1[00:00:14] <Maws> somiaj, yea, tried to do it as root as
well....I'm on debian testing but still....
2[00:00:21] <somiaj> !buster su
3[00:00:21] <dpkg> To start an interactive session with
<su>, you should use "su -l" or "su -";
this has always been the case, but is now enforced in Debian 10
'buster', where the PATH is no longer reset in
non-interactive sessions. Alternatively, you can ALWAYS_SET_PATH yes
in /etc/login.defs. See
replaced-url
11[00:02:46] <somiaj> there are reasons you don't want su
- though, they are both needed for different reasons
12[00:02:57] <somiaj> but yea, this change is affecting lots of
users
13[00:03:23] <annadane> i mean i just use su - for everything,
i'm unaware of dangers or downsides to doing so
14[00:03:33] <annadane> i know the *difference* between the two
15[00:03:59] <Maws> hmm, looks like the depmod error is gone
after doing su -, but then this error occure, any tips on that ?
" snxterm.c:12:10: fatal error: ncurses.h: No such file or
directory
16[00:03:59] <Maws> #include <ncurses.h>
17[00:03:59] <Maws> "
18[00:04:16] <somiaj> Maws: this is due to missing the -dev
package for that libary
19[00:04:25] <Maws> (it's for installing drivers for a
rs232 card....)
27[00:05:41] <Maws> Thnx, at least is seem that it installed
properly :), cheers!
28[00:07:16] <annadane> "For backward compatibility, su
defaults to not change the current directory and to only set the
environment variables HOME and SHELL (plus USER and LOGNAME if the
target user is not root). It is recommended to always use the
--login option (instead of its shortcut -) to avoid side effects
caused by mixing environments."
29[00:07:32] <annadane> well, shit.
30[00:09:10] <annadane> despite the fact - -l and --login are
all listed as equivalent, under options
31[00:09:41] <Maws> Guess I need to do a restart to get this
mod loaded.....:)
37[00:12:23] <annadane> i don't think i do anything wacky
with my environments in any case, both should be fine
38[00:12:49] <somiaj> yea, most people make very little use of
their enviorment outside of say PATH, HOME, and a few other small
things
39[00:13:28] <somiaj> But very little software these days uses
enviorment variables to get settings, or if it does, users
don't use it.
40[00:13:29] <annadane> it's just tracking it that becomes
an issue, if you do something in january and then forget about it
and then in june you use the "wrong" invocation of root,
and end up screwing something up irreparably...
41[00:13:30] <voldial> Anyone see my mistake? I cant get gpg to
use a proxy. The docs and everything I find online seem to indicate
that --keyserver-options "http-proxy=replaced-url
58[00:18:20] <annadane> i think the unprivileged user's
environment is prone to change more frequently and is more
immediately familiar to the user so you should just probably use su
unless you require a utility in /sbin, and use root's
environment for particular edge cases that you know you need
59[00:18:55] <annadane> just a cleaner separation that way
69[00:21:52] *** Quits: Night-Shade (~TimF@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
70[00:22:17] <annadane> it seems counterintuitive, because why
use the *user* environment as *root*... but i feel that's
probably good general advice, but this is just conjecture
71[00:23:31] *** Quits: The_Loko (~The_Loko@replaced-ip) (Remote host closed the connection)
72[00:25:13] *** Lord_of_Life is now known as Lord_of_Life_
75[00:27:02] <somiaj> annadane: maybe you need enviroment
variables that you ahve setup as that user, but to run an operation
as root. Again edge case which is why the manual you linked say most
cases use -l
81[00:29:00] <annadane> it says use --login instead of - to
avoid mixing environments but it's also a linguistic ambiguity
because it's not clear whether the man page means "always
assume root's environment in general" or "always use
--login instead of -l" or both
85[00:29:36] *** Quits: Burek (~Burek@replaced-ip) (Remote host closed the connection)
86[00:29:44] <annadane> i believe it means the former, always
have a consistent root environment... but to use --login rather than
- or -l
87[00:30:10] <annadane> which is still confusing, because it
doesn't tell you *why*... and if you scroll down to options -
-l and --login are all listed as having the same capabilities
88[00:30:28] *** Quits: xcm (~xcm@replaced-ip) (Remote host closed the connection)
89[00:31:20] *** Quits: Gigglebyte (~Giggles@replaced-ip) (Remote host closed the connection)
90[00:31:27] <annadane> also, this su man page is dated july
2014, before the Buster changes
97[00:33:23] <somiaj> annadane: as the factoid said, this has
always been the case, buster is just now enforcing it
98[00:33:25] <xtc_> no probs
99[00:33:53] <somiaj> basically the command was allowing users
to use it how it was not enteded and this 'bug' was fixed
in buster
100[00:34:12] <somiaj> unsure if this was due to some patch in
debian, or upstream.
101[00:35:29] <annadane> so much is shoved into /sbin anyway,
it's practically worth it to just always use su -, edge cases
be damned... i just guess the fear is that if you forget about
something critical in your environment you'll end up doing
something really bad
105[00:36:34] <somiaj> the only thing in sbin should be stuff
that requires root. In my expreince ifconfig was the only thing that
was in /sbin that I would want to run as anormal user, since
ifconfig can give you info on intefraces as a normal user, but
can't modify.
106[00:36:37] <annadane> i guess because *more* changes in the
user environment - users set environment variables regularly -
it's better to just assume the root environment as it's
cleaner
107[00:36:47] <annadane> so anyway, sorry for rambling
126[00:38:34] <_ioannis> I used to feel the same with systemd
127[00:38:39] <_ioannis> but now I love it
128[00:39:02] <_ioannis> now to figure out that NetworkManager
error :( I don't like having errors in logs :(
129[00:40:35] <annadane> anyway i think i have it backwards, i
think one should use su - for generic system tasks and su for edge
cases... but it still confuses me that the su man page directly
tells you to use --login and not - when they're both listed as
having the same behavior, but again that man page is also from 2014,
years before whatever changes buster introduced
130[00:40:58] *** Joins: deb (~unlike@replaced-ip)
131[00:41:02] <annadane> i think *someone* isn't on the
right page
132[00:41:09] <annadane> probably worth a bug report
143[00:50:14] <annadane> we need a 'becomerootpls' and
if you're trying to do something you're not allowed to the
system tells you 'dothisinsteadyouidiot'
153[00:56:11] <LtL> i'm just surprised this whole 'su
-' thing is such a big deal. like somiaj said, buster just
enforces it as far as root goes. 'su username' is the
'other way to use 'su'
154[00:56:40] <alei1180> 8
155[00:57:13] <LtL> but 'su root' is a bad idea imo,
never did it.
156[00:57:31] <annadane> just su by itself you mean
157[00:57:37] <LtL> yes
158[00:57:43] <annadane> mhm
159[00:58:04] <annadane> i guess i'm just confused why the
man page says use --login instead of - or -l when it goes on to
describe that they're the same things
160[00:58:06] <somiaj> I just think most didn't realize the
differnce between su username and su - username
161[00:58:10] <LtL> you can become another user that way, and
'cd' to their directory.
162[00:58:34] <annadane> well, only if you know their password,
but point taken
163[00:58:42] <LtL> yes annadane
164[00:58:47] <somiaj> since their limited use of su, and the
fact it changed the PATH in the past didn't affect them.
165[00:59:00] <somiaj> you don't have to know their
password if you are root
166[00:59:13] <somiaj> I use su - username all the time as root
on systems I manage to beomce other users to check things out
167[00:59:19] <_ioannis> I happened to need that function today,
su from root to my main user
168[00:59:25] <annadane> but yeah the root environment will
intrinsically be cleaner so it makes sense to do system
administration in there especially as buster has now introduced this
change
169[01:00:07] <somiaj> annadane: I have no clue what (if any)
the differences between using -l and - is, maybe it is a shell thing
and it just clearn to use full conflig flags vs a single dash
170[01:00:18] <LtL> when uncertain about environment, type
'env'
172[01:00:40] <annadane> -i isn't an option, apparently
173[01:00:42] <annadane> i think in sudo it is
174[01:01:03] <LtL> yup with sudo, login versus interactive
175[01:01:44] <somiaj> I wonder how many users actually take
avantage of the login vs interactive shells, my experince is very
few.
176[01:01:49] <annadane> but read 'man su', it says
basically "please use --login and not -" then proceeds to
list - -l and --login as equivalent, doing exactly the same things
177[01:01:54] <annadane> it directly contradicts itself
191[01:05:18] <somiaj> annadane: my man page in stretch clearly
explainst he differences, "When - is used, it must be specified
before any username. For portability it is recommended to use it as
last option, before any username. The other forms (-l and --login)
do not have this restriction."
192[01:05:48] <annadane> right, in buster it doesn't say
that
194[01:06:16] <_ioannis> is it related to Preserve the entire
environment, i.e. it does not set HOME, SHELL, USER nor LOGNAME.
This option is ignored if the option --login is specified.
195[01:06:30] <_ioannis> because it specifically says it's
ignored if --login is specified
196[01:06:44] <_ioannis> just trying to understand what "
to avoid side ef‐
197[01:06:46] <_ioannis> fects caused by mixing environments.
267[01:41:52] <_ioannis> Sorry for asking again but perhaps
someone else is looking now, I'm seeing
((src/settings/nm-settings-connection.c:361)): assertion
'<dropped>' failed
270[01:42:11] <_ioannis> Googling didn't help much, if
anyone has any ideas I would appreciate it :)
271[01:42:49] <Eryn_1983_FL> what if my package cant use 11?
272[01:42:51] <somiaj> I'm unsure, what network card, could
you be missing firmware? Is it wifi/wired, do you have this card
mentioned in your interfaces file?
273[01:43:00] <Eryn_1983_FL> what if i only get support for 8
not 11 on my tool?
274[01:43:10] <somiaj> Eryn_1983_FL: then you are on your own,
download 8 and put it in /opt, or just use stretch
275[01:43:16] <Eryn_1983_FL> hmmm
276[01:43:21] <_ioannis> somiaj: It has an onboard wired and
wireless card but I'm specifically asked it to ignore the wifi
277[01:43:24] <somiaj> but I don't think even oracle
supports java 8 anymore, your app needs to update
278[01:43:28] <_ioannis> I've*
279[01:43:37] <Eryn_1983_FL> indeed so does the rest of my
company lol
280[01:43:50] <_ioannis> Wired works fine
281[01:44:12] <_ioannis> Wireless I don't really
care/don't want it :D
282[01:44:13] <somiaj> java is even up to version 13 now
284[01:44:35] <somiaj> Eryn_1983_FL: well that is why people
still run stretch and stretch will have lts support for another 2-3
years, so there is time
287[01:44:51] <BCMM> Eryn_1983_FL: then you should talk to
whoever maintains that package, because java 8 is dying in general,
not just in debian.
288[01:44:56] <somiaj> but if you want an older version of java
in buster, you have to install/manage this yourself, and also be
incharge of security issues (of which there are lots)
301[01:47:50] <somiaj> Eryn_1983_FL: yea, oracle is kinda a pain
here, they do provide much better support for their closed java
(which you often have to pay for), but the openjdk only gets support
for recent versions
302[01:48:04] <Eryn_1983_FL> yeah
303[01:48:06] <somiaj> if your company is going to stick with
java 8, you may want to have them pay for oracles version and
install that
358[02:13:07] <somiaj> Thedarkb: you can't just install
random .debs in buster, where did you get this .deb, what makes you
think it is compadable with buster
359[02:13:10] *** Oksana_ is now known as Oksana
360[02:13:30] <Thedarkb> somiaj, I got it from the Jessie repo.
361[02:13:46] <somiaj> yea, you can't expect a jessie .deb
to be installed in buster
362[02:14:01] <somiaj> you need to have whatever software
requires this outdatded lib be updated to use the modern libpng
368[02:14:58] <Thedarkb> I'll just roll my own build and
statically link libicu52 and libpng12
369[02:15:01] <somiaj> or if the software will actually work
with a modern libpng, just put a link to force it.
370[02:15:07] <somiaj> Thedarkb: yea that works too
371[02:15:08] <Henry151> i don't know if i want to get into
a can of worms messing around with upgrading this thing.
372[02:15:34] <somiaj> Henry151: debian 9 will have lts support
for another half year or so
373[02:15:39] <Thedarkb> Wait, put a link and force it, somiaj ?
374[02:15:44] <Thedarkb> How do I do that?
375[02:15:53] <somiaj> Thedarkb: ln the libpng you have to
libpng12.so.0
376[02:16:05] <somiaj> that way the file is in place, but it is
actually pointing at the libpng on your system
377[02:16:06] <Thedarkb> Ahhh, thanks.
378[02:16:15] <somiaj> some software will just work, others
won't
379[02:16:30] <LtL> Henry151: why do you repeat non-free so
much?
380[02:16:31] <somiaj> depends on what calls the software makes
and how backwards compadable libpng is
381[02:17:05] <somiaj> Henry151: oh wait, you said 9 (got
confused with 8), debian stretch is supported until next summer, and
lts will support a limited set of packages until 2022, so you have
time before you upgrade.
382[02:17:08] <Henry151> LtL: probably had no idea what i was
doing at the time and was just putting it every-which-where
388[02:18:43] <somiaj> Henry151: sometimes if things are a mess
(due to learning), a backup reinstall is worthwhile to ensure you
are following good practices
389[02:18:43] <Henry151> i've tried to do all sorts of
crazy things on it while not knowing what i was doing, and yet, i
achieved some things with it that i don't really know how to
achieve again, so i don't want to break anything, it's all
working
393[02:20:08] <Henry151> uptime is 759 days :) it is an OVH VPS.
I'm actually realizing that it's where my irssi i'm
using to talk to y'all right now is running
415[02:30:51] <somiaj> LtL: yea, I got confused, for some reason
I thought 'doesn't support buster' was some of those
old openvnz vpses running on a shared 2.32 kernel or something like
that
416[02:31:03] <somiaj> as opposed to the company just
dosen't support buster
417[02:31:30] <somiaj> arg 2.6.32 kernel, but there have been a
few of those having trouble upgrading to buster because of that
418[02:31:39] <somiaj> since libc6 now requires 3.2 or newer
419[02:32:57] <LtL> somiaj: ovh may support buster by now, not
sure. i installed stretch on mine which was jessie prior to that.
420[02:33:19] *** Quits: Theroxat (~quassel@replaced-ip) (Remote host closed the connection)
439[02:47:55] <LtL> Henry151: you should really look to see if
you have a newer kernel in / and reboot the thing, better to apt
update && apt upgrade first.
496[03:38:29] <trui> when i looked up a firewall tutorial, it
used iptables and i only learned that they switched it after,
haha... but i still access it with /usr/sbin/iptables-legacy
505[03:39:49] <eblip> i looked on debian wiki and it tells how
to change to legacy..if you want ..and hten tells how to build a
file
506[03:40:01] <eblip> i built a file and saved it .and now i
have my rules.
507[03:40:09] <eblip> but at the back end i believe it uses
nftables
508[03:40:20] <eblip> in fact ..now i am totally confused
509[03:40:25] <eblip> buit they are working
510[03:40:29] <trui> fortunately now that i'm on debian, i
can not worry about my system and use that stress-free time to
actually learn things ;)
511[03:41:07] <trui> vs other distros where i
'learned' as things broke, and often forgot what i did
exactly
512[03:41:32] <eblip> yes same here...i was using gentoo for a
few years..but not its a business machine and my screen went...
513[03:41:53] <eblip> i plugged in another screen and half the
stuff was off the screen..so i had to dig around for drivers..etc.
514[03:42:06] *** Quits: Tom01_ (~tom@replaced-ip) (Remote host closed the connection)
515[03:42:18] <eblip> on top of that i had to do a serious
update and i just cant afford that any more on a business machine
..although i do love gentoo
516[03:43:04] <trui> for me it was distro-hopping syndrome, but
no longer
517[03:43:30] <eblip> yes ...im sticking with debian now..spent
a few days getting my exact same gentoo set up and everything in
place
518[03:43:36] <eblip> but will be much quicker next time
525[03:52:14] <trui> i actually used a live puppy linux usb for
longer than i'd like to admit (harddrive failed and i kind of
liked it), but debian is a dream in comparison
749[07:28:27] <oiaohm> Not a new kernel feature. Bluetooth
devices alter their speed based on noise.
750[07:28:43] <oiaohm> and the controller chips get stuck.
751[07:28:57] <oiaohm> once they have turned a device to min
they stay there.
752[07:29:20] <k-man> oh interesting
753[07:29:26] <k-man> so a reboot may improve things?
754[07:29:33] <oiaohm> k-man: that what you are pointing to is
most likely not a kernel bug. Its like someone put in a new access
point near by that is now interfering.
755[07:29:54] <oiaohm> Radio interference with bluetooth is a
serous pain in the ass.
756[07:30:05] <oiaohm> Due to its auto band reduction in case of
noise.
757[07:30:17] <k-man> oiaohm: i only just purchased this mouse
and turned it on. this is my first experience with it
758[07:30:29] <oiaohm> Its not like RF mouse that will try to
transmit stronger to attempt to overcome the noise.
759[07:30:40] <oiaohm> and then have shorter battery life.
763[07:32:07] <oiaohm> I tried using the mouse on another
computer with bluetooth and it works fine, so the mouse
shouldn't be the problem. << Like this line here is
classic. You move to a different computer you have movied to another
physical location with a different noise profile.
764[07:32:11] <k-man> thanks for the good news aib
765[07:32:17] <k-man> oiaohm i mean
766[07:32:52] <oiaohm> I am not saying absolutely that it not a
kernel fault but rf noise is a major problem.
767[07:33:11] <oiaohm> I use a lot of wired.
768[07:33:15] <k-man> so this mouse also comes with a usb dongle
769[07:33:22] <k-man> i just tried that - works much better than
bt
876[09:23:22] <somiaj> Guest76987: I would check the bug reports
or maybe the themes, also look in what debian patches were added
between -4 and -8, I see very little difference, so maybe some
regression due to a bug fix
952[10:31:58] <Eryn_1983_FL> hey guys how secure is the upstream
for snapd?
953[10:32:10] <Eryn_1983_FL> are the packages gpged like apt
repos?
954[10:32:22] <Eryn_1983_FL> who maintains them?
955[10:32:38] <jeanpica> I use an extra repository for some
packages, apparently one of those mirrors is corrupted, but my
Debian 9 keep chosing that again and again so apt upgrade keep
failing
replaced-url
956[10:33:00] <jeanpica> is there a way to tell apt to avoid a
certain mirror?
957[10:33:58] <Eryn_1983_FL> comment it out?
958[10:34:22] <Eryn_1983_FL> go into
/etc/apt/sources.d/packagerepo.list and put a # in front of the
replaced-url
964[10:40:05] <jeanpica> Eryn_1983_FL I need a package from that
repository, the problem doesn'T seems to be the whole
repository but only one of its mirrors
998[11:14:45] *** Quits: DelPhyinf (~DelPhyinf@replaced-ip) (Remote host closed the connection)
999[11:15:42] <ws2k3> would it be possible to block arp requiest
based on ip range? i have a use case where an application is sending
out arp on all interfaces. and this is causing some routing issues
1006[11:19:11] <Eryn_1983_FL> it worked fine on the host same
exact one liner
1007[11:22:17] *** Quits: eipip1e0 (~quassel@replaced-ip) (Quit: No Ping reply in 180 seconds.)
1008[11:22:35] <th0r> Eryn_1983_FL, I don't see where you
tell gpg what to do with the key....shouldn't there be a
--import or some such in that command?
1026[11:27:47] <th0r> Eryn_1983_FL, the error says the repo
isn't signed. That isn't an issue with gpg, it is a
problem with the repo. I don't tinker like I used to, but I
think you are just running into one of the 'new and
improved' safety measures.
1167[12:53:05] <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>.
1168[12:53:07] <Eryn_1983_FL> hey guys am i doing systemd wrong?
i got actual file in /etc/systemd/system instead of links..
1169[12:53:32] <Eryn_1983_FL> and systemd is saying they are
masked and i cand unmask, these would be my samba mount services i
made.
1186[13:05:19] <Eryn_1983_FL> it is not telling me much
1187[13:05:49] <zamuro> !flood
1188[13:05:49] <dpkg> It's considered impolite to paste many
lines of text on IRC. Please don't do it. Pasting one line is
fine. Pasting two lines you can usually get away with. Pasting three
lines will get you insulted. Pasting four or more lines will get you
kicked. If you want to paste, ask me about <paste>
1189[13:06:02] *** Quits: Mrbuck (~Mrbuck@replaced-ip) (Remote host closed the connection)
1331[14:25:48] <Lady_Aleena> Hello. How many times can my kitten
step on the power button on my battery backup, which powers down my
computer, before booting back to Debian becomes a mess?
1352[14:30:21] <_ioannis> so to understand better it steps on a
UPS?
1353[14:30:22] <phogg> Lady_Aleena: but if you need cheap get
some card stock and scissors. You can get much the same effect with
a few cuts and folds, a little bit of tape
1354[14:30:23] <Lady_Aleena> I am hoping the cat gets too big to
get back there sometime soonish.
1355[14:30:52] <Lady_Aleena> It has been about a month since it
last happened.
1356[14:31:11] <_ioannis> I've seen a lot of types of UPS
devices and I can't think of a way for a kitten to do it
1371[14:34:11] <Lady_Aleena> However, I grew up in a time where
putting a key into a power socket was almost normal. Parents
weren't so over protective back then.
1373[14:35:39] * Lady_Aleena stuck a bobby pin into a power socket
once, never again. 8)
1374[14:35:42] <Unit193> I think we're straying off course.
The answer to the question is "You should prevent the cat from
getting back there"
1375[14:35:52] <_ioannis> ^ agreed.
1376[14:35:58] *** Quits: xsisec_ (~xsisec@replaced-ip) (Remote host closed the connection)
1377[14:35:58] *** Quits: XsiSec (~xsisec@replaced-ip) (Remote host closed the connection)
1378[14:36:27] <Lady_Aleena> Unfortunately, there isn't an
easy solution for that in this case.
1379[14:37:07] <brokencycle> Hi! I am having massive trouble with
gmtp. It takes ages to connect my phone (30 minutes?), and it
crashes on the earliest opportunity. Any ideas?
1380[14:37:18] <brokencycle> I'm on Buster, btw.
1381[14:38:06] <_ioannis> Lady_Aleena: join a pet group and ask
for ideas there I'm guessing ? I don't currently even own
a pet :(
1382[14:38:06] <brokencycle> But the problem has been persistent
ever since I tried to use gmtp, probably in Lenny even.
1383[14:38:52] <Lady_Aleena> _ioannis, you don't own pets,
pets own you. 8)
1384[14:39:03] <brokencycle> If I switch to a different virtual
screen, then switch back, gmtp sort of starts over, as if I first
invoked it.
1391[14:43:01] <Unit193> Wow, there's a new version of gmtp
but no viable changelog, news file, nor release
announcement...Can't tell if the version in testing might have
a fix..
1392[14:44:03] <Unit193> Oh! It's just that old, cool. No,
seems to be just French translation updates.
1413[15:01:26] *** Quits: pyfgcr (~pyfgcr@replaced-ip) (Remote host closed the connection)
1414[15:01:34] <Eryn_1983_FL> hey guys im running lxd/lxc on my
new debian buster and i cant get samba to mount properly, works on
all my other hw systems and my old stretch lxc os
1441[15:13:24] <rocketmagnet> i have compiled python3.7 on my own
and now have remove all files that are related to python3.7, how can
i make dpkg or apt reinstall python3.7 ?
1454[15:19:15] <rocketmagnet> i compiled python on my own and
delete all python files and now apt thinks it's still
installed, how can i tell apt that it's not installed anymore
so i can reinstall it ?
1631[17:25:51] <bernyrd> hey has anyone used switched PCIe to
NVMe adapter? obviously there is multiplexing involved so lower
bandwidth, but I am unsure if this cuts into drive speed
1635[17:26:37] <bernyrd> like you can get x16 to 4x NVMe drive,
and it is one x4 per NVMe, but there are few company that sell x8 to
4x NVMe (as opposed to usual 2)
1636[17:26:44] <bernyrd> has active silicon
1637[17:27:16] <LtL> ychaouche: you must have grub-legacy, mine
grub installs are shown here,
replaced-url
1638[17:28:03] <LtL> ychaouche: if it boots i wouldn't worry
about it
1662[17:35:19] <LtL> VG234: asking in #ubuntu would be wise
1663[17:35:39] <VG234> I'm on debian.
1664[17:35:42] <asmos01> Welcome. I have a problem. I migrated
from Ubuntu to something lighter. I have Debian Stable. I read about
backports. I would like newer versions of programs. Can you give me
the name of the repository for this? Because on the Internet there
are different names for these packages.
1708[17:50:41] *** Quits: krabador (~krabador@replaced-ip) (Remote host closed the connection)
1709[17:51:01] <VG234> OK. I remove the call for it. My update
now at least passes. My upgrade at least has no errors: 0 upgraded,
0 newly installed, 0 to remove and 0 not upgraded.
1710[17:51:29] <VG234> Is there anything I should check? Also, I
use sudo with apt-get, should I be doing that?
1729[18:02:15] <VG234> trui: Basically, I'd like to have it
work the same or better than my Cygwin under windows7. It's
seems to be better in some ways. Currently, I'm working with
Json and Yaml doing DevOps work. That is why I want the yq tool.
This is what broke it:
replaced-url
1741[18:05:54] <VG234> LtL: I have that, but it's not the
same. It doesn't slice and dice yaml.
1742[18:06:12] <VG234> LtL: at least, I don't think it does.
1743[18:06:55] <trui> ah, i was wondering due to the above. but
yeah, debian and ubuntu are different. i think ubuntu has been
pushing snaps more lately iirc
1746[18:08:49] <VG234> trui: The things I need to do is to be
able to train myself across DevOps tools and cloud environments. I
plan to do this using python, terraform, aws, etc.. Each brings with
it the need to learn and use different sub tools.
1757[18:12:09] <VG234> As a last question: I've basically
shut off some packages such as yq and virtualbox in their
source.lists. Should I follow that pattern instead of trying to
remove them? Sometimes, I can't remove packages.
1762[18:13:33] <VG234> somiaj: Yes. That is why I asked about
snap.
1763[18:13:39] <somiaj> virtualbox has an okay repo that is
compadable with debian, so that would be fine, but ppa's shoudl
be avoided.
1764[18:14:11] <somiaj> I personally don't use snap or
flakpack or any of these tools, I really try to stick to software
that is in debian, it does contain most things one will want, and
compile stuff from source that isn't tehre.
1765[18:14:20] <VG234> somiaj: 👍
1766[18:15:14] <somiaj> linux distros aren't like windows
were one can jsut download random software from random sites and
expect it to be compadable and work.
1767[18:15:44] <somiaj> snap/flatpak are ways to try to improve
this, but I haven't found a need for such things, but my
software needs are simple compared to most (I don't even use a
DE)
1769[18:16:46] <VG234> somiaj: That can be an lengthy path. From
the unix book I have, Porting Unix software it states something
like, someone comes into your office and ask you to port something
to another OS, "Obviously, killing them is out of the
question". This always make me laugh, but the amount of work is
true.
1770[18:17:56] <somiaj> I wans't talking about porting here,
I was just talking about downloading random binaries that run on the
linux kernel
1778[18:19:49] <VG234> somiaj: If they work and the tools are
supported on the system, yes.
1779[18:20:17] *** Quits: conta (~Thunderbi@replaced-ip) (Quit: conta)
1780[18:20:40] <somiaj> often the biggest issue with ranodom
binaries is linux uses the shared libary approach, vs static linked
liabaries. That is what flatpack/snap does, but it makes binaries
way larger
1782[18:21:05] <somiaj> so if you compile it against the libaries
you have it often works (though depends on the actual software,
really well written software works well)
1793[18:23:45] *** Quits: VG234 (~user@replaced-ip) (Remote host closed the connection)
1794[18:24:25] <trui> i am curious if you can upgrade to debian
10 on windows 10, but i don't have windows and couldn't be
bothered to install it anyways
1810[18:31:54] <somiaj> I personally like vms, but I guess WSL is
for those who need to develope stuff on linux but don't need a
full vm while using windows.
1826[18:35:57] <laptop> I wish there was a rolling release debian
1827[18:36:32] <trui> have you tried debian sid/unstable?
1828[18:37:17] <laptop> no has anyone tried it and did it work
from stretch to buster
1829[18:37:23] <trui> it's certainly rolling enough. too
rolling for lots
1830[18:37:39] <somiaj> !moving target
1831[18:37:39] <dpkg> Sure, testing might be shinier than stable,
but are you prepared to be continually updating your system? Things
that worked today will break tomorrow. Configuration file formats
will change and you'll have to fold your changes in yet again.
Testing is a moving target and if you'd rather work *with* your
computer than working *on* your computer, you might not want that.
See <testing security>.
1832[18:37:56] <somiaj> It really isn't a release, but it is
constantly being upgraded, with all new bugs included.
1833[18:38:31] <somiaj> and many people use testing/sid, and if
you use it, learn to use the BTS and report bugs, it is users of
testing/unstable that help prepare debian for its release
1932[19:29:40] <dpkg> From 2006 to 2016, Mozilla Firefox was
known as "Iceweasel" in Debian because Mozilla would not
permit Debian to use the Firefox name. Firefox packages are now
available for both jessie and sid; Firefox in Debian <stable>
follows the Extended Support Release (ESR) branch. For the latest
versions, ask me about <mozilla.debian.net>.
1937[19:33:41] <dpkg> From 2006 to 2016, Mozilla Firefox was
known as "Iceweasel" in Debian because Mozilla would not
permit Debian to use the Firefox name. Firefox packages are now
available for both jessie and sid; Firefox in Debian <stable>
follows the Extended Support Release (ESR) branch. For the latest
versions, ask me about <mozilla.debian.net>.
1938[19:33:54] <GSock14> sdfds
1939[19:35:34] *** hwm4rgs7 is now known as hwm4rgs
1945[19:40:58] <dpkg> From 2006 to 2016, Mozilla Firefox was
known as "Iceweasel" in Debian because Mozilla would not
permit Debian to use the Firefox name. Firefox packages are now
available for both jessie and sid; Firefox in Debian <stable>
follows the Extended Support Release (ESR) branch. For the latest
versions, ask me about <mozilla.debian.net>.
1984[20:08:03] <Gigglebyte> Is there a way to expand a separator
so that it expands all the way across instead of just a segment of
the bottom menu bar in xfce4? I recall in the past there was a way
to adjust how much a separator could be expanded, and now I have to
use three separators to move things over. It takes longer to load,
and I am already on a laptop that is a bit short in horsepower.
1985[20:08:25] <Gigglebyte> I am on Debian Stable Stretch.
1994[20:13:00] <primvibes> I apologize for taking advantage of
the excellent debian community to solve an issue on kali-rolling,
but my situation is desperate. I'm in a remote location, my
phone broke yesterday and today something went wrong during
dist-upgrade, which rendered the system unusable. The only tools I
have is on recovery mode: logs, lynx for internet, irssi and tmux.
Telinit can only bring the system up to runlevel
1995[20:13:06] <primvibes> 3. I'm only an intermediate user
and cannot see anything wrong in the logs, not that I understand
much about video drivers. Regular boot gives a black screen with a
mouse cursor and Alt+Fn2-12 doesn't work. I've tried
reconfiguring all packages and reinstalling the linux headers and
images, with no success. Please help
2019[20:23:47] <jhutchins> primvibes: We'd really rather you
didn't. You'll probably have to repeat whatever steps you
did for the initial installation, but for assistance and support
you're going to have to rely on the slower response and smaller
community of kali supporters.
2020[20:23:59] <jhutchins> primvibes: You will probably want to
try their mailing list.
2119[21:09:11] <Gigglebyte> how do I adjust the sleep settings in
debian 9 stretch? My laptop logs off quite quickly due to
inactivity. I am looked online, but only find info on debian 7, 8,
and 10.
2120[21:10:18] *** Quits: Adbray (~Adbray@replaced-ip) (Remote host closed the connection)
2192[21:40:06] <CarlFK> I'm adding docs to gst's build
instructions. is this an OK way to add to PATH? printf
'\nexport PATH=${PATH}:~/.local/bin\n' >> ~/.bashrc
2204[21:43:46] <rwp> It's a problem. Because there are so
many hackers on Windows and Mac OS and they do not have good package
managers. So upstream communities often disparage downstream
packages.
2207[21:45:00] <rwp> On the good side pip --user should only
touch $HOME and nothing else. So the system is safe from it. On the
bad side people forget they have done that and then later we are
debugging why their system has some old package being used instead
of the one installed.
2208[21:45:25] *** Quits: gry (~sveta@replaced-ip) (Ping timeout: 268 seconds)
2244[22:02:51] <primvibes> madage: you're a lifesaver, the
workaround was a success. In future I'll need to pay closer
attantion to debian testing bugs, since that's what kali is
based on.
2270[22:16:08] <Encrypt> I'm currently struggling to make
NFS (v4) work between a Raspberry Pi and my Debian 9 computer, with
differet UID/GIDs (therefore, with idmapd)
2297[22:26:38] <blastwave> The usual odd problem wherein after a
reboot there is audio for some applications and none for Firefox.
The tool 'Pulse Audio Volume Control' seems to not help
here at all and the alsamixer also does not help. Does pulseaudio
need a reset or what could be the issue ??
2302[22:28:31] <blastwave> oh also .. screensave locks up the
whole XFCE desktop so badly I needed to hook up a reset button to
the motherboard front panel .. that can't be right
2321[22:35:43] <rwp> blastwave, I sometimes have to tell Firefox
video to un-mute. I think it saves muted as a default sometimes.
2322[22:36:39] <blastwave> hrmmm .. let me check
2323[22:36:43] <rwp> That screensaver problem sounds nasty. I
would be disabling the screensaver as quickly as I could! (I
don't use them by default. I just blank the screen here.)
2333[22:39:02] <rwp> Curious... What graphics card are you using?
2334[22:39:35] <blastwave> NVidia Quadro K4200 and nvidia-smi
shows the driver has not changed since last apt-get update or reboot
.. seems to be working greak
2335[22:39:38] <blastwave> gret
2336[22:39:44] <blastwave> great <-- spelling is an issue for
me
2339[22:40:06] <blastwave> damn .. I have no idea how to
'unmute' firefox which should not be muted
2340[22:40:36] <rwp> If the kernel upgrades and there isn't
a reboot then I have had random problems with modules that are then
mismatched between the running older kernel and the newer on disk
files.
2342[22:40:54] <blastwave> am trying to test with my own stream
from twitch
replaced-url
2343[22:41:22] <blastwave> hrmm kernel mismatch on the kernel
modules for nvidia ? sounds unlikely
2344[22:41:24] <rwp> Un-mute example: Play a youtube video. Mouse
around the video window. The audio is muted. Un-mute. That's
what I find I sometimes need to do. Even when *I* haven't muted
it.
2345[22:41:35] <blastwave> I am using the debian packages here
and not the drivers from NVidia
2355[22:43:02] <rwp> Good on using the Debian packages. But that
may still be out of sync version wise after a package upgrade and
before a reboot to the new kernel.
2356[22:43:08] <blastwave> there were a lot of people watching
live as I destroyed a hard disk while a machine was runnng ...
saturday evening fun
2394[22:50:15] *** Quits: f476 (~f476@replaced-ip) (Remote host closed the connection)
2395[22:50:28] <blastwave> okay .. get this ... I start up VLC
and get Star Trek running ... playing and it sounds fine .. the
pulse audio control shows VLC as a valid application
2397[22:50:55] <rwp> if you suspect that it might be a file in
$HOME somewhere... then you can always create a new test user and
log in as that pristine new test user and see if the pristine files
work. If so then you know it is is some config file in your $HOME
causing the problem.
2404[22:52:55] <blastwave> SerajewelKS: you are looking to
transfer a filesystem from one disk to another and only grab the
data ?? See Jorg Schillings 'star'
2405[22:52:59] *** Quits: dtux (~dmtucker@replaced-ip) (Remote host closed the connection)
2407[22:53:14] <rwp> SerajewelKS, Opens up the box of Pandora and
mentions that ZFS will sync and being a file system it knows exactly
which blocks need to be transferred. Really very cool.
2408[22:53:16] *** Quits: pyfgcr (~pyfgcr@replaced-ip) (Remote host closed the connection)
2409[22:53:34] <blastwave> well ZFS is awesome of course
2410[22:53:42] <rwp> Or do you mean sparse files?
2411[22:53:46] <blastwave> however he is on ext4fs from the
distant dark path
2412[22:54:15] <rwp> I did say I was reaching into "the
box". And down that path we all go one day... :-)
2418[22:55:26] <SerajewelKS> the context is migrating an existing
volume into luks. as a security consideration, the device luks is
used on should be random-filled before adding a filesystem.
obviously this only affects the unused space of a filesystem that
has been used. so the logical way to do it would be: random-fill the
underlying device, create the luks container, copy only the used
blocks into the container.
2422[22:55:55] <SerajewelKS> however, i just realized that
i'm using pvmove so that doesn't matter; lvm doesn't
have a mechanism to do that anyway.
2423[22:55:57] <blastwave> I have no idea what " luks "
is
2424[22:56:05] <SerajewelKS> !luks
2425[22:56:05] <dpkg> Linux Unified Key Setup (LUKS) is a disk
encryption specification with a platform-independent format.
replaced-url
2426[22:56:23] <blastwave> Linux Unified Key Setup ?? cool
2427[22:56:27] <rwp> Nice blastwave. Two side by side monitors.
But only 1080 vertical pixels is too short for me. I would rotate
both 90 degrees! :-)
2428[22:56:36] <blastwave> " platform-independent format
" even better
2429[22:56:43] <SerajewelKS> blastwave: if pulse does not notice
firefox then the logical conclusion is that firefox is not talking
to pulse
2430[22:56:54] <blastwave> rwp: well a rotation on one of the is
not easy
2431[22:57:03] <blastwave> it is a PQLabs massive touch screen
2432[22:57:11] <blastwave> the other is a normal 24inch monitor
2433[22:57:25] <blastwave> SerajewelKS: k .. and what next ?
2434[22:57:49] <SerajewelKS> blastwave: hmm, you'd have to
figure out what firefox is using. my guess would be alsa, and that
alsa is not configured to use pulse.
2435[22:58:17] <rwp> My solution is usually to purge all pulse
audio from the system. And then everything Just Works.
2438[22:58:27] <SerajewelKS> you could test this theory by using
something else that only talks to alsa (or can be configured to)
2439[22:58:48] <SerajewelKS> e.g. if you run a command-line ALSA
client with no device switches, does it show up in pavucontrol?
2440[22:58:57] <SerajewelKS> aplay for example
2441[22:59:20] <rwp> SerajewelKS, If you are doing pvmove then I
think internally it will copy the entire size of the source
partition no way to avoid it. Pretty sure.
2442[22:59:21] <SerajewelKS> if it does not then the likely
explanation is that firefox is using alsa and alsa is not using
pulse
2443[22:59:33] <blastwave> wow ... I killed FF and then restarted
it .. *now* even youtube video doesn't work
2444[22:59:37] <SerajewelKS> rwp: right that's what i meant
by "lvm doesn't have a mechanism to do that anyway"
2445[22:59:44] <blastwave> time to go back to normal release
firefox to test
2451[23:00:32] <rwp> SerajewelKS, I do like LVM and pvmove. It
makes upgrading disk sizes to larger disks very simple. All on the
fly while the system is running too. FTW! :-)
2452[23:01:02] <SerajewelKS> rwp: so this would mean: failing a
disk from the raid1, creating a new (degraded) raid1 on that disk,
creating a luks container on the new raid1 with a detached luks
header in /boot (so that the header doesn't eat space,
otherwise i have to shrink the volume which can't be done
online)
2454[23:01:32] <SerajewelKS> rwp: add the container as a PV to
the VG, pvmove everything off of the old PV, tear down the old
raid1, put the disk from that mirror into the new mirror
2455[23:01:52] <SerajewelKS> obviously there is no redundancy for
the duration
2456[23:02:11] <rwp> Question: Can you add an additional disk?
(Avoids the need to degrade the raid and then no need for the
detached header.)
2457[23:02:39] <SerajewelKS> the additional disk would prevent
degredation but would not alleviate the need for detached header
unless you are suggesting reshape to raid5
2459[23:03:18] <rwp> I work with lvm quite a bit but only work
with luks and encrypted volumes using the debian-installer to create
them on installation.
2460[23:03:24] <blastwave> also going to blow away the FireFox
profile on disk
2461[23:03:32] <rwp> save your bookmarks
2462[23:03:51] <SerajewelKS> if keeping it as raid1 then the
final device size is still the same
2463[23:03:57] <SerajewelKS> you'd need some extra KBs for
the header
2464[23:04:22] <rwp> I will take your word for it. I haven't
worked with that part of things.
2465[23:04:33] *** Quits: Guest95420 (~iulian@replaced-ip) (Remote host closed the connection)
2466[23:04:36] <rwp> Your overall plan sounds good to me. Seems
well thought out and should work.
2468[23:04:54] <SerajewelKS> well the concept is simple: luks
needs to store metadata somewhere. it usually stores it in the
underlying device, which obviously means the mapped container is a
bit smaller than the existing device.
2469[23:05:09] *** Quits: todd_dsm (~todd_dsm@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
2470[23:05:09] <SerajewelKS> so adding luks reduces the size of
the device by the amount needed to store the header
2472[23:05:52] <rwp> Right. So I would make the new partition
slightly smaller in order to free up space. Then pvmove onto it. But
I almost never allocate 100% of a VG.
2475[23:06:07] <SerajewelKS> right that's the problem is
that this PV is 100% allocated
2476[23:06:13] <blastwave> wow .. stable FireFox
"Quantum" normal download does not work at all .. no
youtube video and no audio
2477[23:06:14] <SerajewelKS> the other PV isn't redundant
2478[23:06:17] <blastwave> nice
2479[23:06:18] <rwp> And if I did then I would *temporarily* add
an additional disk to make up the extra. Then compact back down to
the right size as an additional step.
2480[23:06:34] <SerajewelKS> "compact back down"
requires downtime to resize the ext4 volume
2486[23:07:08] <SerajewelKS> i'll just steal some swap space
for the luks header
2487[23:07:15] <rwp> I cannot recommend ever downsizing an ext*
volume. I once tried that and it took more than 10 days (12 days?)
to run to completion.
2488[23:07:33] <SerajewelKS> yep that's why i wanted to
avoid it by using detached header but i'll just shrink swap
instead
2489[23:07:48] <rwp> I let it run and it did eventually finish. I
imagine some exponential algorthm that could be optimized inside
there somewhere.
2490[23:07:50] * blastwave thinks .. use ZFS
2491[23:08:17] <SerajewelKS> i'd love to temporarily have a
third disk for redundancy but i don't have a 6TB sitting
around. nor any combination of disks i could combine into 6TB
2492[23:08:18] <rwp> Now I copy everything off to a new file
system file by file.
2493[23:08:36] <SerajewelKS> blastwave: i'll add a tick in
the box that counts the number of times that's been suggested
to me, thanks
2506[23:10:11] *** Quits: jnavila (~jnavila@replaced-ip) (Remote host closed the connection)
2507[23:10:14] <SerajewelKS> but generally yeah not gonna convert
an entire system to a new filesystem just because it might make
things easier. that's exactly the kind of project that
isn't easy to begin with.
2509[23:10:34] <rwp> I am in a position now with a system with
only 2 SATA ports total and 2x 1T drives. I think for that system
upgrade I am going to shutdown, pull both disks, install in a 4-port
SATA system, add two new disks, partition, mdadm raid, pvcreate,
pvmove, everything else is online, move data to the new disks. Then
shutdown and shuffle the disks back into the original 2-SATA system
again.
2510[23:11:07] <SerajewelKS> rwp: if the SATA controller is
hotswap you can do it online if you don't mind degredation.
assuming you have a mirror right now.
2511[23:11:21] <SerajewelKS> this is how i grew this system from
3TB to 6TB
2512[23:11:29] <rwp> Love the hot swap SATA capability.
2513[23:11:32] <SerajewelKS> yep
2514[23:11:35] *** Quits: BlueByte (~walther@replaced-ip) (Quit: This computer has gone to sleep)
2515[23:12:22] <SerajewelKS> fail drive 3TB A, swap in drive 6TB
C. add C to mirror, wait for resync. fail drive 3TB B, swap in drive
6TB D. add D to mirror, wait for resync. tell mdadm to resize the
array. resize2fs.
2516[23:12:23] <rwp> But moving from 2x 1T disks to 2x larger
disks (TBD) and would prefer not to need to break the raid while
doing it. And I have a 4-port SATA system available for the use. So
might as well make my life easier and use it.
2517[23:12:36] <SerajewelKS> with the extra step of growing the
LV anyway
2518[23:13:01] <SerajewelKS> if you have a 1TB USB drive you
could use that temporarily too. flag it write-mostly.
2519[23:13:25] <rwp> SerajewelKS, Sounds like a plan. It's a
Towers of Hanoi process but works within the system you have. Good
deal.
2520[23:13:28] <blastwave> rwp: fyi --> Alsa support was
dropped starting Firefox 52.0 and later.
replaced-url
2521[23:13:34] <SerajewelKS> assuming its actual capacity is
really >= the rest of the drives
2522[23:13:55] <somiaj> I belive debian's packages still
contain alsa support, it was just mozilla's package that droped
that build config
2523[23:14:04] <rwp> Yes. I could add a USB disk. However I have
not had the best of luck with USB drives for some reason. Probably
my proclivity of buying cheap crappy USB adapters. So probably
self-inflicted. But that makes me shy away from them.
2524[23:14:08] <blastwave> ahhh .. good to know
2525[23:14:08] <somiaj> AT least it was that way for a while,
unsure if it has changed.
2526[23:14:09] <rwp> External eSATA is perfect though.
2529[23:14:31] <SerajewelKS> rwp: i have a couple of
odroid-hc2's that i can use in cases like this, when a system
has no extra drive bays but i need to add a drive either temporarily
or permanently. throw a drive in the hc2 and export it over iscsi.
2531[23:15:10] <SerajewelKS> rwp: FWIW my backup volume is a set
of mirrored 2TB USB drives. haven't had any issues. raid has
been intact since i set it up months ago.
2532[23:15:26] * blastwave sees that Debian release FireFox ESR and
that will work fine !
2533[23:15:28] <Maws> hi!, is there any way to get this mod
loaded: "snx: loading out-of-tree module taints kernel.
2536[23:15:30] <rwp> SerajewelKS, That sounds cool! I should look
into something like that.
2537[23:15:58] <SerajewelKS> rwp: iscsi is supported by linux
(open-iscsi in debian) and, i didn't realize until recently,
windows 10 has a built-in iscsi client as well. i figured that would
be a server feature.
2539[23:16:20] <rwp> I had a dual disk USB adapter that I used
for a while. But it would drop offline every so often. In a 3 month
window many twice. Never stayed online more than 4 months.
2540[23:16:55] <SerajewelKS> rwp: it makes all kinds of options
available. i really hate windows' idea of mirrored disks. so i
can do the mirroring on a debian system and export the raid device
over iscsi. windows doesn't know or care about the raid.
2541[23:16:57] *** Quits: openface (~of@replaced-ip) (Remote host closed the connection)
2542[23:17:04] *** Quits: CtrlC (~CtrlC@replaced-ip) (Remote host closed the connection)
2543[23:17:05] <rwp> On the other hand I have had really good
luck with USB network adapters. I expected them to be trouble. But
actually they have been very reliable when I have used them.
2545[23:18:11] <rwp> blastwave, somiaj, I am using Debian's
Firefox 60 esr with alsa okay and everything works. But good to know
if it was dropped upstream.
2553[23:19:58] <SerajewelKS> rwp: in my case i'm using it to
add storage for a machine i use as a hypervisor for qemu-kvm VMs.
it's particularly nice because qemu supports iscsi as backing
for a virtual disk.
2555[23:20:28] <SerajewelKS> rwp: so i don't even have to
configure iscsi on the host machine. i just add a line to the VM
launch script to configure the iscsi disk. the VM doesn't know
or care that it's remote.
2562[23:22:18] <rwp> SerajewelKS, Thanks for sharing it now.
There many things like that. Gems of tools that are really useful
but not something learned all at once.
2563[23:22:55] <rwp> Talking about ssh tunnels and gems of
tools... Have you used 'sshuttle'? It's one of those
I picked up for creating quick and simple vpn tunnels.
2564[23:23:09] <trui> blastwave: are we playing with sid in
#debian? *teases*
2589[23:27:35] <rwp> SerajewelKS, Now I show up at a corporate
site with a restricted firewall and http proxy but has ssh
available. I launch "sshuttle -v -x 10.0.0.0/8 -r myservername
0/0" and then have full private web access without triggering
their corporate http proxy restrictions.
2590[23:28:08] <SerajewelKS> rwp: nice, i did something similar
but much more involved
2591[23:28:09] <rwp> Where 10/8 represents the local subnet. That
is excluded. I can still access local subnet IPs. But all others get
tunneled.
2593[23:28:49] <rwp> Needs local sudo root to set up the local
iptables. Does not need root on the remote system. Works very well.
2594[23:29:01] *** debhelper sets mode: +l 1550
2595[23:29:06] <SerajewelKS> rwp: it involved running a linux VM
on windows, bridging the physical network to the VM but leaving the
bridge interface on the host unconfigured, adding a host-only
network interface to the guest, and having the guest route it
2597[23:29:22] <SerajewelKS> rwp: the guest could be used to
selectively route traffic out the real network or a VPN
2598[23:29:30] <blastwave> oh bloody hell .... the released and
packaged Debian FireFox ESR ... *just works* .. well gee
2599[23:29:48] <rwp> An advantage over -D is that I assume with
-D you configure things to use the proxy? With sshuttle nothing
needs to be configured because it uses iptables and therefore
everything Just Works.
2601[23:30:21] <rwp> Isn't that the way things should be
blastwave? Just Working? :-)
2602[23:30:24] <SerajewelKS> rwp: right. i use -D sometimes
though because in some cases i want only some programs to be
proxied. you could do the same thing in iptables with -m owner or
something
2603[23:30:56] <SerajewelKS> ssh -w gives you a tun interfaces.
which again is, i assume, how sshuttle works.
2604[23:31:05] <rwp> In any case I was mostly just trying to
repay the favor of pointing me to open-iscsi with another gem of a
tool that I use a lot but no one knows about. :-)
2605[23:31:06] <SerajewelKS> it's probably ssh -w but sets
up the iptables rules for you.
2606[23:31:15] <SerajewelKS> yeah i might have the occasion to
use that
2607[23:31:16] *** Quits: hele (~hele@replaced-ip) (Ping timeout: 248 seconds)
2608[23:31:28] <SerajewelKS> it's always good to learn about
new things
2609[23:31:44] <SerajewelKS> lately for me that's been
docker/podman
2610[23:32:09] <rwp> A good hacker is always learning. [[ Pilots
have a saying, "A good pilot is always learning." :-) ]]
2624[23:37:22] <SerajewelKS> rwp: speaking of new stuff i've
found, i also do some audio engineering work mostly for fun. imagine
my surprise when i found ardour and jack. i'd known about jack
for years but not had a use for it.
2633[23:39:20] <SerajewelKS> i've used a multiband
compressor on a conference call to compress the snot out of the high
bands because one of the ladies on the call had a really piercing S
that was actually physically hurting my ears if i had it loud enough
to where i could understand everyone else
2634[23:39:46] <SerajewelKS> and i sat through it for five
minutes before realizing that i had the solution staring at me on my
other monitor
2635[23:40:35] *** Quits: dtux (~dmtucker@replaced-ip) (Remote host closed the connection)
2645[23:44:21] <jhutchins> Heh. I tried putting my guitar through
jack and some multi-effects thing and it was like a tape delay.
Amusing but not useful.
2646[23:44:37] <jhutchins> It's an old Core2 duo with a
soundblaster.
2647[23:44:56] <SerajewelKS> hmm. depending on how many effects,
that could have worked.
2648[23:45:03] <SerajewelKS> maybe the period size was too large?
2649[23:45:17] <SerajewelKS> i've managed to get an rpi
processing several channels with comp+eq
2650[23:46:12] *** Quits: Christian75 (~Christian@replaced-ip) (Remote host closed the connection)