1[00:05:02] <quadrathoch2> lenswipe if I remember correctly 9
-> 10 was the switch to systemd-resolved, which had dns leak
issues. but I am not sure if they were ever fixed
121[01:35:51] <dpkg> If you have a question, just ask! For
example: "I have a problem with ___; I'm running Debian
version ___. When I try to do ___ I get the following output ___. I
expected it to do ___." Don't ask if you can ask, if
anyone uses it, or pick one person to ask. We're all
volunteers; make it easy for us to help you. If you don't get
an answer try a few hours later or on debian-user@lists.debian.org.
See <smart questions><errors>.
122[01:37:07] <b1ack0p> what if i am just greeting?
166[02:16:19] <cuddylier> I have IPs bound to an interface
called 'dummy1' but I'm unable to actually get any
application to use them as their source IP, all applications
externally show their source IP as the IP on the main interface
(e.g. eth0). Any ideas why this might be?
168[02:17:01] <cuddylier> Inbound traffic to them is fine but
any outbound requests e.g. if I use 'wget --bind-address=[IP]
ipinfo.io/ip', it returns the primary interface IP rather than
the IP I specified.
171[02:18:08] <cuddylier> The primary IP being 45.xx.xx.82
172[02:18:33] <RoyK> is the dummy1 address on the same network
as the primary ip?
173[02:18:50] <cuddylier> What would you say 'network'
means in this context?
174[02:19:17] <cuddylier> The dummy IPs are BGP announced IPs by
bird on this same server. The primary IP is assigned to this server
via a VLAN so external to this server.
175[02:19:33] <RoyK> as in if you had a network mask of /24 and
the dummy1 address were on 45.x.x.240 or something
176[02:19:53] <RoyK> so the same IP subnet
177[02:20:27] <cuddylier> Ah okay, different IP subnets
180[02:21:23] <cuddylier> This server just has a single network
uplink with one vlan yeah. The BGP announced IPs on the server
aren't specified in the VLAN of course, just the primary IP /29
181[02:21:44] <cuddylier> Nothing outside the server though that
would rewrite IPs/block anything.
182[02:22:05] <cuddylier> if anything was blcoking then I should
have no connection with e.g. wget when using --bind-address
183[02:22:39] <cuddylier> it's the same with all
applications, Nginx, wget etc all just show their source IP as the
primary bond0 interface IP no matter what I specify in their
configs.
184[02:22:55] <RoyK> I beleive I have seen that happen. Linux
just uses the primary for everything outgoing. Anyway - mixing
different IP networks on the same L2 network isn't really
recommended
185[02:23:41] <RoyK> also - using a VM or container of sorts
would probably separate things a wee bit better
186[02:23:56] <cuddylier> VM definitely 'fixes' this
issue yeah
187[02:24:01] <cuddylier> I just can't use VMs in this case
hmm
188[02:24:16] <RoyK> not a VM? not docker? not anything? why?
189[02:24:49] <RoyK> But please consider using different VLANs
for different IP subnets - otherwise it'll just be a mess
190[02:24:54] <cuddylier> Well, I guess it may be possible to
just use one large VM but it seems sorta pointless to introduce an
overhead for such an issue but if it's really the only solution
then maybe I've no choice.
191[02:25:00] <cuddylier> But the IPs aren't specified on
the VLAN end.
192[02:25:09] <cuddylier> The BGp announced IPs aren't VLAN
related at all
193[02:25:23] <cuddylier> The only IPs in a VLAN is the assigned
/29 to bond0
195[02:25:54] <cuddylier> I don't trunk VLANs down to the
server either
196[02:26:11] <RoyK> there isn't much overhead on a kvm vm
these days and things like LXC or Docker are about zero on the
overhead
197[02:26:24] <cuddylier> E.g. if I assign multiple subnets on a
VLAN and add them to the server's bond0 interface, I can bind
to those IPs totally fine, it's only an issue it seems with
using a separate interface such as dummy1
198[02:26:27] <RoyK> well, do - it's not really hard to
configure
199[02:26:58] <RoyK> or just setup a VM for each role the server
plays
200[02:27:10] <RoyK> isolation isn't only about corona :D
201[02:27:49] <RoyK> may I ask what the different roles of this
server are?
202[02:27:59] <cuddylier> Its sole purpose is running a Nginx
proxy
203[02:28:39] <cuddylier> So traffic comes from remote web
servers to the various IPs inbound then the server uses its primary
interface IP to return the connection.
204[02:28:50] <RoyK> just setup a base system, rather minimal,
and then two (or more) kvm VMs. Put them on different VLANs and you
have good separation
205[02:29:18] <RoyK> the overhead really isn't much
206[02:29:34] <RoyK> modern CPUs just virtualise in the CPU
itself
207[02:30:01] <cuddylier> Yeah, in this case though, I
don't need any separation whatsoever, the current setup is
perfect except I just need all outbound connections to use a single
IP from the ones that are on the dummy1 interface, I just wanted to
avoid more complexity if it wasn't necessary.
208[02:31:54] <RoyK> I beleive there are some iptables hacks to
allow changing the source address, but I never managed to make that
work (haven't tried in years, though). Just use some VLANs and
VMs. It's easy and easy to document so that whoever takes over
the machine management later can understand it without digging too
far into things.
209[02:32:05] <RoyK> Sometimes simplicity helps a lot ;)
225[02:50:32] <RoyK> cuddylier: the point is, if you have two
addresses on different ip networks and you send out a packet, it
will be sent to the default gateway, of which you can only have one
(at least in practice). It's possible to fix that with some
tricky routing, but really, the best thing is to isolate the two in
VMs on different VLANs
226[02:51:00] <cuddylier> I tried the iptables SNAT option but
sadly that seems to not be changing the source IP either.
244[02:58:29] <annadane> ralink is generally packaged in debian,
i'd probably try to find the debian package for it first before
running make from a vendor's website
245[02:58:56] <RoyK> yeah
246[02:59:01] *** debhelper sets mode: +l 1142
247[02:59:06] <RoyK> I didn't know that was RALink, though
248[02:59:13] <jennie> its not connected to internet, i dont
have any cable to connect this computer to internet
249[02:59:28] <jennie> i cant do anything on this debian until i
get wifi usd work
250[02:59:44] <jennie> how do i copy text from terminal?
254[03:03:08] <annadane> i guess the thing to do if you have
connection on a separate computer (and you must or you wouldn't
be talking to us) is to download the .deb and transfer it via usb
stick or something to the non-internet computer and install it that
way
255[03:03:13] <annadane> unless i'm mistaken
256[03:03:32] <jennie> where is the .deb please tell me
257[03:03:32] <annadane> RoyK can back me up, or not
307[03:11:32] <annadane> just choose any (relatively local)
mirror
308[03:11:53] <annadane> copy to usb, transfer it to the other
machine, and dpkg -i it on that other machine
309[03:11:58] <NiDieu> they aren't critical logins, but
lazy to paste them over to other instances
310[03:14:09] * annadane wonders what genom is
311[03:15:49] <annadane> jennie, the thing with vendor
instructions is that yeah while you're getting the instructions
from an "official" website as opposed to some shady thing,
there's a lot of cheap hacks around linux and people will
happily tell you to mix releases or sudo make or this or that; when
in doubt, ask us first
312[03:16:13] <annadane> just because something is on the
internet doesn't automatically make it correct, in the case of
linux and specifically debian which has a well defined policy of
where stuff goes
313[03:16:56] <annadane> or, the debian wiki (which is hit or
miss, but still)
314[03:17:05] <jennie> dpkg - i firware.deb error
315[03:17:05] <dpkg> jennie: I don't know, could you
explain it?
316[03:17:15] <jennie> ldconfig not found in PATH or not
executable
320[03:17:33] <dpkg> When you get random packages from random
repositories, mix multiple releases of Debian, or mix Debian and
derived distributions, you have a mess. There's no way anyone
can support this "distribution of Frankenstein" and
#debian certainly doesn't want to even try. Ask me about
<reinstall>
321[03:17:36] <jennie> expected programs not found
322[03:17:43] <annadane> i wonder if that's a su path issue
323[03:18:02] <RoyK> annadane: sudo -i
324[03:18:09] <jennie> root's path should usually contain
/usr/local/sbin, ....
325[03:18:23] <dvs> sounds like "buster su"
326[03:18:36] <annadane> difference between /usr/sbin and
/usr/local/sbin
327[03:18:44] <RoyK> jennie: erm - what do you have in
/usr/local/sbin ?
328[03:18:44] <annadane> try becoming root via su - or su -l
instead
329[03:18:56] <annadane> !buster su
330[03:18:56] <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...). To approximate the previous behaviour, put
"ALWAYS_SET_PATH yes" in /etc/login.defs. See
replaced-url
331[03:19:17] <jennie> RoyK its empty
332[03:19:23] <RoyK> it should be
333[03:19:34] <jennie> how to install? whats going on?
334[03:19:36] <RoyK> unless you have installed something by hand
335[03:19:54] <jennie> i did make file 1 hour ago on that driver
which i downloaded earlier
336[03:19:57] <RoyK> jennie: sudo -i or su - and then, when
you're root, dpkg -i <filename>
337[03:20:15] <jennie> i am at root@debian terminal
338[03:20:35] <dvs> and how did you get there?
339[03:20:35] <RoyK> echo $PATH
340[03:21:02] <jennie> dvs i did su 20minutes ago
341[03:21:15] <dvs> and hence buster su ^^^
342[03:21:17] <RoyK> jennie: as annadane pointed to above, you
don't get root's environment with 'su' alone,
you need 'su -' or 'su -l' to get that
347[03:22:43] <RoyK> jennie: what buster has reverted to has
been the standard for 30+ years. There were only a version or two
where su picked up root's environment without it being
speicified before that.
349[03:22:55] <RoyK> jennie: was that $path or $PATH?
350[03:23:04] <jennie> $PATH
351[03:23:19] <RoyK> so dpkg -i thatfile
352[03:24:12] <annadane> and then because it's firmware a
reboot is likely required before the wifi can work
353[03:24:18] <RoyK> btw, dpkg should be under /usr/bin so it
should be available in the path of a normal user as well (but not
with the rights to install or remove anything)
355[03:24:38] <jennie> ok, i rebooted it, now i am normal user,
not root, i did sudo -i with normal user and entered password, and
now its saying that user 'deb' is not in the sudoers file,
i will be reported
356[03:24:58] <RoyK> jennie: then use su -
357[03:25:06] <RoyK> jennie: you can fix sudo later
358[03:25:13] <annadane> su - <enter> password
<enter>
371[03:35:17] <annadane> (slackware probably would be a
legitimate time to sudo make but no one includes slackware in their
install instructions so...)
382[03:41:00] <annadane> every distro has its cheerleaders i
suppose
383[03:41:13] <annadane> "i love
slackware/debian/ubuntu/mint/gentoo/manjaro/openbsd and i
wouldn't ever use anything else"
384[03:42:38] <RoyK> annadane: hehe
385[03:42:47] <jennie> how do i copy text from terminal?
386[03:43:09] <RoyK> annadane: I just use debian and sometimes
ubuntu these days (but at work, mostly centos/rhel)
387[03:43:11] <annadane> there's a package for that but i
don't recall what it is
388[03:43:59] <RoyK> jennie: just mark the text you want to copy
and switch to the other window and press the mouse wheel if you have
one to paste - or rightclick and choose paste
389[03:44:19] <RoyK> jennie: but please use a pastebin
390[03:44:39] <RoyK> jennie: pastebinit is a nice command to
pipe data directly from a command to a pastebin
403[03:46:20] <RoyK> jennie: please understand that while some
of us speak very directly, we all have to start somewhere and you
have just started. Welcome :)
481[04:02:45] <RoyK> and preferably, add 'contrib' and
'nonfree' at the end of these lines
482[04:03:06] <jennie> what do you mean by comment out, are they
already not commented out? i mean what I understand by commenting
out is to put # in the beginning of the lines
483[04:03:07] <RoyK> so basically, just start from scratch in
that file
484[04:03:39] <dvs> !tell jennie about buster sources.list
489[04:07:36] <annadane> a # at the start of a line means
"this is just text not to be executed", essentially, or a
comment (programmers can comment their code to make it more clear to
people reading their code)
490[04:07:40] <annadane> so uncomment means remove the #
556[05:09:13] *** Quits: yonder (~yonder@replaced-ip) (Remote host closed the connection)
557[05:11:05] <alexrelis[m]> I have a Thinkpad that supports
booting from USB, but all the USB ports are 2.0. At the same time, I
have an ExpressCard to USB 3.0 adapter, which allows me to utilize
USB 3.0 peripherals at their full potential. Does anyone know if
it's possible to boot a LiveUSB from the ExpressCard USB 3.0
port?
558[05:12:36] <annadane> no idea, i imagine so
559[05:12:44] <annadane> !tias
560[05:12:45] <dpkg> TIAS is "Try It And See".
561[05:13:25] *** Quits: Grldfrdom (uid391113@replaced-ip) (Quit: Connection closed for inactivity)
570[05:31:30] <Jmabsd> When you have a freshly installed Debian,
may it have outdated packages repository, so that if you install the
nonfree firmware for instance, you would get a *too old* version?
571[05:32:40] <Jmabsd> As in is there any advantage to
"aptitude upgrade; aptitude update; aptitude install -y
intel-microcode", *OVER* just doing "aptitude install -y
intel-microcode"?
582[05:49:40] *** Quits: sklei4 (~sklei4@replaced-ip) (Quit: see you later space cowboy)
583[05:52:12] <nickdaly> Jmabsd: for the second case: if your
packages are sufficiently outdated, the locally-known version may no
longer exist on the package servers, and you'll get a
can't-find-this-package error. Fix it by running the first case
:)
635[07:04:15] *** Quits: xcm (~xcm@replaced-ip) (Remote host closed the connection)
636[07:04:25] <bracham> so i'm not getting much time during
the week to work on setting up my new server system, but at least it
seems that each time i step back here for a few minutes i figure out
something else i'm using on the old (ubuntu server) system that
i'll have to set up on the new (debian) system, so that's
a plus. although it's starting to feel like a bigger project!
641[07:07:14] <RoyK> bracham: I spent some time using ubuntu on
servers and I went back to debian as well. I still use ubuntu on
laptops and such, though
642[07:07:20] <RoyK> annadane: ROTFL
643[07:07:29] <RoyK> annadane: the mouseover there is brilliant
652[07:12:29] <RoyK> bracham: well, Randall usually releases
three per week, so just go there and click backbackbackbackback :)
653[07:12:44] <bracham> when i came to linux from windows i
started with ubuntu. i found i could fairly easily set up a home
server, so i did that with ubuntu. now, as much as ubuntu is easy to
use, i'm working on getting away from them as they're
getting too much like windows, IMO
654[07:13:07] <bracham> RoyK, honestly i would, but as it is i
already don't have enough hours in the day!
658[07:13:27] <search_social> ironically i found the arch linux
wiki to be more helpful for installing debian than the debian wiki
659[07:13:33] * RoyK started off with slackware 2.1, but that was some
time ago
660[07:14:01] <search_social> and the one extra tip is all kinds
of things don't work for no reason (desktop, wirless) unless
you upgrade from the included 4 kernel to a 5 kernel
661[07:14:27] <RoyK> ubuntu is nice for things like laptops,
since it's more bleeding edge-ish
662[07:14:39] <RoyK> debian is nice for servers since it's
more focused on stability
663[07:14:49] <bracham> i'm a simple guy though, i just
want something that will work
664[07:14:58] <search_social> debian has supported all the
hardware on my very new laptops .. AFTER using a usb stick to apt
install the 5 kernel
665[07:15:04] <qman__> ubuntu server used to be a great system,
but it got more and more bloated and buggy as time went on
666[07:15:29] <bracham> i will be happy when i get my debian
server all set up and can switch to it!
667[07:15:30] <nkuttler> i find that running sid is far more
stable than running ubuntu though
668[07:15:32] <qman__> particularly when they started adding
stuff like plymouth, and the switch to systemd didn't go so
well
670[07:15:48] <sponix> search_social: I take the lazy route and
use MX Linux AHS branch
671[07:16:09] <search_social> i was annoyed at getting several
"snaps" installed for no reason - for example
gnome-calculator was a "snap" for some reason
672[07:16:18] <bracham> so with my debian server, should i be
looking to upgrade the kernel?
673[07:16:32] <RoyK> bracham: not if you don't need to
677[07:16:43] <sponix> bracham: normal rule of thumb for kernels
is -- if all hardware is working well - leave it alone
678[07:16:54] <bracham> sponix, that's kinda what i figured
679[07:17:04] <bracham> thanks guys
680[07:17:08] <RoyK> if it works, don't fix it!
681[07:17:11] <search_social> thanks sponix
682[07:17:20] <bracham> RoyK, exactly!
683[07:18:09] *** HellCatBluesyCat is now known as sleepingWellCat
684[07:18:25] <RoyK> And then, there's this spurious error
in dmesg, should I upgrade? Probably not :)
685[07:18:34] *** sleepingWellCat is now known as spinningCat
686[07:18:36] <bracham> also, i'm kinda trying to take my
time with this setup, as i've learned a few things in the
meantime so i want to set some things up a whole lot differently
687[07:18:52] <RoyK> bracham: what sort of setup is this?
695[07:22:10] <bracham> apparently the mainboard only recognizes
up to 4GB of ram, so that's a bummer, but if i have some spare
cash laying around one day and / or i find 8GB of reasonably priced
ddr2 ram, i might try it
696[07:23:03] <RoyK> ddr2 is the type of thing that people throw
away
697[07:23:14] <bracham> it's just a home server for print,
scan, media streaming, http, sftp, and more
698[07:23:28] <bracham> well that's what my board takes
701[07:23:54] <laidback_01> dang. I have a virtual that will
drift time ... about a minute every 24 hours. I have to reboot it
daily to keep it in sync. virtual is Debian running on FreeBSD
bhyve. I put in the systemd.network NTP=, and they sync at boot now,
but is there a way to keep the clock synced the way that ntp-client
used to pre-systemd?
702[07:23:57] <RoyK> and for that use, probably not much need
for memory either
703[07:24:21] <bracham> no i don't seem to ever be running
out of memory
704[07:24:29] <laidback_01> is there an equivalent in the
systemd world which will work like ntpd in the pre-systemd debian?
706[07:24:37] <bracham> although it would be nice to have more
if i run into it and it works
707[07:24:40] <RoyK> laidback_01: hm - running ntpd or just
systemd timesync?
708[07:25:03] <RoyK> some people prefer chrony
709[07:25:27] <laidback_01> well, I'm not used to timesync,
but I have to use it because ntpd is not available in 10.4
710[07:25:35] <bracham> i looked through the compatibility list
on the manufacturer site and max is 4GB. if that's up to date,
that sucks. oh well, it's old.
711[07:25:40] <laidback_01> only if you are running a server do
you want ntpd according to the docs now
712[07:25:40] <Abrax> the gentoo chan still exists
713[07:25:46] <RoyK> ntpd is a wee bit touchy for drifting
"hwclock", which can be rather lousy on a vm
714[07:25:53] <Abrax> woops
715[07:26:11] <RoyK> chrony is the preferred thing on
rhel/centos 7 and 8 for that reason
757[08:25:02] *** Quits: edlou (uid413273@replaced-ip) (Quit: Connection closed for inactivity)
758[08:25:46] <colo_work> is there a "tiny" httpd
implementation in Debian (like thttpd once was) that can be
configured to serve up directory listings and files from only the
command line, and that offers proper support for byte-range
requests?
767[08:30:28] <rudi_s> colo_work: You coul always write a
minimal Go program; Go has a powerful built-in webserver; all you
need is http.Handle("/",
http.FileServer(http.Dir("."))) and
http.ListenAndServe(":80", nil) to serve static files from
the current directory
768[08:30:52] <rudi_s> I also haven't tested if it supports
byte-ranges but I suppose so
769[08:30:57] <colo_work> sponix2ipfw, it requests the server to
serve only a number of subsets of the resource requested. think,
like, the first 100 bytes of a 10GB file, instead of everything in
one go.
770[08:31:12] <colo_work> rudi_s, I think I'll go with
replaced-url
774[08:31:42] <rudi_s> colo_work: A (not-well tested) web-server
in C, are you crazy?
775[08:32:16] <rudi_s> From a person learning C "I am
reading Computer Systems: A Programmer's Perspective. It
teachers me how to write a tiny web server in C." ...
776[08:32:34] <colo_work> don't panic; I use it to stream
video via a browser browser from host A to host B for a few minutes
at a time ;)
777[08:32:57] <colo_work> (on my LAN)
778[08:33:00] <rudi_s> Still, why would you use such a software.
779[08:33:12] <alexrelis[m]> annadane: I've tried but it
won't boot. I'm guessing it could be possible if I had
another flash drive that went inside the USB 2.0 port that booted to
the ExpressCard USB 3.0 port.
846[09:11:04] <ratrace> ws2k3: and really, twiced a day is too
much. with default 30 or less to expiration, weekly is more than
enough, though I prefer daily checks with 15 days to expiration
886[09:58:16] <ueb> from Debian: dig +short myip.opendns.com
@resolver1.opendns.com
887[09:58:29] <ratrace> ueb: only if there's NAT going on
888[09:58:45] <ueb> this command responds with 154 (same of win
dev)
889[09:58:56] <ratrace> ueb: now wait, you've introduced a
whole different concept, DNS, into the story
890[09:59:00] <ratrace> !xy
891[09:59:00] <dpkg> Slow down for a bit! Are you sure that you
need to jump through that particular hoop to achieve your goal? We
suspect you don't, so why don't you back up a bit and tell
us about the overall objective... We know that people often falsely
diagnose problems because they are too close to them -- it's
easy to miss that there is a better way to proceed. See
replaced-url
892[09:59:36] <ueb> curl ifconfig.co, responds with 153
894[09:59:50] <ueb> dpkg ok, what do you suggest then?
895[09:59:51] <dpkg> ueb: are you using Windows?
896[10:00:12] <ratrace> oh this should be fun.... do we tell
them.... :)
897[10:00:15] <ueb> dpkg on dev machine Windows, on prod server
Debian
898[10:00:15] <dpkg> ueb: I'm not sure, is it larger than a
breadbox?
899[10:00:26] <ratrace> ueb: dpkg is a bot here :)
900[10:00:39] <ueb> hehehe :)
901[10:00:52] <ratrace> _anyway_ ... you're mixing DNS and
IP addresses into the same problem, so perhaps start at the
beginning. HOW are you testing WHAT exactly
902[10:01:16] <ratrace> "public" (ie. subnets
"outside" your LAN/WAN gateway) would see your
"public" IP __only__ if there is NAT going on at the WAN
gateway
905[10:02:06] <ratrace> and if you can ping an external subnet,
and get your "LAN" IP visible there? then you don't
have NAT at all, and you're NOT on a RFC-1918 non-routable IP
subnet
911[10:03:47] <ratrace> okay, the word "caller" does
not exist in networking, so we're talking about sources and
destinations here. what's your source subnet, and what's
the destination subnet?
923[10:06:07] <ueb> with that php script I checked the IP of
caller machine, with variables HTTP_CLIENT_IP, HTTP_X_FORWARDED_FOR
and REMOTE_ADDR
924[10:06:51] <Unit193> whatismyip.com and ifconfig.me are two
handy sites for that, yep.
925[10:07:32] <ratrace> ueb: so what are the IP addresses here?
if they're RFC1918 subnets, you don't have to censor and
you can post them here
926[10:08:01] <ueb> no they're not subnets
927[10:08:41] <ratrace> there are always subnets involved....
928[10:09:01] *** debhelper sets mode: +l 1157
929[10:09:16] <ratrace> eg. if your IP is 192.168.1.3,
you're in a 192.168.0.0/16 subnet, or a 192.168.1.0/24 subnet,
or some other segmentation, but there's _always_ subnetting
involved.
930[10:09:20] <ueb> 2.X.X.153 (Debian)
931[10:09:28] <ueb> 2.X.X.154 (Windows)
932[10:09:53] <ueb> 192.168.1.42 (Debian)
933[10:09:54] <ratrace> alright, so 2.x.x.x is a routable IP.
oyu probably don't have any NAT going on at the WAN gateway
934[10:10:02] <ueb> 192.168.1.109 (Windows)
935[10:10:51] <ratrace> 192.168.1.* are non-routable IPs, so
you'll need NAT going on the WAN gateway.... so which IP
address you get, when you "call" from your Debian box, to
a public site?
936[10:11:54] <ratrace> by "public" I mean something
"outside" of your LAN, something not in the 192.168.0.0/16
subnet
937[10:11:56] <ueb> dig +short myip.opendns.com
@resolver1.opendns.com (replies with 2.X.X.154)
938[10:12:07] <ratrace> no, don't use dig. leave DNS aside
for now.
939[10:12:21] <ueb> curl ifconfig.co (replies with 2.X.X.153)
940[10:12:36] <ratrace> from which box?
941[10:12:44] <ueb> from Debian
942[10:12:56] <ratrace> the one that has 2.x.x.153 public IP on
its network interface?
956[10:18:53] <ratrace> ueb: okay, so that box is only in the
192.168.0.0 subnets. there's probably NAT going on if it
translates to 2.x.x.153 public IP .... so far so good. so what
exactly is the problem here?
958[10:19:51] <ratrace> ueb: so maybe your network setup is such
that you have multiple public IPs in the 2.x.x.y subnet, and your
router is NAT'ing individual public IPs to individual LAN IPs
960[10:20:53] <ratrace> you'll have to query the windows
box for same data as `ip a` on Debian (I think it's `ipconfig`
in a windows terminal) , I don't know windows networking
really, and see if behaves the same: has only 192.168.x.x
address(es) assigned.
961[10:21:09] <ratrace> then you check your router
config/admin-panel/whatever and see how NAT'ing is set up
1030[11:11:04] <dpkg> Bug Tracking System for Debian packages,
replaced-url
1031[11:11:11] <ratrace> debian doesn't have Bugzilla
1032[11:11:34] <xvilka> Yeah, quite sad. And while you have Salsa
you seem not to use it.
1033[11:11:47] <xvilka> You should move eventually to something
more user-friendly
1034[11:12:32] <ratrace> which ain't gitlab
1035[11:12:49] <enyc> xvilka: i don't think reportbug is
that complex myself
1036[11:13:01] <ratrace> infact, I like Bugzilla very much. it
also supports hooks and email based APIs so hardliners can still
manage bugs from their mutts :)
1037[11:13:01] <enyc> colects some key information
1038[11:13:15] <xvilka> ratrace: Bugzilla is fine. BTS is not
1039[11:13:31] <ratrace> yup
1040[11:13:37] <enyc> xvilka: become a debian developer an make
the bridge/update ?
1041[11:14:07] <ratrace> you can't without committee
involving, voting and entire dev community deciding, which they
won't.
1062[11:26:42] <Azrael_-> i'm running debian 9 with full
system encryption. if i want to upgrade the whole system, do i have
to look out for caveats or can i just start the upgrade with apt?
1091[11:52:47] *** Quits: tmroland (~tmroland@replaced-ip) (Remote host closed the connection)
1092[11:54:25] <martinus__> apt question : I want to pin a
package from a third party repository (gitlab-ce), and force the
admin to do something specific when it's time to upgrade. I
thought about pinning the package but then I don't see how I
can use the '-t' option to force apt to upgrade
1093[11:54:26] *** Quits: bolt (~r00t@replaced-ip) (Remote host closed the connection)
1109[12:06:05] <eblip> ok i have found a version 79 of google
chrome...if i uninstall version 83 and install version 79,..when i
update it will resort back to 83 ..how do i stop this
1130[12:23:10] <ksk> eblip: why would you want to install an
older version of the browser?
1131[12:23:38] <eblip> ksk becasue some programs i need to run
wont work on this new version
1132[12:23:41] <ksk> also, chrome is third party, not part of
Debian. You would need to redirect your support request to google
please.
1133[12:24:00] <TuxCrazy> are other distros based on Debian as
secure as Debian?
1134[12:24:10] <ksk> !based on Debian
1135[12:24:10] <dpkg> Your distribution may be based on and have
software in common with Debian, but it is not Debian. We don't
and cannot know what changes were made by your distribution (compare
replaced-url
1136[12:24:20] <ksk> TuxCrazy: We cannot (do not want to) tell.
1137[12:24:27] <TuxCrazy> there are so many lesser known distros
based on Debian.
1139[12:25:03] <TuxCrazy> how can I get packages of newer
versions on Debian stable?
1140[12:25:05] <ksk> I personally would put some thinking into
the "how does this forks dev team look like compared to debian,
debian-security"-question
1141[12:25:14] <ksk> !backports
1142[12:25:15] <dpkg> A backport is a package from a newer Debian
branch, compiled from source for an older branch to avoid dependency
and <ABI> complications.
replaced-url
1144[12:25:43] <TuxCrazy> for example, if I need a newer or the
latest version of LibreOffice and I am using Debian stable.
1145[12:26:04] <ksk> TuxCrazy: In general: I would not get newer
version of XY just because (shiny new things syndrom, I dont recall
the bots trigger for that, right now..)
1146[12:26:17] <EdePopede> !sns
1147[12:26:17] <dpkg> Shiny New Shit Syndrome is a serious
disorder, which usually breaks out into an epidemic every time
something new is released. If you have SNS, ask me about
<backports> and <ssb>; these are better options than
upgrading to <testing> because it is a <moving target>.
1148[12:26:46] <eblip> ksk ..we use debian for business systems,
programming, lots of stuff..debain should not struggle to handle 3rd
party stuff...most companies require them to some extent
1156[12:28:26] <ksk> TuxCrazy: If you _really_ need version XY
you could always look into getting it from upstream (as in: Does
libreoffice supply debian packages/mirrors, or an installer for
linux?)
1157[12:28:43] <TuxCrazy> are packages from debian backports as
reliable and secure as those from debian stable?
1158[12:28:45] <ksk> You are leaving the Debian ecosystem at that
point of coure.
1215[13:08:52] <nlpqda> encrypted debian hdd failed to boot,
tested with smartmontools no problem, ext4magic no problem, but not
able mount partitions when I boot from other drive! tried to replace
superblocks showing errors and but not least cryptsetup luksOpen is
not doing any good
1254[13:48:14] <ksk> nlpqda: can you share details?
paste.debian.org if you have a network-enabled system running (use
for example pastebinit (install via apt))
1255[13:48:22] <ksk> !paste
1256[13:48:22] <dpkg> Do not paste more than 2 lines to this
channel. Instead, use for text:
replaced-url
1313[14:27:56] <ConnectingMedia> I asked this question over in
##kernel and was pointed here. How can I set the maintainer for the
.deb package when compiling a kernel with `make deb-pkg`?
1355[15:04:34] *** Quits: tijara (~tijara@replaced-ip) (Disconnected by services)
1356[15:05:12] <ws2k3> so im having a kind of weird thing.when i
run dig ns mydomain +short returns emty while on some other domains
it works just fine
1374[15:14:51] <ws2k3> ksk: no domain is in production just fine
1375[15:15:06] <ws2k3> domain works in the broser i can ping it
etc that looks fine
1376[15:15:08] <greycat> I believe ksk is trying to say,
"point dig directly @ the authoritative server for this
domain"
1377[15:15:57] <ws2k3> greycat: ksk did that. still emty
1378[15:16:06] <greycat> ws2k3: it is impossible to help you if
you do not reveal enough information for people to duplicate your
results. We would need to know the actual domain name you are
resolving. If it's private (inside some corporate firewall
where we can't see it), we can't help.
1379[15:16:27] <ksk> I would then think you dont have NS records
for that domain. I was asking for the domain so I could check myself
if I found anything odd about it.
1404[15:22:04] <ksk> (and spoiler: have fun configuring that in
bind!)
1405[15:22:04] <deadrom> hi
1406[15:22:05] <greycat> ws2k3: at this point we are trying to
figure out what MISUNDERSTANDING you are holding on to that is
preventing you from seeing reality
1408[15:22:44] <greycat> Did you think, perhaps, that if you do
an NS query for foo.bar.com and get no-such-thing as an answer, it
will recurse and try bar.com instead?
1409[15:22:50] <ws2k3> let me explain a bit more. we use dig ns
domain @1.1.1.1 +short to check if we manage the dns for that
domain. cause if so. then ican renew its ssl certificate by dns
1417[15:24:19] <ws2k3> greycat: can i pm it to you if you realy
want to know. irc is getting logged and i would like to not be
associated with the domain to much
1418[15:24:26] <greycat> *plonk*
1419[15:24:55] <ws2k3> greycat: i think the domain is not
relevant to the issue cause bla.google.com is showing the same
behavior
1420[15:25:06] <ksk> Do you read what I write at all?
1421[15:25:11] <greycat> I'm going to take that as a
"yes" on the "I thought dig NS bla.google.com would
be equivalent to dig NS google.com" theory.
1422[15:25:15] <greycat> It's not.
1423[15:25:36] <deadrom> I run a 6 drive NAS, and it's a bit
noisy and draws power all the time, drives need to wake up when I
need it etc. So my plan: move RAID server to attic, build a small
system with only one 16TB drive and cron that machine to WoL the
raid server as a backup facility and rsync to it, then shut it down.
voila, silence and finally proper backup. any
1424[15:25:37] <ws2k3> greycat: yes I thought dig NS
bla.google.com would be equivalent to dig NS google.co
1431[15:28:14] <ksk> deadrom: you can still cut out the small
system and let your computer like wakeup at $date and run this
wakeup and backup from there? I dont see the point of addind a
second server here.
1433[15:29:08] <ksk> ws2k3: what does "manage the dns"
mean? its not 100% portable to dns ressource records.
1434[15:29:37] <deadrom> ksk: "cut out the small system and
let your computer like wakeup at $date and run this wakeup and
backup from there" <- sorry, what?
1435[15:29:42] <ws2k3> ksk: we have alot of domain. if we manage
the dns for the domain. that means we can renew it by dns
1438[15:31:17] <ksk> either do that on a per server basis (hello
letsencrypt, certbot, traefik, whatnot) - or do it centralized. for
the latter do rely on a list of domains you own, which is stored in
a central place in your logic, do not query dns to find domains you
own!?
1439[15:31:47] <ksk> deadrom: Why do you need the "small
Server"? What are you really trying to do?
1440[15:32:33] <ksk> Why cant you just move the NAS into the
basement and keep copying backups from your personal computer to it,
just like you do now? (ignoring the fact that it is, in fact, in a
different location)
1442[15:32:49] <deadrom> ksk: 1disk-miniserver is supposed to sit
in living room and be a media center while at it. so silent
operation is key. 2nd server: backup.
1446[15:34:23] <deadrom> ksk, 2nd machine does not exist yet.
that riad server sits in my living room and is quite a bit loud. I
don't have a basement, only attic. the requirements are
somewhat fixed, there is no good place other than the attic for a
noisy machine.
1517[16:14:56] *** Quits: quadrathoch2 (~quadratho@replaced-ip) (Remote host closed the connection)
1518[16:15:31] *** Quits: pwerti (~pwerti@replaced-ip) (Quit: My MacBook Air has gone to sleep. ZZZzzz…)
1519[16:16:30] <unborn> DJDan: is your key having correct
permissions on your system? is it deployed correctly? normally you
should not have had any issues.. it should just work..
1520[16:17:02] <greycat> I'm guessing "oh finally
yay" means "I found my mistake and fixed it"
1540[16:25:02] <deadrom> DJDan : your system has to be up against
an attack whether it is visible or not. hiding from ping sweeps
simply is security by obscurity.
1541[16:25:03] <DJDan> because when i view my icmp traffic people
are trying to udp find/attack sip and other stuff
1542[16:25:08] <unborn> DJDan: what icmp attacts could do to you
or your platform in those days... think about it
1543[16:25:34] * themill has fond memories of the ping of death
1544[16:25:58] <DJDan> i have notice alot of activity when
tcpdump icmp looking for various udp ports and sip etc...
1545[16:26:20] <unborn> DJDan: and?
1546[16:26:52] <greycat> If you have a *web server*, that means
people know you exist, so why would they need to probe for your
existence with ping?
1547[16:27:02] *** Quits: mthe878 (~mthe@replaced-ip) (Quit: Lost terminal)
1548[16:27:03] <deadrom> DJDan, your system is on the internet,
people scan it.
1549[16:27:22] <DJDan> well you could have various other
services. does portscanners use icmp
1550[16:29:48] <greycat> ...
1551[16:29:48] <unborn> DJDan: they will scan it for sake to see
how capable your services are and what you can offer as well.. its
normal
1556[16:29:48] <greycat> The best thing I can say at this point
is, at least you switched from completely wrong statements, to
clueless questions. Questions are an improvement.
1557[16:29:48] <unborn> DJDan: not true...
1558[16:29:48] <deadrom> DJDan: if I look for a target of
opportunity I'd rather see what kind of service my tool set is
able to attack and scan for those.
1559[16:29:48] <greycat> Oh, whoops, you switched back to
absolutely wrong statements again.
1568[16:29:48] <unborn> DJDan: if I would scan anything I would
do for my knowledge as novadays everything
1569[16:29:56] *** Joins: antoine (~antoine@replaced-ip)
1570[16:29:57] <deadrom> greycat and fuck you double, you
miserable asshole
1571[16:30:05] <greycat> !icmp
1572[16:30:05] <dpkg> i heard icmp is Internet Control Message
Protocol. or DO NOT BLOCK ALL ICMP or someone will thwack you and
you will have deserved it., or use --state RELATED in iptables to
selectively block and still have legal icmps pass
1573[16:30:11] <deadrom> I remember when you were helpful
1574[16:30:13] <unborn> is filtered.. no one scans you for attack
that's insane
1581[16:30:27] <DJDan> i asked weither i should block icmp cause
someone mentioned he did on his vps and i thought it was stupid
cause then you couldnt check if your server is actually active when
your webserver is down
1600[16:32:51] <unborn> DJDan: i asked weither i should block
icmp cause someone mentioned he did on his vps - goes same with me
as I am diabetes type 1 shall I stop sugar rise and potatoes? - you
can see, it all depends on your personal settings - you should not
ask here.. do your research about this..
1604[16:34:27] <unborn> DJDan: do as you pleased, you are in
control of your own thingy.. also this should not be even asked here
as its off topic.. btw as I am saying
1605[16:34:31] *** Quits: cupcake90 (~cupcake90@replaced-ip) (Remote host closed the connection)
1613[16:37:09] <greycat> Saw that the web page in question was
http, switched it to https in my browser, the window didn't
change, pasted the https URL into the factoid, and *then* my window
gave a timeout error.
1615[16:39:12] <unborn> same here.. also js blocked then and meh,
time outed as well on my end
1616[16:39:29] <greycat> even though some people think I'm
"not helpful"
1617[16:39:49] <greycat> funny how the ones who say that
aren't helping anyone, either
1618[16:40:13] <beardy> !helpful
1619[16:40:50] *** Quits: cupcake90 (~cupcake90@replaced-ip) (Remote host closed the connection)
1620[16:40:54] <beardy> !hurry
1621[16:40:54] <dpkg> it has been said that hurry is Please be
patient, we're all just fellow users, mostly with other things
going on.
1622[16:40:57] <unborn> anyway, man is own god, he can control
whatever he wants out there.. let it (him) be in its own world.. how
would you speak to me in slovak when you cannot...
1623[16:42:00] <beardy> That used to be more helpful.
1624[16:42:04] <unborn> - I can say and claim - I am not help
here.. possibly it will give me an ban, been there and got used to
it.
1626[16:43:04] <unborn> anyway folks lets be helpful and lets
focus on issues with debian again :)
1627[16:43:08] <deadrom> greycat: I've known herer you for
10 years or more and while your lectures were always harshly dealt
they were worth listening to because one actually learned from it.
these days you lash out at people at the slightest irritiation as if
they pissed on your shirt. look at the above right now. and then at
a small hint that your aggresion is
1628[16:43:09] <deadrom> unnecessary you return "fuck
you".
1633[16:45:24] <unborn> deadrom: hes been harsh and he had
changed.. I can say it from person by my self. - and his advice
touch seen it be harsh, he actually helped me.
1636[16:47:41] <deadrom> unborn: I've been coming here since
the the early 2000s and observe the change was for the worse. the
above pattern I observed a lot lately.
1637[16:47:44] *** Quits: ich (~ich@replaced-ip) (Quit: Ex-Chat)
1644[16:51:15] <deadrom> on-topic: how to prevent bit rot? I
thought: if a file system would hashsum files while writing and
enter that data in the file system's metadata one could scrub
at file system level: hash the file: is the hash the same? no? has
mtime changed? yes? all good, probably. if it diffres, the kernel
says "I did not write to that file at al", but yet the
1645[16:51:16] <deadrom> hash changed -> possible rot.
haven't fully fleshed out the idea yet.
1647[16:51:39] <deadrom> but maybe I'm beind the cutting
edge and there are FS that do that already?
1648[16:51:51] <unborn> deadrom: same here.. I've been
banned most of the times or perhaps muted but since forced alone I
learned a lot also everyone should be realising that this channel is
nice and good but official debian channel is somewhere else.. I can
agree with you but I would stand in defence of greycat - that guy
showed me my potential regards the bash and linux it self. I still
think even same observation on both of out ends - some respect is
need to paid on... he
1649[16:51:51] <unborn> is dealing with others by his own way and
its good.
1650[16:53:03] *** Quits: JohnML (~john1@replaced-ip) (Remote host closed the connection)
1651[16:53:46] <deadrom> we can discuss this to no end but I see
no gain in it. It's just that setting people on ignore disrupts
the communication flow so I try to avoid that. this is no official
debian channel anymore?
1652[16:54:27] <unborn> often I was here with idiotic questions -
how often someone can be annoyed by them? - we are all humans - end
of the day I love greycat even his uglyness in a same way I love you
and my beloved code or our os.
1653[16:55:34] <deadrom> we have attitude free, stone cold logic
templates for this, like "how to ask smart questions".
there is no need to be ugly.
1657[16:56:22] <deadrom> pointing out people being annoying
results in either of two: they go away, rebuffed by the scolding or
they introspect and better.
1658[16:56:48] <deadrom> differently worse or better?
1659[16:56:49] *** Acheron is now known as ancient_user
1660[16:57:03] <greycat> Since there was actually a question
asked, I will try to answer it. "does portscanners use
icmp" -- No, a port scan cannot use ICMP, because ICMP does not
use or have any concept of port numbers.
1662[16:57:18] <unborn> deadrom: for me I would ask as much as I
need within my brain capability - as english is not mi 1st language
I often was sounding here very offending - and tush been banned but
its okay
1665[16:59:24] *** ancient_user is now known as infant_user
1666[16:59:37] <deadrom> unborn: if your native language is
listed in DeepL my advice is to use their translations. the quality
of their translations exceeds google/translate quite a bit, even for
complex structures.
1667[17:00:08] *** infant_user is now known as Acheron
1673[17:01:06] <greycat> irc.debian.org is an alias for
irc.geo.oftc.net.
1674[17:02:32] <unborn> deadrom: I kindly passed issue of my own
language issues, now I speak fluently 8 languages as human so -
thank you for advice.. I do not use google as well since 2007..
1675[17:03:02] *** MathAmphetamine is now known as CrystalMath
1686[17:05:10] <beardy> However, people doing portscans
(portscanners) very often sends an ICMP Echo request (ping), and
awaits a reply. If no reply, some of them skips doing a scan of TCP
and UDP ports.
1687[17:07:03] <greycat> In the best possible case, all that
would do is avoid a little bit of extra traffic from the most stupid
wannabe hax0rs. Any competent person would not be deterred by this.
1735[17:31:33] <b1ack0p> in image 2 it holds and counts the
seconds until it finishes at " A start job is running for
/dev/disk/by-uuid/xxxxxxx (16s / 1min 30s)
1736[17:31:45] <b1ack0p> what can i do about these?
1737[17:32:30] <b1ack0p> sorry for the bad images. they look
normal in image viewer but after uploaded to imgur they changed
position
1738[17:33:29] <b1ack0p> i didnt notice on my earlier debian
installs.. i noticed them after my fresh install yesterday
1739[17:33:42] <b1ack0p> what are they about i dont know.
1740[17:34:08] <dob1> but ls -la doesn't distinguish case,
is this the point maybe
1741[17:34:10] <b1ack0p> i want to boot and shutdown debian quick
as before
1742[17:34:13] <unborn> b1ack0p: nothing will change on igmur -
its raw as you would post it.. I would have look at the script which
would take that much time ( I did not look at your images - please
note ) however if you not need those scripts remove them.. if you
are not sure what they are doing - refer to your script creator..
1743[17:34:19] <greycat> dob1: depends on your locale.
1748[17:34:45] <greycat> In the C locale, ls will use traditional
ASCII ordering. In the en_US locale, for example, it will treat
"A" and "a" the same.
1749[17:34:56] <intx> how to find a debian mirror with testing
sources?
1750[17:35:00] <greycat> Because this is how locales (LC_COLLATE)
work.
1751[17:35:10] <intx> I tried like 10 different ones none of them
work
1752[17:35:16] <b1ack0p> where is /scripts/local-block directory?
1753[17:35:17] <greycat> !testing sources.list
1754[17:35:17] <dpkg> Using "testing" instead of the
current codename (e.g. bullseye) in your sources.list isn't as
bad as using stable in your sources.list; you will just seamlessly
continue to track the "testing" distribution following a
release. Note that for a few months just after a release, testing
can be a pretty bumpy ride... perhaps running stable would be a
better idea? See also <stable sources.list>, <bullseye
sources.list>.
1756[17:35:31] <unborn> dob1: wen you log into system you should
be fine to see it with default setting in mc
1757[17:35:35] <beardy> dob1: The difference is that mc's
default is Name, and case sensitive. Uncheck "Case
sensitive" and it will sort as ls's default.
1758[17:35:38] <greycat> well, that was a disappointing factoid
1759[17:35:54] <annadane> did you try deb.debian.org?
1760[17:35:58] <annadane> !bullseye sources.list
1761[17:35:59] <dpkg> A suitable /etc/apt/sources.list for
"Bullseye" has three lines: "deb
replaced-url
1786[17:41:11] <annadane> intx, when you say "none of them
work", what output do you get? paste.debian.net
1787[17:41:40] <intx> ahh okay. you guys are right, dns
isn't working. will need to look into why
1788[17:42:00] <intx> thanks!
1789[17:42:13] <b1ack0p> in forum somebody addressed swap
partition:
replaced-url
1790[17:42:22] <b1ack0p> "Recently I've encountered a
similar problem of repetitive "local-block", and it turned
out to be a modified swap partition UUID which need to be updated in
/etc/initramfs-tools/conf.d/resume."
1791[17:42:46] <b1ack0p> so both issues are related each other?
1792[17:43:01] <b1ack0p> script/local-block and running job
uuid..
1837[18:07:05] <dpkg> extra, extra read all about it, su is
switch/set user. It is used to change User ID's and/or gain
super user access. Since Debian Buster, "su -" or "su
-l" is needed to access programs located in /sbin (see
<buster su>). It provides an root environment as if the
superuser had logged in directly. See "man su".
1838[18:07:29] <Scou_moune> yea i have read that
1839[18:08:26] *** MathAmphetamine is now known as CrystalMath
1840[18:08:51] *** Quits: behanw (uid110099@replaced-ip) (Quit: Connection closed for inactivity)
1841[18:10:38] <Scou_moune> I have make a /var /tmp /home /
/boot, but that first time i make a /var /tmp in other partition
1842[18:11:11] <Scou_moune> Maybe the error: command failed: Too
many open files in system (-23) due to that ?
1843[18:11:19] <Scou_moune> or that a bug with iw
1844[18:13:39] <Scou_moune> RoyK: ok su - root lol
1867[18:21:16] <tharkun> RoyK: :) Be safe. I suppose it
eventually will be but I need to configure it connecting directly to
it using the usb connection But I do not know the tool to do so. I
currently use it through the usb port but it is not an option for
other devices.
1870[18:22:17] <RoyK> Scou_moune: that's normal - it just
means everything is using the defaults
1871[18:22:35] <Scou_moune> "In the future iw will become
the canonical command line tool for wireless configuration and
iwconfig/wireless-tools will no longer be required. See
/usr/share/doc/iw/README.Debian for a more detailed overview of iw.
"
1872[18:22:47] <Scou_moune> yea for me iw bug actualy xD
1873[18:22:59] <Scou_moune> RoyK: ok
1874[18:24:25] <tharkun> !tell Scou_moune about lsof
1901[18:44:41] <Scou_moune> i have only the error: command
failed: Too many open files in system (-23), when i use this cmd: iw
phy phy0 interface add wlp4s0 type managed
1902[18:44:54] <Scou_moune> So at this moment i have 0 file open
1903[18:45:44] <greycat> You do NOT have zero files open.
1904[18:45:52] <greycat> You have LOTS of files open. At least 3
per process.
1905[18:46:42] <greycat> However, it might be IMPLIED that you
are running that iw command AS ROOT. (It would be nice if you SAID
so, especially after other people talked to you about checking the
number of open files for the USER having the issue.)
1906[18:46:58] <greycat> Per-user system resource limits on open
files probably aren't in effect for root. Maybe.
1931[18:51:45] <EdePopede> Scou_moune: every program you run
opens files. just look into the number subdirectories of /proc, in
their fd subdirs the links all point to open files
1932[18:52:07] <EdePopede> just like you don't have to start
a single program after boot but a lot are running already
1936[18:52:52] <RoyK> Scou_moune: try to run ulimit -n 30000 or
something before running that command
1937[18:52:58] <tharkun> I think there is a language barrier
here. Scou_moune What is your natural language?
1938[18:53:00] <RoyK> as root, obviously
1939[18:53:18] <EdePopede> maybe a primer on the topic:
replaced-url
1940[18:53:20] <Scou_moune> yea i'm french but i try to
understand
1941[18:53:55] <Scou_moune> and i'm scary i have missed the
install, but that can't possible
1942[18:53:57] <RoyK> Scou_moune: all programs have open files,
streams etc. Some open a *lot* of them. You just don't see it.
1943[18:54:15] <Scou_moune> ok
1944[18:54:27] <greycat> For the record, on my system right now,
if I do "lsof -u root | wc -l" I get 1328 open files
*JUST* as root. 33 of them are from lsof itself (verified by sudo
lsof -u root | grep -c ^lsof).
2067[19:32:31] <c-c> - Now, I got tired of figuring it out, and
just selected autopartitioning from the same GUI installer from the
same Live Buster 10.4 image, and that just does its thing like it is
supposed to.
2070[19:33:39] <c-c> I've been through debian installs few
dozen times (I was working on a jessie lite-desktop installer so
many many installs), and can figure this out. But I wonder if a GUI
installer should give some error message when it dies.
2075[19:35:09] <rgwu> Which is more useful, catching an exception
and printing an error message, or blowing up and creating a core
dump for a programmer to analyze?
2082[19:37:12] <c-c> I guess I should have hunted for the core
dump, now that I think of that
2083[19:38:25] <shtrb> Could some one remind me what is the
proper way to declare mode switch ? I created a file in
/etc/udev/rules.d/file.conf with
"ATTR{idVendor}=="XXXX",
ATTRS{idProduct}=="YYYY", RUN+="usb_modeswitch
'%b/%k'" but the /dev/sr0 is stil present
2084[19:39:44] *** Quits: tmroland (~tmroland@replaced-ip) (Remote host closed the connection)
2089[19:45:14] <c-c> shtrb: my modeswitch-fu is basically
nonexistent, so I'm bit confused by your details, but can you,
at your own risk of listening to my sillyness $ eject /dev/sr0
2090[19:45:33] *** Quits: dvs (~hibbard@replaced-ip) (Remote host closed the connection)
2111[19:57:20] <shtrb> I have a Marvell(?) usb modem , normally
when I just plug such a device it will switch blindly to the
ethernet mode and I surf . but here what ever I try that /dev/sr0 is
still there (and SPAMMING me about blk_update_request: I/O error,
dev sr0, sector 0 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
until I do eject /dev/sr0)
2136[20:13:12] <c-c> shtrb: I think my best input was already
shared, maybe someone else can help: I have problems with unstable
system so its bit diff to help
2150[20:24:31] <maniqui> Since yesterday (and likely after dpkg
got upgraded to v 1.20.4 on my Debian Sid, system), I noticed that
some plain text files (related to apt/dpkg processes) are getting
badly corrupted.
2151[20:24:33] <RoyK> maniqui: probably this one
2152[20:24:40] <RoyK> !debian next
2153[20:24:46] <RoyK> hm
2154[20:25:14] <sney> !#debian-next
2155[20:25:14] <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
2163[20:27:44] <sney> it's "when it breaks, you get to
keep all the pieces." It doesn't always break. I prefer
the mild filter of testing though, personally.
2170[20:29:44] <maniqui> 1. dpkg/apt seems to be somehow involved
with the files getting corrupted.
2171[20:29:57] <lupulo> in gentoo the op bans after one person
breaks the rules
2172[20:30:03] *** spinningCat is now known as spinningCat-away
2173[20:30:16] <sney> if you think it's a hardware problem,
try to rule that out before getting into the weeds with software.
many disk manufacturers offer diagnostic tools, and there's
smartctl and various bootable vendor-agnostic tools as well
2174[20:31:02] <maniqui> 2. the new contents for a corrupted file
seem to be coming from the same source: a very, very large
hosts.block file (managed by the hostsblock script, an
ad/malware/xxx sites blocker). And the corrupted file gets converted
from plain text into binary.
2176[20:32:00] *** Quits: voidSurfr (~todd_dsm@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
2177[20:32:39] <maniqui> sney yes, I''ve it pending to
do some further SSD testing. At least, smartctl shows an attribute
in red ("Media Wear Out Indicator").
2178[20:33:24] <maniqui> Can't join #debian-next, it seems
one needs to be invited there.
2236[21:20:53] <lifostack> hi all i'm trying to update my
nvidia drivers and it's warning me that i have an x server
running and that it MUST be closed before updating, is this as
simple as CTRL+ALT+F2 or do i actually need to kill the process?
2246[21:22:42] <lifostack> it may not be a quadro, i thought it
was though
2247[21:22:52] <lifostack> most certainly is legacy though, this
Dell is from like 2010
2248[21:23:02] <sney> the nvs series was branded as quadro iirc.
are you using the debian package for the driver?
2249[21:25:22] <lifostack> well i didnt know there was a debian
package, i just downloaded a .run file from Nvidia, which i found
out was an incompatible version so it threw an error telling me to
go download the legacy .run file and warned me about the x server
and that it must be closed to update
2250[21:25:39] <sney> !why nvidia installer sucks
2251[21:25:39] <dpkg> nvidia-installer works fine for you today,
granted. Let's say tomorrow Xorg is updated -- the nvidia
driver will break. If you try to uninstall nvidia (nvidia-installer
--uninstall), then you break Xorg badly. nvidia-installer overwrites
files at random, and has NO CONCEPT of package management. We
recommend against using it based on extensive experience. USE THE
DEBIAN PACKAGES. Ask me about <nvidia>.
2252[21:25:52] <lifostack> only reason im even bothering with the
update is because when i run a simple hello world app in android
studio with the AVD virtual emulator im getting a soft freeze
2263[21:27:36] <lifostack> i get no response or update from the
screen, no mouse movement etc
2264[21:28:17] <sney> the free nouveau driver supports most
features of most nvidia gpus but it does fall short in some
situations. installing the proprietary driver will probably fix your
issue
2265[21:28:19] <lifostack> only when i compile the program and in
turn launch AVD emulator and the emulator settings wont let me
change the Emulator GFX settings from Auto to CPU only so as to
disable GPU rendering of the emulator
2266[21:29:31] <lifostack> sney, any way i can check which nvidia
drivers are currently installed on my system?
2267[21:29:41] <lifostack> sorry if that sounds ultra noob
2268[21:29:41] *** Quits: Grldfrdom (uid391113@replaced-ip) (Quit: Connection closed for inactivity)
2271[21:30:22] <sney> lspci -k would show you a list of devices
and what drivers they are using, yes. or lsmod. you most likely have
only the nouveau driver, which comes with debian
2272[21:30:54] <sney> and if you install the nvidia legacy driver
as per the instructions on that wiki page, it will automatically
clean up any messes left by the nvidia .run file and also disable
nouveau for you.
2273[21:32:22] <lifostack> ok i'm almost finished reading
that wiki page
2357[22:36:23] <tharkun> Good $DAY, be safe everyone. I need to
install a wifi printer which is an hp2545 Someone earlier today
suggested installing hplip which I did. I am unable to still access
the printer to configure the wifi and set up the wifi-password on it
so it is available for other devices. Any suggestions. I ran out of
options on my google skills.
2362[22:39:40] <sney> tharkun: it sounds like you have a few
things going on there. the wifi on the printer will need to be
enabled before you can connect to it from your computer
2363[22:40:13] <sney> it's usually pretty easy to find the
service manual for a given hp model, if you're locked out of
the control panel or something
2396[23:12:33] <jhutchins> tharkun: iirc you often have to
connect with a cable, then browse to a pre-set IP. If it's pure
wifi, then you need to connect to it's preconfigured network. I
would be surprised if you could not find setup instructions on the
web, specifically at the HP support site.