106[00:41:06] <hatter_> I have noticed in dmesg and on the
console, hard disk errors like this show up : ata2.00: exception
Emask 0x0 SAct 0x4000 SErr 0x0 action 0x0
107[00:41:17] *** Quits: dmtucker (~dmtucker@replaced-ip) (Remote host closed the connection)
108[00:41:19] <hatter_> How do I get these messages to log to
syslog ?
131[00:53:51] <crestfallen> hi would using ' sudo apt-get
install snapd ' be a valid and safe way to install bitcoin core
app .. on debian 9 64bit? thanks
132[00:54:14] <crestfallen> snap .. I've never used it to
install anything
143[01:00:56] <dvs> doughy, by editing your
/etc/network/interfaces file to include a "iface xxxx inet
static" clause in it, followed by an address and netmake line
144[01:01:04] <doughy> yeah, I did that
145[01:01:31] <doughy> I tried doing it through network manager
too on the desktop
201[01:38:33] <m0rd3cai> Has anyone had any issues with Xwindows
display and kernel 4.9.0.9-amd64? When i boot it will show the
windows but no buttons to close windows nor does anything respond
except opening a terminal. I can't even type there. Had to boot
4.9.0.8 just to get the machine back up.
235[01:54:01] <joepublic> quoting eric raymond at the ready. not
bad.
236[01:54:38] <annadane> debian is a good operating system for
anybody for many purposes, not just "hacking". so, sure,
install debian because it's awesome, not too sure about the
hacking bit
237[01:55:06] <annadane> a chat room can't teach you
243[01:57:29] <klys> zazagx, but you want to learn to code
hacks?
244[01:58:08] *** Quits: behanw (uid110099@replaced-ip) (Quit: Connection closed for inactivity)
245[01:58:40] <joepublic> I was inferring "zazagx wants to
know what it takes to be an elite hacker" no reason specified;
you seem to be inferring "ha! I know why this person wants to
know that. They, themselves, want to learn 'code
hacks'!"
246[01:58:56] <joepublic> interesting.
247[01:59:01] *** debhelper sets mode: +l 1526
248[01:59:09] <LtL> No troll feeding rule is in effect last I
heard.
249[01:59:13] <klys> hacking is an intelligent and useful
activity.
260[02:01:06] *** Quits: de-facto (~de-facto@replaced-ip) (Quit: See you around.)
261[02:01:15] <klys> ^
262[02:01:28] <annadane> people want to learn "how to
hack" without asking themselves the (necessary) corollary
question, "won't this get me into trouble?"
263[02:01:29] <joepublic> crack is "commit a crime, usually
that of unauthorized access"
265[02:01:58] <annadane> if you want to hack and face the
consequences, go right ahead, but that means you're an
industrious sort of person who likely wouldn't poke around IRC
channels asking how to do it
266[02:02:48] <LtL> doughy: i do believe if network-manager is
running it will ignore anything set in /etc/network/interfaces and
the latter is what you want to set static although it may be doable
with NM.
267[02:03:07] <doughy> I read the same thing
268[02:03:12] <dvs> LtL, he's been told that
269[02:03:33] <LtL> dvs: i see, my mistake, thanks.
270[02:03:35] <doughy> something's changed since last
debian install I had
271[02:03:49] <doughy> something in 10
272[02:03:56] <dvs> doughy, run "ip link"
273[02:04:10] <joepublic> I think networkmanager was introduced
with jessie?
274[02:04:26] <doughy> I don't remember code names just the
numbers
321[02:26:02] <dvs> joepublic, wait until the next release,
you'll be on target every time!
322[02:26:41] <klys> this afternoon I installed a 4-port
rtl8111h, and it gave me a firmware error, so I had to install the
latest firmware-realtek from non-free.
323[02:26:49] <doughy> ok so I got that to work. I thought
I'd let you know in case someone else has this problem
350[02:44:12] <ZaZaGX> i gotta admit. The installer has more
options and it seems to take longer to install in Debian compared to
Ubuntu
351[02:44:48] <jim> coffeecow, debian installing is pretty
easy... does the machine have a wireless card?
352[02:44:49] <coffeecow> I'm not sure if I'm being
overly paranoid tho... not totally sold on moving over to debian. I
haven't used it in years. I've used BSD and Linux for like
10 years.
353[02:45:00] <coffeecow> Yes it's a thinkpad, a recent
one; x390.
354[02:45:19] <jim> coffeecow, do you have some free disk space?
355[02:45:20] <ZaZaGX> what kind of wireless card?
358[02:46:16] <jim> coffeecow, if you're runnng linux now,
could you run... lspci -nn | grep -i net # and look at the output
359[02:46:51] <jim> coffeecow, you're looking for the
pciid, which looks like this: [1234:abcd]
360[02:47:48] <jim> how many lines do you get from that?
361[02:47:49] *** js is now known as js__
362[02:48:11] <ZaZaGX> well, if you want, you can download the
firmware with Debian. so it'll be like an Ubuntu install.
it'll just work with the wifi card
441[03:15:18] <doughy> I'm putting debian on two different
PCs and one has ens4 and eno1, the ohter is like what you're
getting
442[03:16:58] <jim> doughy, if the debian installer can see the
cards, it will write the /etc/network/interfaces file for you, no
need to bother with it, I guess
505[03:53:51] <ZaZaGX> if i want to use debian testing. how do i
do it? should i just upgrade from command line or download the iso?
i can't find the link to download the iso
506[03:55:03] <dvs> !iyhtays ZaZaGX
507[03:55:03] <dpkg> ZaZaGX: If You Have To Ask, You
Shouldn't.
559[04:37:18] <rant> alex89: you don't have a problem with
MTP, MTP has a problem with MTP.. its trash..
560[04:37:24] *** Quits: dtux (~dmtucker@replaced-ip) (Remote host closed the connection)
561[04:37:41] <rant> alex89: for small transfers I put the phone
in photos mode but as a general rule ADB is the way to go
562[04:38:18] <rant> in photos mode you'll often get an
error dialog about unable to start the camera.. but the files will
be visible and it'll mostly work
563[04:38:26] <alex89> so i need to do it in command line ?
564[04:38:34] * dvs just transfers the files because I'm
reckless!
565[04:38:41] <rant> you can, but not necessarily..
566[04:38:49] <rant> alex89: what is your use case.. what sort
of thing are you typically doing?
567[04:39:05] <rant> I just wrote my father a little script to
pull stuff off his device
568[04:39:19] <alex89> i just need to do a backup of the files
in my phone (OnePlus6T) to my computer (Debian 10 Buster)
569[04:39:36] <rant> ah, well for a large grab and adb pull is
best, yes
570[04:39:53] <rant> if you do it often and want to simplify it,
its not difficult to make a helper script to suit your needs
571[04:40:11] <rant> adb is not only faster but its more
reliable and less error prone
576[04:41:14] <KNERD> With the typical open source project, you
extract the project files, and for the most part just do ->
configure make & make install. How could I just make the
excutable binary?
577[04:41:28] <rant> you'll need to go into your settings
on android, to the About Phone section, and click like 9 times on
the build number to enable developer mode and then you'll have
a developer options in the settings where you can enable ADB then
you install adb with apt
578[04:42:00] <rant> then you'll need to authorize the
computer's rsa id to access the phone over adb.. when you first
try to connect
579[04:42:09] <rant> a dialog will appear on the android
581[04:42:37] <alex89> I have done all of that but i get stuck
in what command i need to type ...
582[04:42:41] <rant> KNERD: thats what the make step does
583[04:42:59] <rant> alex89: do you know the path of the dir you
want to pull? sounds like you want all of /sdcard
584[04:43:14] <rant> alex89: which would be "adb pull
/sdcard" which will pull all to the current directory
585[04:43:41] <rant> alex89: if you dont know the path you can
do adb shell and look around on the device to figure it out run df
or such see the mount points and use ls
586[04:44:23] <KNERD> rant: make , makes object files, not
executables
587[04:44:52] <dvs> ???
588[04:45:19] <KNERD> rant: i have seen examples of people
running gcc on those files, but I am seeing multiple ones
589[04:45:32] <rant> KNERD: it compiles whatever there is
typically if not sounds like the makefiles arent correct.. but there
is also something called checkinstall which you can use to wrap the
process and make a deb package out of it
590[04:45:38] <alex89> it work thanks, i just have a problem
with a folder named A.0 Home it says adb: error: failed to stat
remote object '/storage/emulated/0/A.0': No such file or
directory
591[04:46:57] <rant> alex89: idk, try adb shell, do some ls -l
/storage/emulated/0/A.0 or file /storage/emulated/0/A.0 for some
clues
593[04:47:15] <rant> could be its a dangling symlink, not a
regular file, or perhaps a permission issue
594[04:48:21] <jim> ZaZaGX, maybe you could make a copy of your
stable debian and upgrade the copy
595[04:49:49] <ZaZaGX> like you mean install it on a different
partition?
596[04:49:57] <rant> KNERD: typically the configure step
configures the source/makefiles, make builds and links everything
producing not only objects but executables, and make install moves
everything to the proper locations on the system
597[04:50:35] <alex89> Problem solved, i just rename it to Home
much more simplier. Thanks for your help
601[04:51:38] <rant> no problem.. android is annoying.. the
removal of USB Mass Storage mode and the whole MTP and media scanner
daemons and shit are a hot mess
602[04:51:48] <rant> ADB is the closest thing to a sane way to
transfer data
603[04:53:48] *** Quits: argusbr (~tls@replaced-ip) (Remote host closed the connection)
604[04:53:53] <rant> there are a lot of adb filemanagers out
there but they've all become abandonware.. idk why someone
hasnt continued to develop such things
612[04:57:25] <rant> I haven't really tested them
extensively I just use whatever works.. but I dont really trust
android as far as I can throw it.. I currently just have a Galaxy S7
I just use as a mobile pc to save power to do simple things like
reading ebooks, listening to music, watching shows, etc.. I have no
sim in it, keep it in airplane mode with all the radios off :P
618[04:59:15] <rant> there is #android and the folks in #h3droid
are very nice and knowledgeable folk.. they develop an android
version for the allwinner h3 found on many of the orange pi boards..
if you were to get one and help them test I'm sure they'd
love you for it.. and they are very helpful in all things android
619[05:00:04] *** Quits: alexandros_tab (~alexandro@replaced-ip) (Remote host closed the connection)
633[05:10:28] <KNERD> forget spyingware Android. Something new
and open sourece is coming out at the end of the year. The Pine
Phone with different Linux based OSes you will be easilly install
like any Raspberry Pi
663[05:43:34] <jim> Mr_Queue, by not having that file/
664[05:43:35] <jim> ?
665[05:43:48] <rant> KNERD: I haven't heard of it, but I
won't hold my breath.. Saygus, Librem, etc have been promising
open mobile platforms for years.. I've yet to see it
684[05:59:02] <rant> KNERD: you can always follow up on
#debian-offtopic as this is not really on topic here.. but the issue
with most these efforts seems to be govt regulations and such
773[06:52:51] <dfcnvt> It was never like that long before I
remember using terminal many time. Untils this recent month ago. I
didn't mind or pay any attention until I starts to realize it.
860[07:49:28] <friendlyGoat> god im sorry, i wrote down what to
do if another power outage happened cause i didnt want to bother
anybody but holy fuck my electricity went out again. im trying to
fix my DNS resolving thing AGAIN and nothing in my note i made is
working to fix it
861[07:50:15] <friendlyGoat> i deleted resolv.conf, recreated it
with nameserver 127.0.0.1 in it and its just, not resolving at all
862[07:50:24] <friendlyGoat> gonna try fixing things on my own
but godd what a bother
863[07:50:47] *** Quits: Prints (~333@replaced-ip) (Excess Flood)
895[08:00:07] <alkisg> friendlyGoat: you should probably use a
proper solution, like ayekat suggests
896[08:00:15] <friendlyGoat> also i have no idea why, its just
that whenever the power goes out it fucks up my resolving.
897[08:00:20] <friendlyGoat> and yeah i should look into it.
898[08:00:27] <alkisg> You have network manager AND resolvconf
and possibly systemd-resolved, and you manually edit resolv.conf? :)
899[08:00:29] <ayekat> diogenes_: then I'd be *very*
interested in seeing their setup, because with resolvconf and NM
running, I can't see how resolv.conf will remain untouched
901[08:01:24] <diogenes_> ayekat, ask friendlyGoat to give some
outputs.
902[08:01:58] <ayekat> friendlyGoat: would you mind posting your
personal "note" that you've made on this issue?
903[08:04:03] <ayekat> (note that I know neither resolvconf nor
NM very well, but fiddling around with resolv.conf manually is
rarely a clean solution)
936[08:10:33] <friendlyGoat> thanks! also i have resolvconf for
pihole
937[08:11:12] <alkisg> When resolvconf is used, resolv.conf is a
symlink: /etc/resolv.conf -> ../run/resolvconf/resolv.conf
938[08:11:27] <alkisg> So that resolvconf picks up the dns from
network manager, and updates it properly
939[08:12:05] <alkisg> Nowadays, systemd-resolved is used
instead, as a superset of resolvconf, and the file is a symlink to:
/etc/resolv.conf -> ../run/systemd/resolve/stub-resolv.conf
950[08:14:53] <diogenes_> but no major distros do.
951[08:15:50] <ayekat> well, users are free to configure their
setups however they like (and I chose to replace ifupdown with
networkd) :-)
952[08:16:15] <ayekat> but anyway - there shouldn't be any
reason to manually touch /etc/resolv.conf
953[08:17:40] <diogenes_> that doesn't harm and fixed his
issue.
954[08:17:52] <alkisg> diogenes_: by default, debian
doesn't use resolvconf nor systemd-networkd. But when one needs
resolvconf to manage DNS on a per-link basis, why wouldn't he
use systemd-networkd instead, since he's going to select one of
them?
956[08:18:17] <alkisg> If he has hardcoded DNS, it cancels
resolvconf, it's like he didn't install it in the first
place
957[08:18:25] <ayekat> diogenes_: it only "fixes" it
for now, but who knows when it will break again, and I think it
makes sense to understand why such a hacky workaround is necessary
in the first place
958[08:18:50] <diogenes_> alkisg, that's what i meant, by
default no one uses it, only if someone wants to tinker something
specifically, he is free to choose whatever he wants.
959[08:20:45] <diogenes_> ayekat, i gave his a quick solution to
fix the issue right now and i never claimed it was the ultimate
solution and besides, i was waiting for other more advaced
specialist than i to step in and try to resolv the core of the issue
and since nobody bothered to troubleshoot the core of the problem,
it is what it is so, if you can fix his core issue then go ahead.
960[08:21:10] <alkisg> diogenes_: what I'm saying is, the
method you proposed, invalidates his setup. He no longer uses
resolvconf anymore, since he removed the resolvconf symlink
961[08:21:25] <alkisg> So he *thinks* he's using
resolvconf, but he isn't
962[08:21:28] <diogenes_> alkisg, yeah read what i just said.
963[08:22:08] <diogenes_> go ahead and fix his core problem
instead of commenting on my solution.
964[08:22:24] <alkisg> Sure; hopefully friendlyGoat will
understand that he still has an issue to resolve (no pun intended);
I see he lost interest in this :D
965[08:22:30] <diogenes_> friendlyGoat, are you still there?
971[08:23:32] * diogenes_ waits for the process to take on
972[08:23:53] <ayekat> diogenes_: I was merely commenting
because the discussion kinda went into "oh, it fixed it,
thanks!" and there was no followup whatsoever à la
"warning! this is only a workaround, and this should definitely
be fixed"
975[08:24:47] <diogenes_> ayekat, and your comments are
irrelevant when they are addressed to me because i did what i could,
your questions should be addressed towards the user with the issue.
976[08:24:48] <ayekat> also again, I can't help debug this,
because I use neither NM nor resolvconf
977[08:24:59] <diogenes_> and no warning there since it
doesn't break anything.
978[08:25:22] <friendlyGoat> alright so i have to make a
symlink? /etc/resolv.conf ->
../run/systemd/resolve/stub-resolv.conf yeah?
979[08:25:29] <alkisg> friendlyGoat: no
980[08:25:45] <alkisg> If you want to properly resolv this,
please explain first why and how you're using resolvconf
981[08:25:59] <alkisg> You have a pihole somewhere in your local
network?
982[08:26:05] <alkisg> Or you have it installed in this host?
983[08:26:16] <friendlyGoat> pihole is on this host
984[08:26:47] <ayekat> diogenes_: but it leaves the user in a
situation where they wrongfully believe it's fixed "until
the next blackout", so pointing that out explicitly seems
necessary to me
985[08:26:49] <alkisg> Is this a raspberry pi with raspbian?
986[08:27:12] <friendlyGoat> nope this is debian on an older
machine of mine
987[08:27:15] <ayekat> (but I don't want to argue, we are
not contributing to solving, so I'm gonna stop)
988[08:27:30] <alkisg> Debian 10?
989[08:27:39] <diogenes_> ayekat, no buts there, you have no
right to say "but" unless you have a better solution, you
don't understand what i'm saying.
992[08:28:47] <alkisg> I don't know pihole. I'm
guessing that it's using resolvconf to manage DNS on a per link
basis.
993[08:28:58] <alkisg> Does this file contain your dns servers?
cat /run/resolvconf/resolv.conf
994[08:29:07] <friendlyGoat> yep, resolvconf if a requirement.
also let me check real quick
995[08:29:09] <ayekat> diogenes_: no, the problem is on
different layers - it's fine to help out and give a solution,
even if it's not permanent - but it's necessary to point
out that it's a temporary solution
1001[08:30:00] <friendlyGoat> yep that file exists but theres
nothing in it
1002[08:30:21] <friendlyGoat> (BRB)
1003[08:30:30] <alkisg> friendlyGoat: you *might* need to first
create the symlink and then restart resolvconf/network-manager for
this file to be populated,
1004[08:30:46] <diogenes_> ayekat, it is relevant because you
don't know the problem, you don't know how to troubleshoot
it, you don't know nm, you don't know resolconf and as a
result you can't even judge whether it's a permanent
solution or not.
1005[08:30:53] <ayekat> diogenes_: I just stepped in because
"edit resolv.conf" - "it's fixed!" - (no
follow-up), so I made a comment pointing out the issue
1006[08:31:07] <alkisg> which would be: sudo ln -sf
../run/resolvconf/resolv.conf /etc/resolv.conf && ...best
reboot, to be really sure it works or not,
1007[08:31:28] <ayekat> diogenes_: I don't need to know NM
or resolvconf to judge that manually modifying /etc/resolv.conf is
not a permanent solution
1008[08:31:32] <alkisg> friendlyGoat: but if this command
doesn't fix your dns, then pihole does more tricky things, in
which case you'd ask in #pihole instead of #debian
1009[08:31:46] <diogenes_> ayekat, and that's more relevant
than yor comment because the user has internet right now and if it
was stuck with ou helping him, he wouldn't have any positive
reult bby niw.
1015[08:33:04] <diogenes_> and i know that in a normal setup you
don't edit resol.conf but i told you, in this particular case
it's ok to od it.
1016[08:33:09] <ayekat> diogenes_: it's permanent *by
accident*, doesn't mean it's fixed - and again, I'm
not criticising the help - I was just pointing a flaw based on my
understanding, and now I'm suddenly getting all the flack
1017[08:33:21] <alkisg> diogenes_: ayekat is right; helping users
is something to be proud of; all of us here trying to help others
are good people trying to do the right thing; but, notifying that
"this solution isn't proper, it's just a
workaround" is also necessary, and users may choose to not do
anything more anyway
1021[08:35:30] <diogenes_> ayekat, alkisg hehe you're trying
to look politically correct now or what? so the issue is that i
didn't mention it's a temporary solution? show me the
regulations that i have to say that or mention anything or something
like that and prove it's wrong, and also i'm still waiting
for youboth to fix his issue once and forever.
1024[08:36:28] <alkisg> diogenes_: nah, I'm trying to
prevent 2 good people from fighting regarding "how to best help
others"; it's silly to hurt someone's feelings when
he's trying to help
1025[08:36:43] <alkisg> I know the feeling, that's way I
wrote that previous sentence, not to be politically correct
1026[08:37:17] <alkisg> *why
1027[08:37:51] <diogenes_> and did i argue about anything in the
first place, i gave the advice i knew and i was minding my own
business until someone tried to tell me i'm wrong because bla
bla bla without either corecting me or giving an elternative
solution.
1037[08:40:11] <ayekat> diogenes_: I just reread the
conversation, and nowhere did I point out you were wrong
1038[08:40:35] <ayekat> I just noted that it's not a
permanent solution, because that wasn't mentioned anywhere
1039[08:41:16] <diogenes_> ayekat, the only thing i'm
obliged to do, is not to harm the user that's all, all other
stuff like mentioning this or that is not my business.
1040[08:41:57] <ayekat> diogenes_: that's correct, but
getting all defensive when someone points out a flaw is not really
helping anyone either
1044[08:43:28] <diogenes_> ayekat, this "flow' is not a
flaw but an opinion which is not even a technical one and opinions
are different from person to person and you was tryung to say that
your opinion is the right one and i disagree because i have mine.
1045[08:43:46] <ayekat> what
1046[08:43:51] <ayekat> this is silly now
1047[08:43:54] <ayekat> have a nice day
1048[08:44:00] <diogenes_> you too
1049[08:44:04] *** Quits: fl3x1t (~fl3x1t@replaced-ip) (Remote host closed the connection)
1054[08:50:44] <alkisg> friendlyGoat: note that this is a
suggestion on how it would solve the problem with resolvconf. If
pihole is doing more weird stuff, it might again break your dns, in
which case you'll need to hardcode it again etc like diogenes_
said, and ask help from #pihole,
1055[08:50:59] <alkisg> so, the command is: sudo ln -sf
../run/resolvconf/resolv.conf /etc/resolv.conf
1063[08:53:33] <finn0> In a fresh installation of Debian which is
file provided by default? (~/.profile or ~/.bash_profile). AFAIR,
~/.profile is present in the home directory why is so?
1064[08:54:25] <klys> one style for user and the other for root
1065[08:54:38] <alkisg> finn0: many packages may put things in
/etc/skel; those are copied to home dirs when users are created
1066[08:54:47] <alkisg> So you can just ls -lha /etc/skel
1070[09:00:30] <alkisg> friendlyGoat: nothing, does dns work?
1071[09:00:30] <finn0> okay, Debian provides ~/.profile. Since,
Debian default shell is bash and bash looks for file in this
/etc/profile -> ~/.bash_profile, and if finds it, it read that.
If not, look for ~/.bash_login, and if it doesn't find that,
look inot ~/.profile otherwise stop searching.
1072[09:00:45] <alkisg> friendlyGoat: does `cat
/run/resolvconf/resolv.conf` have things?
1078[09:02:29] <finn0> If ~/.bash_profile is found, bash stops
looking for files. Am I correct? And why does Debian provide
~/.profile instead ~/.bash_profile? Is there any specific reason?
1079[09:02:59] <ayekat> finn0: because ~/.profile is sourced by
all kinds of shells, so it makes sense to provide a generic file
rather than a bash-specific one
1080[09:03:28] <ayekat> finn0: also, it reads ~/.bash_profile no
matter what /etc/profile says (but otherwise your observation there
is in line with what bash(1) says)
1084[09:06:39] <ayekat> to be honest, I think bash's
sequence of sourcing files is pretty weird, but I guess we're a
couple of decades too late to change that now ^^
1179[09:57:34] <centrix> I have removed all samba packages plus
all "unnecessary" package apt reports about. The installed
all the samba packages again but got:
1240[10:18:10] <half-beard> ratrace, are you around bud? I
finished installing buster last night. A very simple installation.
Just grub-pc on ext4 in it's own partition, luks on another
partition with ext4 / inside it, debootstrapped. Setup fstab, passwd
tasksel standard etc, linux-image, installed cryptsetup-initramfs I
setup /etc/crypttab ran update-initramfs -u and grub-update etc.
There were no errors. Grub is installed, but it says it cannot find
the UUID of my
1241[10:18:10] <half-beard> ext4 partition. The initramfs or grub
is not using cryptsetup to decrypt my luks.
1242[10:19:32] <half-beard> oh shit. I think I forgot to bind
mount /boot!
1257[10:23:43] <ratrace> make sure you check boot/grub/grub.cfg
if it points at correct uuids for both the device containing /boot
and for rootfs itself
1258[10:23:44] <half-beard> oh, I see what you're saying
1259[10:23:54] <ratrace> half-beard: you can mount /mnt/boot even
from outside the chroot
1260[10:23:59] <half-beard> yes. That's probably the problem
1261[10:24:08] <half-beard> yes, i understand what you meant now.
1262[10:24:10] <ratrace> inside the chroot, you see all the real
mountpoints via /proc/mounts
1263[10:24:41] <half-beard> ratrace, I've seen people doing
fancy switches to update-initramfs. I normally just do -u. But
I've seen people mentioning -k and -v etc
1264[10:24:52] *** Quits: timahvo1 (~rogue@replaced-ip) (Remote host closed the connection)
1265[10:25:15] <half-beard> is there anything special I need to
do now besides mount boot properly and then update-initramfs?
1272[10:26:59] <half-beard> or should it happen automatically by
virtue of the fact that I've got the cryptsetup-initramfs
package installed and /etc/crypttab done?
1274[10:27:16] <half-beard> okay will redo with /boot mounted and
re-check.
1275[10:27:43] <half-beard> /boot was NOT mounted when I did this
last night. So that's probably the only thing that was wrong.
1276[10:28:02] <ratrace> "search" should contain the
uuid of the ext4 containing /boot, and "linux" line should
have root=UUID=<uuid of ext4 root which is insde the luks
container>
1277[10:28:23] <ratrace> then crypttab should have the uuid of
the partition that's the luks container itself, in the second
column
1278[10:28:39] <ratrace> fstab / either by uuid or /dev/mapper/
name
1279[10:28:46] <ratrace> (which is the first column of crypttab)
1280[10:28:55] *** Quits: stefanos82 (~stephen@replaced-ip) (Quit: Quitting for now...)
1286[10:30:26] <ratrace> half-beard: cryptsetup-initramfs should
make sure initramfs contains all the required hooks and scripts to
unlock the container
1299[10:32:46] <centrix> alkisg, I am bit under pressure so I
grabbed /var/lib/samba /run/samba from the working box and replaced
in the failing. Now at least smbd starts without complains.
I'll see what comes next (AD join ...)
1300[10:32:46] <ratrace> yeah... but wait... uh... i think you
actually need to reinstall the versioned linux-image- packages
1309[10:33:52] <ratrace> linux-image-amd64 is a metapackage
1310[10:34:32] <ratrace> oh but careful, if that's within
chroot, that uname -r might be wrong, so just make sure it's
linux-image-4.19.0-5-amd64 for buster
1311[10:35:26] *** Quits: MrShared (~mrshared@replaced-ip) (Remote host closed the connection)
1312[10:35:31] <half-beard> ok, I normally just see what's
there with `dpkg -l | grep linux-image` then remove everything
listed
1322[10:40:40] *** Quits: BlueByte (~walther@replaced-ip) (Quit: This computer has gone to sleep)
1323[10:40:53] <half-beard> ratrace, I've seen people doing
fancy switches to update-initramfs. I normally just do -u. But
I've seen people mentioning -k and -v etc
1329[10:42:26] <ratrace> half-beard: -u suffices ; -k is if you
want specific or all kernels' initramfs-es updated
1330[10:42:44] <half-beard> okay
1331[10:42:46] <half-beard> thanks
1332[10:43:09] <half-beard> ratrace, I read this last night, is
it necessary? To force the cryptsetup modules to the initramfs image
you have to set CRYPTSETUP=y in /etc/cryptsetup-initramfs/conf-hook.
1333[10:43:31] <half-beard> (I've not done that)
1334[10:44:14] <ratrace> half-beard: read that file :)
1335[10:44:46] <ratrace> half-beard: note that in
initramfs-tools, hooks are functions exec'd when you construct
the initramfs, and scripts are execed when initramfs is run
1379[10:57:08] <half-beard> haha, ratrace: I just remembered to
check the UUIDs in /boot/grub/grub.cfg I've got search
--fs-uuid blah blah then the UUID of the /boot partition
1427[11:14:18] <half-beard> I forgot to create a user account and
now need to setup networking etc but just connected to IRC from my
phone now to say thanks!
1428[11:14:37] *** Quits: MrShared (~mrshared@replaced-ip) (Remote host closed the connection)
1429[11:14:51] <half-beard> When I get back to the PC I'll
sort I'll finish the rest of the setup.
1462[11:22:50] <ayekat> finn0: well, users expect bash to behave
in a certain way (in particular: to source a certain set of
configuration files in a certain order)
1463[11:23:30] <ayekat> finn0: if bash suddenly started sourcing
different things in a different order, those expectations would no
longer be valid
1468[11:24:24] <half-beard> Ratrace, you said you run ZFS for
root. Would you do this for your laptop? How do you know the next
kernel update won't break ZFS?
1478[11:26:38] <half-beard> Haha. But then if you've got
urgent work to do you have to plug in your USB boot drive that you
hopefully remembered to bring with you on the road. Boot that and
then proceed to do unexpected maintenance.
1481[11:27:44] <ratrace> half-beard: on the servers where we use
zfs root, we have scripts in initramfs, due to luks unlocking over
ssh, those scripts can easily revert the snapshot if something was
broken
1550[12:17:39] *** Quits: MrShared (~mrshared@replaced-ip) (Remote host closed the connection)
1551[12:17:50] <ZaZaGX> sounds like a debian program
1552[12:17:57] <half-beard> Ratrace, that's interesting,
what kind of scripts? Bash? What's the limit of what can be run
in initramfs? What determines the limits?
1553[12:18:21] <half-beard> Ratrace, is it possible to use
dropbear unlock with WiFi or Bluetooth?
1554[12:18:23] *** Quits: ohwowlol (uid375208@replaced-ip) (Quit: Connection closed for inactivity)
1555[12:18:54] *** Quits: cfoch (uid153227@replaced-ip) (Quit: Connection closed for inactivity)
1556[12:19:12] *** Quits: bolt_ (~r00t@replaced-ip) (Remote host closed the connection)
1561[12:22:20] <ratrace> half-beard: sure, i don't know if
there are wpa_supplicant hooks by default, but one can always write
one, it's just shell scripts
1565[12:25:17] <ratrace> half-beard: initramfs is literally just
an "/init" script in the ramfs, that the kernel executes ;
that scripts loads others that autodetect environment with the sole
purpose to mount the real rootfs and pivots to it
1567[12:25:59] <ratrace> writing your own is a piece of cake too,
if you understand how it all works:
replaced-url
1568[12:26:35] <ratrace> and with initramfs-tools it's even
easier as it provides a framework so you just have to write much
simpler hooks and scripts, not the entire logic of it
1715[13:48:51] *** Quits: killall (~killall@replaced-ip) (Remote host closed the connection)
1716[13:49:44] <centrix> I update permissions for a directory to
a user u:user1:rwx /dir and file like u:user1:rw- /dir/* ; Still he
gets "permission denied". The file content gets updated
though. :) What is wrong. No haul in journalctl
1738[13:57:56] <afernandez> Hello at this time I am using a
debian server, is there a reason or how can i fix why the following
command works from command line but not within a script? rm -- *.zip
1856[14:31:18] <centrix> Originally was the file created by root
and retains mask of 644. I added myself to the ACL of the file,
tested from unx, checked I get the SAME permissions as him ... no
problem.
1857[14:31:23] *** Joins: conta (~Thunderbi@replaced-ip)
1858[14:31:48] <alkisg> AinNero: at some point you have kernel +
compressed initrd + uncompressed initrd, so it needs quite a bit of
ram. Later on only the RAM usage is lower.
1859[14:32:38] <centrix> The filesystem is ext4, the user
connects using Winscp and authenticates/authorizes using AD
(winbind). He is in the ACL of the file... Let me do another test
...
1868[14:34:08] <alkisg> AinNero: kvm -m 192 -kernel linux -initrd
initrd.gz -append 'rdinit=/bin/sh' ==> free ==>
around 90M used after things settle down
1884[14:40:28] <centrix> well_laid_lawn, we can live with that. A
few people will be allowed to edit the file. It makes no sense to
change ownership then
1885[14:40:39] <ws2k3> so i have a ubuntu 14.04 machine which is
on 100 %[working] (apt-get update). and yes i know i should fck off
to #ubuntu. but ubuntu sucks. debian ftw. and the first thing the
ubuntu guys sad was. ubuntu 14.04 is eol. well. its based on jessie.
still supported. anyone an idea what i can/should debug this?
proberly its one repo. but the things is. the repo works fine in the
browser.
1899[14:45:55] <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
1900[14:46:01] <ooAoo> you got debian 10 buster installed?
1901[14:46:12] <ooAoo> heard it is just release recently
1902[14:46:18] <ws2k3> nope ubuntu 14.04 is based on jessie
1903[14:46:21] <greycat> about 4 weeks ago
1904[14:46:34] <ooAoo> greycat: any opinion on this new debian?
1905[14:47:00] *** Quits: fl3x1t (~fl3x1t@replaced-ip) (Remote host closed the connection)
1907[14:47:21] <klys> ws2k3, check out the debootstrap utility,
it may be just what you need
1908[14:48:12] <greycat> It really depends on which
desktop/applications you use. The switch to GNOME/Wayland instead of
GNOME/X11 may be a huge shift for some users, but it does not affect
me (fvwm/X11). The change to su just seems spiteful.
1909[14:48:22] <NetTerminalGene> guys, gedit launches 0.5 second
delay, it is same for you?
2064[16:01:22] <karlpinc> CrazyTux: Debian has a _lot_ of
software available. This can make it harder for a newbie to pick and
choose. Often the newbie might have to learn something, like how to
search for the software they want using, say packages.debian.org, or
the debtags tool, or the apt-file tool, or....
2076[16:06:17] <karlpinc> CrazyTux: The synaptic GUI package
manager has a lot of features, but I can't tell you what they
are. I, like many with more experience, tend to avoid the GUI for
lots of reasons people will be happy to go on about. (Probably best
discussed in #debian-offtopic.) FWIW, in this channel we find text
interfaces way easier to support, although of course we'll help
with anything.
2077[16:06:25] <han-solo> or just `aptituide` :)
2078[16:06:52] *** Joins: Prints (~333@replaced-ip)
2079[16:07:03] <karlpinc> Yah. I like aptitude and it pulls in a
lot of the features of the other tools.
2080[16:07:04] <joshuaBPMan1> so I am having a really hard time
creating a complete install DVD or usb stick. I have tried
installing debian on my Macbook 7,1. It has failed at least 10
times. I have had debian installed on it before....The last time I
tried burning a DVD via the debian 64 netinstall iso, the DVD was
spit out about halfway through.
2081[16:07:20] <greycat> It may also be worth mentioning that
synaptic doesn't fully work (can't do root-stuff) under
Wayland.
2082[16:07:27] <CrazyTux> karlpinc: ok
2083[16:07:40] <joshuaBPMan1> I have then tried burning a usb
stick via "dd if=debian-64-netinstall.iso of=/dev/sdb".
2084[16:07:59] <joshuaBPMan1> The usb sticks evenutally end up
telling me a file is corrupted and it cannot install. What am I
doing wrong?
2085[16:08:11] <han-solo> do a sync ?
2086[16:08:13] <karlpinc> greycat: I don't know anything
about wayland. Is it because it's stupid to run GUIs as root?
2087[16:08:22] <greycat> that's part of it, yes
2088[16:08:24] <joshuaBPMan1> han-solo. I am doing that as well.
2089[16:08:59] <han-solo> or `dd bs=4M if=path/to/input.iso
of=/dev/sd? conv=fdatasync status=progress`
2090[16:09:03] <han-solo> is what i usually do
2091[16:09:10] <karlpinc> joshuaBPMan1: Is this dd on the mac? I
hear tell that MacOS does not reallly do dd, it just has something
called dd that does "special things" depending on the data
it's moving around.
2092[16:09:13] <greycat> Long-term, the correct fix for synaptic
would be to restructure it so the GUI portion runs as *you*, and
communicates with an elevated-privilege daemon that does all the
actual root-stuff.
2093[16:09:15] *** Quits: munshine (~Munshine@replaced-ip) (Remote host closed the connection)
2100[16:11:16] <joshuaBPMan1> karlpinc: I am currently running a
mac 4,1. It is running debian.
2101[16:12:28] <joshuaBPMan1> karlpinc: I just tried using
brasero to burn my DVD. It spit the DVD out. It then told me an
"unknown error occured when trying to burn the DVD.
2102[16:12:38] <joshuaBPMan1> han-solo: Why the 4m?
2127[16:16:42] <CQ> hello, I jsut upgraded my stretch to buster,
and now I only get a normal console, not a desktop. I am using xfce
... any suggestions?
2128[16:16:46] <karlpinc> joshuaBPMan1: It is also remotely
possible you have bad ram in your box. memtest86+ is called for to
test that.
2129[16:16:48] <winny> you might do well to try with cdrecord
directly, it'll tell you error messages that will highlight the
problem
2148[16:22:03] <greycat> Then it sounds like a display manager
either isn't installed at all, or is disabled.
2149[16:22:18] <CQ> rebooting again to make sure we're not
chasing ghosts...
2150[16:22:26] <joshuaBPMan1> karlpinc: I guess my main issues
started happening when I swapped my memory and hard drives from my
macbooks. 7,1 and 4,1.
2151[16:22:30] <humbot> oh weirdly i have graphical.target on
this tty only machine
2152[16:22:37] <greycat> humbot: same here. it's fine.
2153[16:22:48] <greycat> if there isn't a display manager
installed, it can't run one.
2189[16:34:40] <hero44322> /opt/cmake/bin/cmake -version shows
the version of cmake I have since I built from source. How do I add
it to bash so I can cmake -version instead for example.
2206[16:38:11] <Bushmills> hero44322: PATH problem. Consider to
use an alias, or symlink relevant bins to /usr/local/bin.
Alternatively, add /opt/cmake/bin in a wrapper script called for
compilation to PATH
2207[16:39:01] *** Quits: enoq (~enoq@replaced-ip) (Remote host closed the connection)
2213[16:40:11] <karlpinc> hero44322: You might consider asking
the bot more regards below:
2214[16:40:13] <karlpinc> !usrlocal
2215[16:40:13] <dpkg> When hand-compiling software, *always*
create a new directory under /usr/local, then install into it. For
example: mkdir -p /usr/local/mypackage; ./configure
--prefix=/usr/local/mypackage. Otherwise, it can overwrite files
which should be protected by apt, which can break apt, or apt can
break this application's files because it doesn't know
about them. See also: <equivs>, <FHS>, </opt>,
<checkinstall>, <stow>, <make your own .deb>
2229[16:45:08] <rant> Bushmills: no, I haven't. I never
heard of it.. I am not having any issues I was just commenting on
someone saying gedit took half a second to load..
2230[16:46:13] <Bushmills> rant, I'm aware if it, my reply
was also meant as tongue-in-cheek - while the editor is real, the
launch saving time may be marginal
2231[16:46:23] *** Quits: karakedi (~eAC53C340@replaced-ip) (Remote host closed the connection)
2232[16:47:21] <Bushmills> especially on a fast box
2234[16:47:38] *** Quits: hero44322 (~hero44322@replaced-ip) (Remote host closed the connection)
2235[16:48:54] <rant> Bushmills: ah.. well, I always appreciate
humor, I just didnt realize it was a joke.. didnt see any
":P" in there
2236[16:49:20] <Bushmills> that'd be giving it away,
woudn't it?
2237[16:49:42] <rant> Bushmills: my current machine is a Thinkpad
T440, 4th gen Core i5 w/ 8GB ram and a SATA SSD, so I have no issues
with speed really
2240[16:50:53] <Bushmills> rather similar here. i7 2nd gen, other
identical
2241[16:50:57] <rant> considering the best machine I had prior to
this was probably a T61 Core2Duo with maybe 3GB ram, and for years
prior to getting this, I was fooling with first an orangepi lite as
my desktop, then an HP T520 thin client.. this seems super fast to
me
2245[16:51:56] *** Quits: ich (~ich@replaced-ip) (Quit: Ex-Chat)
2246[16:52:05] *** Quits: kila (~kila@replaced-ip) (Quit: Leaving)
2247[16:52:13] <Bushmills> replace orangepi against odroid xu4,
and we're on same page again here
2248[16:52:14] <rant> I'm thoroughly impressed with the
modern-ish thinkpads.. the battery life and speed blows me away.. I
can hardly believe how light weight and cool it is, the fan rarely
ever runs, I am rarely over 5W power consumption, I get typically
16hrs or so battery life.. its just hard to imagine
2250[16:52:23] *** Joins: kila (~kila@replaced-ip)
2251[16:53:00] <rant> idk how the hell they managed to get so
much processing power into a device and still keep it cool without
the fan running and with such low power consuption
2256[16:53:29] <Bushmills> maybe by you not demanding all that
processing power all the time
2257[16:53:51] <Bushmills> idling can be power efficient :)
2258[16:54:56] <rant> I used to demonstrate to people I was
planning to switch to linux the ability of it to juggle.. and
I'd just open up every app in the menu.. to show them it doesnt
bog it down.. and I've tried that on this machine.. many tabs
in the browser with JS hungry websites, opening many large photos in
gimp, videos in kdenlive, audio in audacity.. etc.. I can't get
it to sweat at all even trying to exxagerate the
2259[16:55:02] <rant> load beyond what I'd ever ask of it
2265[16:56:17] *** Quits: MrShared (~mrshared@replaced-ip) (Remote host closed the connection)
2266[16:56:22] <rant> yeah the fan will kick on but the ram/cpu
usage I never really get at or above 50% of what it has to offer no
matter how hard I try
2277[16:57:56] <rant> that fan thing and the low power
consumption baffles me more than the raw processing power of the
machine.. I just really dont understand how the entire system is
using less than 5W
2297[17:02:00] <Bushmills> meaning of that what keeps machine
busy os not for sake of keeping it busy, instead i needed some sort
of reference for a specific kind of copro on another device, meant
to do largest part of what that machine was just busy with. I just
want to see how that copro will fare against a pure software
solution
2299[17:02:43] <Bushmills> if it can make my machine sweat, it
can do yours too :)
2300[17:02:45] <rant> I traded some college kid who is a huge
thinkpad fan, was trying to get a collection going.. traded him an
old t42p and a few broken t61 and t500.. and he decked this machine
out.. fitted it with a 1080p screen which wasnt even an option on
the T440, put in the 3 button touchpad..etc
2302[17:03:38] <rant> I tried to tell him the T42p was really old
and not that impressive a machine but he was thrilled by it.. heh..
I felt like I was rippin him off, but he insisted he was happy
2361[17:29:19] *** Quits: timahvo1 (~rogue@replaced-ip) (Remote host closed the connection)
2362[17:29:36] <jim> davide, "working on debian"? what
do you want to do?
2363[17:29:44] <davide> Need I to install both sid and testing in
order to do development job on Debian ? I mean helping in find and
solve bugs for the next release
2364[17:30:24] <rant> davide: you can use a virtual machine or
chroot for these purposes.. install stable
2365[17:30:39] <jim> probably you want testing, and one thing
there will be to test, is the debian installer
2366[17:31:10] <rant> it makes more sense to just install stable
and do the testing on top of a stable platform
2368[17:31:58] <rant> unless you have a lot of spare hardware
laying around and are going to be doing some kind of
testing/development that requires native access
2383[17:35:51] <rant> davide: do you actually have a question?
something you need support with? because I'm not seeing one.
2384[17:36:06] <jim> btw, I don't see how it makes sense to
test for bugs in a branch that's already released... if you
find bugs there, they're less likely to be fixed unless
they're security-oriented, or otherwise serious
2385[17:36:09] <Psyndrome> I installed an HDD formatted it as
ext4 then i followed a bunch of commands to make it usable but now
every time i relogin to the system it is asking for password again
and again
2390[17:37:17] <rant> Psyndrome: you need to be more specific
about the password.. you included misleading details that one could
interpret as disk encryption.. there are many forms of passwords and
logins.. console, display manager, ssh, etc..
2391[17:37:34] <davide> I have already ask my question
2392[17:37:42] <rant> Psyndrome: are you asking how to make the
system boot to a GUI desktop without a password?
2393[17:38:08] <davide> I mean I have already wrote my question,
thanks for you answers
2394[17:38:38] <rant> davide: perhaps you think you did.. and I
suspect you may not be a native english speaker because while you
put a question mark on that, it was not a question, it was a
statement
2398[17:39:39] <rant> I need to... is a statement. one without
any context in this case
2399[17:39:45] *** Quits: jikz (~k047n0de@replaced-ip) (Remote host closed the connection)
2400[17:40:47] <Psyndrome> Im asking about how to mount the new
installed HDD automatically without asking me for password every
time i accessing the content but if you tell me how to access the
gui without it will be useful as well
2401[17:40:48] <jim> rant, you're not a parsing algorithm...
you have human intelligence you can use to help communicate
2403[17:40:58] *** ravioli is now known as spaghetti
2404[17:42:22] <rant> jim: and I'm not being overly anal
here.. there is no context.. even if I insert "do" in
there to make it a grammatically correct question, its one that can
only be speculated at best.. because we don't know if you do
need to do something, if we don't know what it is you're
trying to DO :P
2408[17:44:30] <rant> Psyndrome: if its a normal unencrypted
drive, you need only make an fstab entry for it.. the specifics
depend on your providing more specifics about what it is
(removable/fixed) and what you want to do with it, i.e. where you
want it mounted
2409[17:44:33] <jim> rant, what I do in situations like that, is
I -gently- ask for the information that's missing... also, I
try to make statements that include as much informatiuon that would
otherwise be missing
2414[17:46:15] <rant> jim: and we were doing that. its not a big
deal.. near as I can tell we satisfied the curiousity with assumed
advice.. which as I already stated, proves there was no actual
question there, no actual issue for support.. they were just looking
for general advice on testing and are satisfied with the responses
2416[17:46:53] <Psyndrome> the HDD is normal but i added it after
the installation of the system and it is keep asking for password
but weather or not is encrypted i dont think so
2422[17:48:07] <rant> Psyndrome: sounds like you don't
really know what you are trying to do.. what is it you want to use
this HDD for.. lets start with that? Is it an internal fixed disk,
or is it something you will be removing like a USB drive?
2428[17:49:48] <rant> Psyndrome: if you are unable to answer
these questions than perhaps you can run some commands to provide us
with some information.
2429[17:49:51] <Psyndrome> The HDD is normal and im already using
it as a stash in some sort but for every reloging need a password to
remount itself
2430[17:50:06] <Psyndrome> just by clicking on the device icon
2447[17:56:33] <rant> yes, well that is why I asked repeatedly
about the specifics and even asked to see lsblk and fstab so I can
figure it out for myself definitively :P
2448[17:58:14] <Psyndrome> i didnt encrypted anything cause i
dont know how but if the gparted tool is doing it by itself i dont
know
2449[18:00:03] *** Quits: tommaso (~pi@replaced-ip) (Remote host closed the connection)
2450[18:01:04] *** Quits: noodlepie (~phillip@replaced-ip) (Quit: Going to do something more productive, I think!)
2452[18:01:33] <rant> if you encrypted it, you'd probably
know.. you'd at the very least have had to create a password
for the disk
2453[18:02:16] <rant> the fact of the matter is, after all
you've said, I dont think your issue has anything at all to do
with this hdd.. you are just asking about not having to keep typing
the password into polkit dialogs.. for installing packages, etc..
2454[18:02:58] <rant> and if thats the case, the question is, do
you just want to eliminate the need altogether and just allow your
user to do these things without a password?
2455[18:03:20] <rant> I've configured my father's
machine to just never ask for a password to use the package manager
or anything
2457[18:03:51] <rant> on my system I just use my built-in
fingerprint scanner for these things
2458[18:05:28] *** Quits: coffeeandporn (~coffeeand@replaced-ip) (Remote host closed the connection)
2459[18:05:38] <rant> if you just want to eliminate passwords
altogether for admin tasks, my recommendation is to install and
configure sudo with NOPASSWD: ALL directive for %sudo and add your
user to the sudo group, disable the root account, and then configure
polkit to respect sudo
2465[18:06:50] <jim> rant, what if that wasn't the case? in
other words, what if he had a drive that he didn't know had
encrypted content?
2466[18:07:59] <Psyndrome> When i insert a flash drive for
example its mounted on the same place as the hdd with the only
difference when i click on the hdd icon needs a password because its
not mounted but just visible from the icon
2467[18:08:05] <rant> disabling the root account is optional but
I recommend it because if a user doesnt regularly use the rootpw
anymore.. they are unlikely to choose a good pw and update it
frequently.. which could introduce a security issue at some point if
a service is isntalled that allows root logins.. where if root is
disabled, it cannot be accessed remotely via a login.. can only be
accessed via sudo which requires a
2468[18:08:11] <rant> login to a sudo user first.. which
you'd be maintaing a better pw policy on
2469[18:08:13] <Psyndrome> When i insert the password it is
mounted and usable
2478[18:10:00] <rant> jim: idk what you're asking
specifically but the user also mentioned installing packages and
such.. which means their frustration with typing passwords often has
to do with polkit in general not specifically this hdd..
2484[18:11:42] <rant> aside from the nopasswords solution I
proposed, there is also use of a keyring.. which is usually default
behavior.. when a polkit dialog comes up it says "forget
immediately, remember until logout, remember forever" if you
dont read the dialog and select an option, then you're just at
a PICNIC
2510[18:19:36] <rant> its not misconfigured per-se its just
default configured.. thats how the system works, you try to do
something that requires root permissions, it prompts for a
password.. like running synaptic, or mounting a volume
2516[18:20:32] <rant> but we're dancing around unknown
issues here.. we never got at the specifics of how this HDD is
configured, what kind of HDD it is.. or what behavior is desired
2527[18:22:36] <rant> and really if you're this
inexperienced of a user you probably wouldnt benefit from sudo
much.. but it doesnt hurt to use sudo cause it not only allows you
to use that to issue terminal commands, but some things still use
sudo, though most GUI are using polkit now
2528[18:22:47] *** Quits: Lingua (~SWM@replaced-ip) (Remote host closed the connection)
2529[18:22:50] <rant> but you can certainly just setup sudo and
tell polkit to use that
2534[18:25:15] <rant> Psyndrome: if the HDD is the real issue you
want to address, I need more information.. which I asked for twice
and gave you a command to provide it.. if you want to just eliminate
root password prompts altogether its a multi-step process we can go
through but it still requires you to listen and do what I'm
telling you.. it involves commands and editing files.. stuff
you've so far not demonstrated the ability
2535[18:25:21] <rant> to do
2536[18:25:31] *** Quits: dahrens (~Srain@replaced-ip) (Remote host closed the connection)
2538[18:26:22] <rant> I am still not clear on if this HDD is
removable i.e. USB or is internal i.e. a fixed disk, where you want
it mounted, or what sort of disk layout it is.. if there is lvm,
encryption, etc
2539[18:27:15] <rant> what I do know is, that in my experience,
plugging in a USB hdd that is not encrypted just mounts to /media/
without any clicking or passwords, or in my case like the USB HDD I
currently have attached, which is encrypted.. I have to as you say,
provide a password to access it
2540[18:28:19] <Psyndrome> It is internal connected with sata
cables
2541[18:28:44] <rant> Psyndrome: as a more general advice.. you
need to understand this is not like MacOS or Windows or such.. in
Debian we have 7 major Desktop Environments, and thousands of other
packages you can mix and match.. one Debian system is possibly
totally different than another.. so we can't just generically
offer advice without details
2543[18:29:35] <rant> Psyndrome: ok, well an internal fixed disk
should not be mounting to /media/ the reason it is in this case is
likely because you never finished setting it up.. you made no entry
in fstab to tell the system where you want it to be.. so thats why
you have to click on it and provide a password
2544[18:30:12] *** Joins: arne (~rusty@replaced-ip)
2545[18:30:19] <rant> Psyndrome: fstab is a configuration file
that tells the system where filesystems are located and where you
want them mounted.. if you tell fstab that this new hdd exists, and
where you want it mounted, you can then set it to auto mount
2575[18:36:18] *** Quits: Guest37199 (~uranium@replaced-ip) (Remote host closed the connection)
2576[18:36:28] <rant> but if thats what you want, you will need
to unmount it first and create the mountpoint as a persistent mount
point
2577[18:36:52] *** Parts: arne (~rusty@replaced-ip) ()
2578[18:36:55] <rant> because the default behavior your are using
now, the mountpoints are created on-demand by udev
2579[18:37:10] <rant> for fstab to automount it, you need the
mount point to be permanent
2580[18:38:07] <rant> if you don't provide the information
I've asked for I can't do much more than generically
explain how it works and how its done
2581[18:39:20] <rant> without an fstab entry the system only
knows the device exists because of udev daemon finding it.. which
makes it show up in your file manager allowing you to click on it
ant provide the root password to mount it
2601[18:42:56] <rant> and I would imagine the explaination isnt
making much sense to the user
2602[18:43:32] <Psyndrome> it is make sense im just tired its
exactly what you say
2603[18:43:51] <rant> its now apparent to me why the password
dialogs are comming up.. if they'd provided the info from lsblk
and fstab an hour ago, I'd have known this and solved it
already :D
2604[18:44:20] <rant> Psyndrome: so then you /are/ experienced
enough to resolve this then, you are just groggy?
2610[18:45:18] <rant> really all you need is a "/dev/sdc
/media/user/foo ext4 defaults 0 0" or such line in /etc/fstab..
but the specifics would be the information from lsblk
2611[18:45:34] <rant> but you also need to unmount the partition
first and create the mountpoint with mkdir so its persistent
2612[18:45:41] <seven-eleven> trek00, i assume `debuild` compares
the build with the signature provided in the .dsc file, if so how is
this reproducible if the package .dsc file is shipped together from
the same repository, could both tampered with?
2617[18:46:20] <rant> yes well I can't be specific without
specific information :D
2618[18:46:30] <rant> I'm just generalizing here
2619[18:46:58] <rant> it would be best to take the info from
lsblk and plug it into blkid and use the UUID
2620[18:47:28] <jim> rant, yeah, that is true... for that,
it's good if the users want to do it, and feel good about doing
so... beating em up for not asking a question correctly will leave
them with a bad taste in their mouth, which they'll recognize,
and the next time it comes, they'll feel bad about asking
anything
2625[18:48:14] <rant> jim: fair enough.. greycat and I both know
and have done such.. but you also gotta understand we're busy
doing other things.. I spent like an hour pising around with this
not doing what I would've been doing otherwise..
2627[18:48:32] <rant> if a user is taught how to ask better
questions and provide better info, then we can just look in and
answer without doing all this
2632[18:49:39] <rant> I know this all too well cause not only
have I been here nearly two decades, but I've proposed a
project to develop a series of software-based solutions to address
this.. its just not getting any traction.. I need programmers to
help make it a reality
2633[18:49:55] *** Quits: volter (~volter@replaced-ip) (Remote host closed the connection)
2637[18:50:20] <rant> its a huge and highly ambitious project..
with no support as of yet
2638[18:50:22] <Psyndrome> sorry about your tme
2639[18:50:24] <Psyndrome> time
2640[18:50:43] <rant> Psyndrome: no worries, I got plenty of
time.. I'm just trying to make a point here about deficits in
our support
2641[18:51:01] <rant> and where the real problem lies..
2642[18:51:17] <rant> us volunteer supporters being more coddling
isn't going to be the solution
2643[18:52:00] <rant> the system is growing rapidly.. we are
getting more and more users, more and more pieces to support.. etc..
we need to adjust the tools we use not the way we interact..
2644[18:52:34] <trek00> seven-eleven: you can rebuild the package
and then compare if they are bit-to-bit comparable
2667[18:58:49] <rant> I have had the vision I outlined in that
manifesto of sorts on that project page for nearly a decade.. it
would I think solve the problem permanently.. but its going to be a
LOT of work.. I vision for the less experienced users, a frontend
similar to Windows(tm) Help and Support Center that includes access
to manpages, wiki, forums, bts, chat and mailing lists, and remote
desktop support options.. and a
2668[18:58:55] <rant> series of backends and such that make that
easy to use interface interact with experienced users on their terms
without them having to change the way they prefer to do things
2677[19:02:26] <trek00> rant: what about improving debian wiki?
even if I never used archlinux, I always uses their wiki to resolve
my issues, is this a sign? :)
2678[19:02:47] <greycat> If you find a Debian wiki page that
needs help, help it.
2689[19:05:07] <rant> I would like to put some effort into the
wiki.. but even with our slow pace, the wiki is so far behind it
needs nearly rewritten.. I have not been able to figure out a system
that suits me to help identify what is outdated and needs to be
changed and to change it.. its not as simple as the pruning factoids
in the bot which I do as it comes up in here
2697[19:06:42] <rant> I dont like having to deal with the web
interface and mediawiki markup crap.. thats not my preferred way of
doing things.. which is why I proposed putting our efforts into
designing better tools that adapt to the needs of each kind of user
2698[19:07:23] *** flaskbck is now known as flashbck
2699[19:08:06] <trek00> rant: I think your project is trying to
address something that is definitively needed, but it seems to me
most of the work is to instructs users how to diagnose and search
for answers
2700[19:08:20] *** flashbck is now known as flaskbck
2701[19:08:46] *** Quits: SPraus (~SPraus@replaced-ip) (Remote host closed the connection)
2704[19:09:18] <aakko> gentoo wiki is really nice. works great on
mobile too
2705[19:09:25] <rant> trek00: yeah, well its trying to bridge
gaps.. is the idea. To allow each user to use whatever method is
preferable to them and still have access to the same resources
2712[19:11:26] <trek00> rant: from my very personal point of
view, i find simpler to fix things instead trying to explain
something to inexperienced users, but i lacks of social skills
2713[19:11:38] <rant> I also recently have been noodling an idea
about making a new screencasting tool of sorts, that could automate
making things like wiki entries.. something that rather than just
captures video, it captures events.. and creates more metadata so
you can just DO something and record it to a manpage, wiki entry,
howto video, pdf, or any format.. so the tool would generate
something with more metadata indexing the
2715[19:11:44] <rant> process by what you DO not just what it
sees, so you could have a TOC attached to a video or extract
screenshots of individual steps, etc.. and would facilitate going
back and changing only pieces of it easily where things have changed
2718[19:12:29] <rant> I am not a very skilled or experienced
programmer, more of a dabbler.. and these kinds of things are highly
advanced pieces of software..
2720[19:13:09] <rant> My best idea on getting it happening is to
get my stuff together, and perhaps invest in making a nonprofit and
raising the funds to pay people to do it
2724[19:15:20] <rant> but I digress.. can bs about this stuff in
an offtopic channel or other places.. I'm always around.. if
anyone has ideas or plans of how to make these things happen
2729[19:17:47] <trek00> rant: "we don't really want
those kinds of users who are inexperienced because they're just
going to become a draw on our resources without contributing to our
project" what's about a stackexchange-like website for
debian? :)
2730[19:19:13] <jken> Hello, I am trying to build a kernel with
this driver built in:
replaced-url
2731[19:19:29] <rant> trek00: yeah, we should talk about it
elsewhere but those sorts of ideas are in there.. having a ratings
based system like that
2734[19:20:07] <trek00> rant: I see now there are many opensource
alternatives to stackexchange
replaced-url
2735[19:20:24] <rant> ,pciid 10ec:818b
2736[19:20:25] <judd> [10ec:818b] is 'RTL8192EE PCIe
Wireless Network Adapter' from 'Realtek Semiconductor Co.,
Ltd.' with kernel module 'rtl8192ee' in stretch. See
also
replaced-url
2737[19:20:55] <rant> ,pciid 10ec:8812
2738[19:20:56] <judd> [10ec:8812] is 'RTL8812AE 802.11ac
PCIe Wireless Network Adapter' from 'Realtek Semiconductor
Co., Ltd.' with kernel module 'rtl8821ae' in stretch.
See also
replaced-url
2739[19:21:18] <rant> jken: my sources suggest this driver is
already available in debian
2740[19:21:23] * rant checks his kernel
2741[19:21:52] <trek00> jken: it should be already included in
standard debian kernel
2742[19:21:59] <jken> I am trying to compile my kernel with that
driver as a built in.
2743[19:22:15] <rant> jken: ah, not as a module you mean?
2744[19:22:20] <jken> right.
2745[19:22:34] <rant> jken: does the one in debian work for you?
the module?
2746[19:22:49] <rant> jken: cause if so, no need to go upstream,
just need to rebuil the debian kernel with different options
2747[19:23:33] <jken> yes it does
2748[19:23:41] <jken> That's exactly what I am trying to do
2752[19:24:42] <trek00> jken: you said you can't find it in
the ncurses configuration, checking theri Kconfig file, you should
find it under "Realtek rtlwifi family of devices"
2764[19:27:10] <rant> jken: is this for an embedded system
application?
2765[19:27:26] <jken> not technically, its an x86 system.
2766[19:27:29] <annadane> Bushmills, basically you got it right,
it just depends the context you're asking in
2767[19:27:42] <jken> basically a nuc type system (but not intel)
2768[19:27:49] <annadane> testing is to test the next stable
release, sid is just the first point of entry for packages
that'll later migrate to testing
2769[19:27:49] <Bushmills> not a question. I'm just lagging
a bit
2770[19:27:58] <rant> jken: cause losing the security of a
maintained kernel just to build in a wifi driver seems ridiculously
irresponsible
2771[19:28:11] <rant> and a lot of work for no real benefit
unless you have contrstaints that warrant it
2772[19:28:24] <jken> lets say I do.
2773[19:28:30] <rant> jken: furthermore building in a wifi driver
means you can't rmmod it if there are issues
2774[19:28:55] <rant> jken: I'm willing to concede the
point, I'm just trying to make sure you've thought this
through and arent doing this to your detriment
2775[19:29:09] <Bushmills> more a belated comment to an earlier
comment calling sid such
2787[19:32:11] <jhutchins> I take experimental to be
"let's see if anybody can get this to work".
2788[19:32:34] <jhutchins> jken: Hey, that's great, that way
you get all the latest bugs!
2789[19:32:50] <jhutchins> Kinda like testing parachutes.
2790[19:33:05] <rant> jken: well your issue seems beyond our
scope.. as it seems to be you want to know why menuconfig is
quashing the config.. which afaik is what its suppose to do.. you
would likely server yourself better contacting the kernel
maintainers, going to #debian-kernel on OFTC, or pinging lkml
2791[19:33:23] <jken> rant, thanks I'll do that.
2792[19:33:35] <rant> jken: oldconfig target is for people
wishing to use an old config, you arent suppose to do menuconfig in
those cases
2801[19:36:14] <jhutchins> When I started using Linux it was just
after the modular kernel. Every howto began with "recompile the
kernel to enable...".
2802[19:36:29] <jhutchins> Kernel compiles were the most common
benchmark.
2803[19:36:37] *** Quits: ansellzsw8 (~sherman85@replaced-ip) (Remote host closed the connection)
2846[19:50:08] <flashbck> I'm trying to use debmirror to
create a local mirror for my network. I'm trying to keep it as
small as possible by using the --exclude-deb-section flag, but
packages in excluded sections seem to be fetched regardless. Am I
using the command wrong? Here's an example of the command that
I'm using:
replaced-url
2854[19:53:43] <trek00> flashbck: I remeber there was an apt
proxy server may be it is apt-cacher
2855[19:54:33] <greycat> I'm assuming "it's
complicated" means "I can't use HTTP directly from
any of the servers, not even through a proxy, so I am
sneaker-netting the packages into my local mirror on a secret USB
stick".
2866[19:56:05] <Bushmills> apt-cacher-ng appeared to function
more reliably than apt-cacher
2867[19:56:08] * dvs uses apt-mirror
2868[19:56:58] <flashbck> My boss told me that I'm not
allowed to use a proxy. I tried to convince him that a proxy makes
more sense and uses far less bandwidth/storage, but that
doesn't seem to matter to him
2869[19:57:22] <flashbck> like I said, it's complicated
2870[19:57:40] <greycat> Doesn't sound *especially*
complicated so far.
2898[20:05:11] <trek00> flashbck: checking apt-mirror configu
files it seems to me it allows only to limit different
architectures, releases or sections (main contrib non-free)
2899[20:05:16] <greycat> or if you're a sudo person, sudo
env LC_ALL=C apt whatever
2905[20:07:25] <trek00> flashbck: may be you can try to type:
debmirror -a amd64 --no-source -s main -h ftp.us.debian.org -d
buster -r /debian --progress --method=http
--exclude-deb-section='.*admin.*' /srv/apt/data/debian
2906[20:08:01] <trek00> flashbck: or simply
--exclude-deb-section='admin'
2910[20:09:40] <flashbck> I limited the list of exclusions for
the sake of brevity, the actually list has more items. I've
tried the single item per flag option and had no success (eg:
--exclude-deb-section=x11 --exclude-deb-section=gnome etc)
2915[20:11:00] <greycat> Is part of your list of exclusions an
executive mandate "no server is allowed to install any X11
packages" by the same boss that said "no proxy"?
2916[20:11:10] <greycat> Just curious.
2917[20:11:30] *** Quits: xcm (~xcm@replaced-ip) (Remote host closed the connection)
2919[20:11:55] <trek00> flashbck: found on the web
--exclude-deb-section='(games|debug|news|gnustep|ocaml|hamradio|gnu-r)'
is this working?
2920[20:12:03] <flashbck> greycat: no, that is just an attempt by
me to limit the overall size of the mirror. these servers will not
need to run x11 so I was trying to exclude those packages
2932[20:14:25] *** Quits: timahvo1 (~rogue@replaced-ip) (Remote host closed the connection)
2933[20:14:41] <flashbck> Yeah, I found that one too. I've
tried "--exclude-deb-section=x11 --exclude-deb-section=gnome
..." and "--exclude-deb-section=(x11|gnome|...)"
2934[20:15:04] *** Quits: Zvmdyv (~Zvmdyv@replaced-ip) (Remote host closed the connection)
3056[21:09:45] <irl25519> NetTerminalGene should dump his cpu
3057[21:10:25] *** Quits: katyperry (~s@replaced-ip) (Remote host closed the connection)
3058[21:11:05] <NetTerminalGene> my cpu beats yours
3059[21:11:32] <cbthree> i installed buster on my samsung
chromebook 3 and everything works just fine. The only issue i have
is that the microphone does not work, that is: no audio input from
built in microphone. Can anyone help me overcome this problem?
3060[21:11:46] <renner> theoretical question: is it possible to
tar a directory from within itself, or will that just run until
system crash?
3061[21:11:57] <Bushmills> that's possibly what the deb-src
lines in source.list are for
3064[21:12:14] <jelly> NetTerminalGene, more seriously, tools
that can actually gain substantial amounts of performance usually
support various builds of libraries and can pick the best based on
runtime checks
3065[21:12:39] <ratrace> i think this is in reference to feh
dorah considering going avx2 as the minimum supported hw base
3067[21:12:52] <trek00> NetTerminalGene: recompiling all the
debian archive will give you little to no performances benefits,
only some packages gain something with architecture specific
optimizations and only a small fraction of the software has
different path for specific architectures: you could rebuild only
them to have the same results
3069[21:13:34] <Habbie> trek00, it's not just about paths -
it's also about what a compiler can do; but i do share the
conclusion, rebuilding just a few should give most of the benefits
3070[21:14:08] <jelly> renner, GNU tar cowardly refuses to
include the destination in the archive. Try it.
3085[21:15:09] <NetTerminalGene> because it doesn't have
important packages
3086[21:15:21] <ratrace> i wonder why
3087[21:15:36] <NetTerminalGene> jelly, ok
3088[21:16:36] *** Quits: we6jbo (~we6jbo@replaced-ip) (Remote host closed the connection)
3089[21:16:37] <renner> trek, jelly: the thought started with
"how to make a system image on a remote virtual server and back
it up".. maybe there is a smarter way, although I suppose one
could tar each top directory individually and keep the result in /
3090[21:16:54] <trek00> NetTerminalGene: rebuilding packages is
nearly automated with dpkg-buildpackage, you can try to rebuild some
package you like
3098[21:19:15] <trek00> renner: if you need a system image, you
should dd the entire partitions, instead if you need to make
incremental backups, via network, rsync/rsnapshot are your friends
:)
3102[21:20:24] <SH2> Hey:), Does anyone know a German Debian irc
channel?
3103[21:20:31] <jelly> !de
3104[21:20:31] <dpkg> Deutschsprachige Hilfe bekommt ihr in
#debian.de (auf irc.oftc.net, irc.freenode.net oder irc.belwue.de) -
German speaking users please go to #debian.de (on irc.oftc.net,
irc.freenode.net or irc.belwue.de).
3105[21:21:01] *** Quits: Prints (~333@replaced-ip) (Ping timeout: 276 seconds)
3106[21:21:10] <SH2> Thank YOU.
3107[21:21:56] *** Joins: Prints (~333@replaced-ip)
3108[21:23:06] <cbthree> someone told me that upgrading to kernel
5.2 may solve my problem as 4.9 may not fully support my hardware.
The thing is that i can not find the kernel 5.2 deb package for
buster
3109[21:23:23] <greycat> There aren't any Debian 5.2 kernel
packages at this time.
3125[21:25:53] <cbthree> tarzeau, i installed buster on my
samsung chromebook 3 and everything works just fine. The only issue
i have is that the microphone does not work, that is: no audio input
from built in microphone.
3133[21:26:40] <jelly> we don't build kernels that way any
more
3134[21:26:42] <tarzeau> but sid has kernel-package
3135[21:26:50] *** Quits: Immanuel (~Manu@replaced-ip) (Ping timeout: 272 seconds)
3136[21:26:52] <jelly> and it's borked
3137[21:26:52] <user___> okay, so my thread was deleted from
debians facebook page few months ago about weboob scandal
3138[21:26:53] <tarzeau> jelly: no? wtf did i miss?
3139[21:27:00] <greycat> upstream kernel sources have some sort
of make target that creates a .deb, so you just use that, unless I
missed another meeting
3140[21:27:09] <annadane> user___, we're not in charge of
debian's facebook page
3141[21:27:15] <jelly> tarzeau, upstream's "make
deb-pkg" and whatever kernel-handbook says to do these days
3142[21:27:16] <annadane> i wouldn't drag #debian into this
3143[21:27:29] <user___> but it is so interesting
3144[21:27:30] <tarzeau> cbthree: luck you, it got so easy
3145[21:27:54] *** Joins: Immanuel (~Manu@replaced-ip)
3146[21:27:55] <humbot> weboob!
3147[21:28:03] <user___> interesting discussion could be held
about ti
3164[21:31:16] <annadane> user___, if you do want to contribute
to debian by starting a certain conversation about policy, let us
know what it is and we can direct you to the proper channels
3169[21:31:50] <dpkg> It's not just the name. It uses
gratuitously sexual icons, and derogatory/inflammatory language in
its output. See <replaced-url
3170[21:31:56] <tarzeau> annadane: it was possible to apt-get
install sex and wine (3 packages), and i think sextractor also still
exists (and is useful)
3218[21:41:24] <cbthree> jelly, i've been told the problem
might have been fixed in 5.2. Might is not certainty
3219[21:41:32] <jelly> !tias
3220[21:41:32] <dpkg> TIAS is "Try It And See".
3221[21:41:33] <user___> hmm
3222[21:41:38] <jelly> !immediate configuration
3223[21:41:38] <dpkg> When upgrading from squeeze to wheezy, apt
may stop with "E: Could not perform immediate configuration on
'$package'." First, try running apt-get upgrade; . If
that fails, try running apt-get dist-upgrade -o
APT::Immediate-Configure=0; instead. Finally, if that doesn't
work, if $package isn't an Essential package, then "dpkg
-r $package; apt-get -f install. You should also read <release
notes>.
3234[21:42:50] <jelly> I guess you can try that anyway
3235[21:42:52] <renner> but on the way from wheezy to jessie,
libdevmapper is somehow unhappy and causing dpkg to fail without
telling me what it needs
3236[21:43:07] <afidegnum> tarzeau: some machine learning stuff
3237[21:43:30] <jelly> cuda is nvidia-specific to the best of my
knowledge
3238[21:43:31] <tarzeau> afidegnum: using tensorflow or pytorch?
3240[21:43:45] <azeem> renner: do you have some more output? Or
maybe just the wohle output of apt-get upgrade?
3241[21:43:48] <renner> jelly: what I'm hoping is the case
is that I can configure apt to skip this issue... not sure what the
consequences are
3242[21:43:50] <afidegnum> pytorch
3243[21:43:52] <tarzeau> afidegnum: i know tensorflow can work
slowly with cpu only just fine (just slower)
3244[21:44:15] <jelly> renner, read what dpkg said above.
3245[21:44:31] <afidegnum> :)
3246[21:44:34] <afidegnum> ok let me give it a try
3247[21:44:41] *** Quits: cryptodan (~cryptodan@replaced-ip) (Remote host closed the connection)
3248[21:45:02] <renner> azeem: "E: Could not perform
immediate configuration on 'libdevmapper1.02.1'. Please
see man 5 apt.conf under APT::Immediate-Configure for details.
(2)"
3265[21:50:37] *** Joins: Prints (~333@replaced-ip)
3266[21:50:44] <tarzeau> afidegnum: i think there's
something called opencl, that is not nvidia specific, but it
didn't work for a friends amd hw (the memtestcl thing)
3267[21:51:04] <tarzeau> afidegnum: i've only worked with
nvidia cuda cards 1080+ti/2080+ti (8/11 gb)
3277[21:52:54] <tarzeau> i know some people use nvcc directly,
some do cuda via pytorch(fb)/tensorflow(google). and that just works
best with nvidia, since 5+ years
3281[21:55:01] <tarzeau> intel has some "retarded"
deals with microsoft (only ms can think of stuff like , i don't
remember the name, but some printer producers created printing
hardware that'd only work with windows)
3326[22:11:17] <trek00> renner: probably Bushmills is right, it
seems startpart was introduced with jessie, to replace standard
sysvinit scripts execution with its own
3328[22:11:52] <renner> Bushmills: yes, that is my understanding,
but I don't know why it can't get past that point. replace
one with the other, done. instead, it stops the process with no
obvious option of how to proceed
3345[22:16:01] <tarzeau> (secretly hoping debian will do it too)
3346[22:16:33] <renner> so consensus is that my current
installation is screwed?
3347[22:16:54] <trek00> renner: if you try to force installation
like dpkg --force-something starpar.deb?
3348[22:17:26] <tarzeau> the only installation screwed i ever
saw, was the one where i run mkfs.ext2 on the root file system (15
years ago), still got the disk, and still no idea how to recover it
3349[22:17:39] *** Quits: CombatVet (~c4@replaced-ip) (Remote host closed the connection)
3350[22:17:58] <greycat> that "Adoption" section looks
rather political... why even include Devuan there at all, since
clearly it will never adopt?
3351[22:18:17] <trek00> rain2: on fstab you should have something
like: proc /chroot/proc proc defaults 0 0
3352[22:18:24] *** Quits: subopt (~subopt@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
3353[22:19:00] <rain2> thanks
3354[22:19:21] <trek00> greycat: adoption means how much was
included at overall, so it's good to list who changed and who
not
3355[22:19:21] <Bushmills> rain2, you may also want to mount
--bind /dev to chroot
3358[22:19:52] <greycat> well, it's a battle I am not going
to join today
3359[22:19:52] <renner> tarzeau: well, this is remote, so if it
doesn't start up again, there are no escapes
3360[22:20:02] <trek00> rain2: yes /dev /proc and /sys are often
required
3361[22:20:56] <tds> rain2: on the subject of systemd, you can
also use something like systemd-nspawn -D /chroot rather than chroot
/chroot to do all the bind mounting etc for you, more or less :)
3362[22:21:03] <tarzeau> renner: just never reboot? :)
3363[22:21:06] <trek00> renner: if you already upgraded some
packages and still in the middle of upgrade process without
finishing it, yes
3364[22:21:17] <rain2> woah cool
3365[22:21:52] <tarzeau> renner: i've got a savegame of
nethack on nethack.alt.org (on the final altar before the planes,
saved 10 years ago). if you don't move you can't die.
3366[22:21:56] <renner> tarzeau: it's managed, they will
reboot it for me periodically, nothing I can do
3367[22:21:58] <rain2> nspawn program does not seem to be on my
system
3368[22:22:06] <greycat> systemd-nspawn is optional, I think
3369[22:22:21] <tarzeau> renner: remove all reboot binaries
3370[22:22:36] <tds> yeah, it's in the systemd-container
package iirc
3371[22:22:46] <greycat> which is ... Priority: optional
3372[22:22:52] <trek00> renner: there are some packages that got
upgraded?
3373[22:23:17] <renner> tarzeau: pretty sure I can't prevent
it with software means
3374[22:23:23] <renner> trek00: yes, there are
3375[22:23:40] <trek00> renner: well try to force startpar
installation to continue
3387[22:27:11] <renner> yeah, I meant, I am at that point... but
the real magic will be in the reboot, I guess - question is whether
to continue upgrading towards stretch and buster and find out later
if the reboot fails, hoping for special healing powers of anything
that might have been broken, or find out the truth immediately
3388[22:27:53] <trek00> renner: you made a backup of the disk
before?
3399[22:29:59] <renner> trek00: yeah, I have all the bits I
should need to reconstitute if a fresh install is the only option;
going back to squeeze and trying again might be adventurous, though
because that disk image is probably not available anymore, nor the
one for wheezy, I bet
3400[22:30:12] <litb> any idea?
3401[22:30:26] <renner> oh, and I think I have an auto backup
facility that I've never used
3402[22:30:28] *** Quits: stefanos82 (~stephen@replaced-ip) (Quit: Quitting for now...)
3405[22:30:48] <renner> so there's probably an automatic
image hanging in that for the next five days or so, come to think of
it
3406[22:30:59] <greycat> apparently they disconnected and
re-joined in the middle of receiving answers
3407[22:31:14] *** Quits: guest567 (5c8a2611@replaced-ip) (Remote host closed the connection)
3408[22:31:19] <trek00> renner: normally i would do: create a new
instance of the system starting from backups, try to upgrade, if
it's all ok, try on the production machine
3409[22:31:36] <trek00> greycat: oops
3410[22:31:39] <litb> trek00, sorry, I disconnected and missed
his answer :/
3411[22:32:01] <trek00> litb: check /var/log/dpkg.log or its
backups /var/log/dpkg.log.[0-9]*
3412[22:32:09] <litb> it appears that the last update was on Sat,
Jun 15 2019 17:45:53 +0200
3434[22:47:24] <irl25519> Can anyone tell me how to forward
traffic from wireguard to a docker container on debian? Presumably
there's an iptables rule for forwarding from wgnet0 to the
ethernet interface of the container?
3435[22:47:32] <irl25519> Been trying but my iptables-foo is weak
3444[22:52:35] <irl25519> ok let me rephrase. all wireguard
traffic needs to go through a specific docker container (which will
basically inspect the traffic) and then the traffic will be
forwarded out the container, which will then hit the default gateway
to leave the machine
3445[22:52:59] <irl25519> i seem to be struggling only with the
first part
3446[22:53:37] <Wulf> irl25519: search for "lartc" /
policy routing
3452[22:56:09] *** Quits: litb (~litb___@replaced-ip) (Remote host closed the connection)
3453[22:56:32] <irl25519> ok, i'll go back to trying the ip
rule add ...
3454[22:56:45] <irl25519> with the custom table
3455[22:57:09] <irl25519> if that's the best way to do
things
3456[22:57:14] <Wulf> irl25519: out of interest: How do you
"inspect" the traffic and why?
3457[22:57:50] *** Quits: magbitang (~siddall@replaced-ip) (Remote host closed the connection)
3458[22:58:04] <irl25519> this is only useful for my personal
unencrypted traffic
3459[22:58:37] <irl25519> content filtering etc
3460[22:58:57] <Wulf> okay. And how?
3461[22:58:59] <trek00> irl25519: do you logs dropped packages to
debug?
3462[22:59:02] *** debhelper sets mode: +l 1557
3463[23:00:00] <Wulf> irl25519: I mean I assume there's a
process running in your container. How will it get access to the
traffic? pcap? nfqueue? transparent proxy?
3485[23:08:42] <trek00> irl25519: usually it's straight
forward to enable forwarding, so if packets do not go over, you
could logs them to debug with a rule like: iptables -A FORWARD -m
state --state INVALID -j LOG
3486[23:08:54] *** Joins: faris (~faris@replaced-ip)
3487[23:09:08] *** Quits: earthundead (~earthunde@replaced-ip) (Remote host closed the connection)
3490[23:09:42] <irl25519> that sounds extremely useful. because
every time i try a forwarding rule that isn't quite correct,
the only thing i can see is that packets don't arrive on the
destination bridge
3494[23:10:56] <trek00> irl25519: sorry I past the wrong command,
you should add this at the end: iptables -A FORWARD -j LOG
3495[23:11:05] *** Quits: alexertech (~xb@replaced-ip) (Quit: Fades into the darkness)
3496[23:11:22] <trek00> irl25519: not at start because it logs
every packet, but only to the end as it would log packets not
intercepted by other rules
3497[23:11:57] <irl25519> thanks for the correction
3502[23:12:33] <irl25519> was just in the middle of trying out
wulf's suggestion. got the infamous "nexthop has invalid
gateway" error. nice and specific.
3516[23:16:49] <petn-randall> benj1: This is also the newbies
channel. :)
3517[23:17:17] <dvs> There's newbies here?!?
3518[23:17:20] * dvs runs away
3519[23:17:51] <trek00> :D
3520[23:17:57] <annadane> how do i computer ???
3521[23:17:58] <benj1> xD
3522[23:18:03] <dvs> heh
3523[23:18:19] *** Guest51921 is now known as Chex-
3524[23:18:56] <petn-randall> benj1: Feel free to ask, we'll
try our best to help you out. If the question has already been asked
very often, we sometimes even have a factoid or documentation
written for that case.
3562[23:46:25] <petn-randall> benj1: Most IRC clients allow you
to type the first letters of a nick, and autocomplete it with tab.
So I typed "ben<tab>" which turned into "benj1:
". And most IRC clients highlight you then.