23:18:53] <Rer> roundcube imap ereror
2063[23:19:11] <Urk> ok I am going to reboot.
2064[23:19:41] *** Quits: Urk (~pdq@replaced-ip) (Remote host closed the connection)