80[01:07:32] *** Quits: frostschutz (~frostschu@replaced-ip) (Remote host closed the connection)
81[01:12:49] *** Quits: Jad (~Nashmi@replaced-ip) (Quit: Benefits I derive from freedom are largely the result
of the uses of freedom by others, and mostly of those uses of
freedom that I could never avail myself of.)
147[02:54:54] <sney> H-var: most system administration stuff is
root-only by default, though you may be able to add your user to a
group that allows it. for instance, 'systemd-journal' is
the group for being able to read system journalctl stuff as a
regular user. there might be one for ufw too.
148[02:55:01] *** interd0me is now known as interdome
176[03:16:39] *** Quits: Tobbi (~Tobbi@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
177[03:18:39] <scruloose> In case anyone's curious about
the Docker filesystem-access-as-root thing I was freaking out about
a day or two ago, I was indeed missing something fundamental: Docker
does indeed allow containers to access the filesystem, as root on
the host, but only to paths you "explicitly" share with
that container.
178[03:18:42] *** Quits: jotaxpe (jotaxpe@replaced-ip) (Remote host closed the connection)
181[03:19:37] <scruloose> I put "explicitly" in
scare-quotes because if you use docker-compose, it'll
cheerfully accept paths defined in .yml files in your pwd as being
"explicitly" shared.
182[03:21:11] <petn-randall> scruloose: docker is not meant to
be used as isolation feature security-wise. A normal docker user has
not only root on all containers, but also on the host itself.
183[03:21:39] <petn-randall> IIRC root in the container also
mean root on the host.
186[03:22:14] <scruloose> So adding a user to the docker group
effectively gives that user full root powers (they can trivially get
a root shell with host system's root dir mounted read/write),
without even sudo's password prompt.
188[03:24:12] <scruloose> New question: is there a way to delete
a group on Debian and prevent it getting recreated by later package
installs/upgrades?
189[03:24:23] *** Quits: Newami (~Newami@replaced-ip) (Remote host closed the connection)
190[03:25:34] <petn-randall> yes; no
191[03:25:36] <petn-randall> !xy
192[03:25:36] <dpkg> Slow down for a bit! Are you sure that you
need to jump through that particular hoop to achieve your goal? We
suspect you don't, so why don't you back up a bit and tell
us about the overall objective... We know that people often falsely
diagnose problems because they are too close to them -- it's
easy to miss that there is a better way to proceed. See
replaced-url
194[03:26:43] <scruloose> I gather that docker does offer *some*
meaningful degree of security isolation in terms of controlling what
the thing *inside* the container can do on the host, but only to the
extent chosen by the user setting up/running the container... yes?
204[03:34:08] <scruloose> From my not-very-deep reading in the
last couple of days, I believe there has been some progress on that
front in the meantime, but yeah, it sure isn't security-first
design.
209[03:40:45] <petn-randall> Their goal is to provide a stable,
reproducible environment, that is easily transferable, to run
binaries in. Security is not on their list.
210[03:41:41] <pasiz> firewall is firewall, application server
is application server...
225[04:05:53] <noisemaker> I just installed the package
`webkit2gtk-driver` but compiling some Vala code I get: fatal error:
webkit2/webkit2.h: No such file or directory
226[04:06:17] <sney> !headers
227[04:06:17] <dpkg> Trying to compile something from source,
and the build script says it can't find a dependency you are
sure you have installed? You need header packages, which have the
-dev suffix. For instance, if it thinks python3 is missing, install
'python3-dev'. See <build-essential>.
228[04:06:56] <sney> you can also use apt-file to find out what
package owns a specific file.
229[04:08:49] <noisemaker> sney: hum, there is no -dev package
for me, could be part of the non-free ?
230[04:09:03] <sney> you can also use apt-file to find out what
package owns a specific file.
231[04:10:28] <sney> !apt-file
232[04:10:29] <dpkg> apt-file is a utility used to find which
package owns a given file - or list files in a package - even if it
is not installed: "aptitude install apt-file &&
apt-file update" ; then "apt-file search filename" to
see a list of packages containing filename (can also use a regular
expression; see man apt-file). You can also search online, ask me
about <pdo>. See also <judd-file>, <auto-apt>,
<dlocate>.
233[04:12:17] <oxek> can apt-file search for things in
buster-backports too though?
234[04:13:01] <sney> it uses the Contents file on the mirror, so
as long as the mirror has one, it'll work normally
248[04:49:20] <EdePopede> seems i found out the identity of the
bad guy on my buster install. it is a "give me everything you
have" install (still DVD1 only, nice), then i logged in to
xfce, made my changes, logged out. which left me back with
TWENTY-EIGHT processes running as that user.
249[04:50:39] <EdePopede> there was a systemd --user which had a
gfvsd subtree and some individual processes. started killing them
one by one, when i reached the last one finally everything including
the whole systemd army disappeared
260[04:53:30] <EdePopede> the next thing is, returning to the
desktop didn't bring back the panel. but i'll look for
this one later. for now i'll see if i can reproduce this one.
261[04:54:46] <EdePopede> and the keypad wasn't working at
all in mc's subshell on tty. looked into the settings dialogs,
even checked the keypad keys in the one, didn't save, but when
i returned they were there. another mystery.
268[05:06:06] <EdePopede> gnome, seriously. the project channel
is on irc.gnome.org which i have no idea if that's freenode or
not, if it is, there's only 2 people sitting there and it
doesn't even have a topic. then i wanted to check their hp for
that info, only the gnome logo top left corner only points to the
discourse start page. discourse, which is nothing but phpbb in shiny
modern colors.
269[05:06:56] *** Quits: xompax (uid200197@replaced-ip) (Quit: Connection closed for inactivity)
270[05:07:17] <EdePopede> they have a really poor idea of
usability and interface design for folks who have the idea of
building the linux desktop of the future.
279[05:14:51] <EdePopede> kline\0: yep, gimpnet. i remember
i've been there years ago. but that's info that really
could be given when they tell people to connect to irc://whatever
since using different domains on the same network isn't that
uncommon.
280[05:14:51] <kline\0> gnome sucks for a lot of reasons but i
dont think this is one of them
282[05:17:44] <EdePopede> back then i was refused to connect to
the network. took me some time to find out that i was already on it
using my default nick. there was no hint at all that it was that
one. yes, this sucks if you're trying to reach someone helping
you out of a problem and all you get is "you stay out."
441[09:43:15] <AimHere> Stealth upgrade decided to upgrade a
library to a shiny new version while I was actually using it. One
minute my application works, next minute it's broken
470[10:02:01] <oxek> I'm just following reverse
dependencies, where I see that unattended-upgrades is recommended by
python3-software-properties, which is a dependency of
software-properties-common/software-properties-kde/software-properties-gtk
472[10:02:36] <oxek> and those packages are unlikely to be
installed outside of initial installation, and very unlikely to be
installed manually by the user
473[10:03:07] <AimHere> Yeah, the dpkg log shows it being
installed at about the same time as a bunch of curl-related packages
+ python3-software-properties + apt-transport-https and related
stuff
474[10:03:19] <oxek> I know for a fact that unattended-upgrades
does not get installed on a minimal installation without a desktop
environment, and not as part of xfce either
475[10:03:45] <oxek> apt-transport-https? What debian version
are you on? That package should no longer exist.
476[10:03:51] <AimHere> sid
477[10:04:07] <AimHere> This update happened in November
478[10:04:14] <AimHere> I haven't noticed until now!
479[10:04:23] <oxek> yeah, sid is not supported in this channel,
and all sorts of crazy things can happen in sid
480[10:04:28] <oxek> !debian-next
481[10:04:28] <dpkg> #debian-next is the channel for
testing/unstable support on the OFTC network (irc.oftc.net), *not*
on freenode. If you get "Cannot join #debian-next (Channel is
invite only)." it means you did not read it's on
irc.oftc.net. See also
replaced-url
482[10:04:42] <AimHere> I'm not asking for support
483[10:04:45] <AimHere> Just venting
484[10:05:11] <oxek> probably no point venting about sid, it
might be configured to use unattended-upgrades so that you always
get upgrades asap
485[10:05:36] <oxek> and "One minute my application works,
next minute it's broken" is behavior I'd expect on
sid
487[10:06:17] <AimHere> My proximate problem is that this rotten
library is broke, and not only is it so new nobody's filed a
bug report, I'm not qualified to file one. It's a library
called by a library called by an app I'm using and I don't
have a clue what the hell it even is
488[10:06:40] *** Quits: timahvo1 (~rogue@replaced-ip) (Remote host closed the connection)
489[10:07:00] <oxek> last week gnutls library broke on sid,
breaking everything. That's typical behavior of sid.
496[10:09:11] <AimHere> Yeah, I know. My main reason for using
sid was that a) Ubuntu wouldn't install when I tried and b)
stable's software was out of date. And sid isn't terrible,
most of the time
505[10:11:48] <oxek> all I'm saying is I definitely would
not expect stable behavior (stable = not changing at random times,
as happened to you) from non-stable versions of debian.
506[10:13:14] <oxek> Lope: 'ip route show'
507[10:13:20] <AimHere> Oh yeah, I know about sid, and unstable.
It's just that until now 'unstable' has meant a bit
more stable than 'app suddenly stops working between successive
invocations'
508[10:13:25] <Lope> oxek, that's not ipv6??
509[10:13:49] <Lope> nevermind: ip -6 route show
510[10:14:02] <oxek> Lope: yeah, that one
511[10:14:31] <oxek> AimHere: on debian, "stable" does
not mean "does not crash" it means "does not
change"
512[10:14:51] <oxek> hence "unstable" means
"changes all the time"
513[10:14:57] <oxek> which is the behavior you observed
514[10:15:42] <AimHere> Yeah. There's still a big
difference between 'changes, sometimes for the worse, when you
ask to upgrade' and 'changes in the middle of the night
for no obvious reasons'
515[10:16:01] <AimHere> The latter is new to me!
516[10:16:43] <oxek> it's a fair point that debian does a
poor job of managing people's expectations.
519[10:19:05] <AimHere> Yeah, but I've been using sid for
at least 15 years. I thought I had reasonable expectations. I'm
not fazed by upgrading into unbootability or the like
522[10:20:20] <carloratm> novice question here, in firefox I see
some random font without antialiasing, what could cause the problem?
replaced-url
523[10:20:36] <carloratm> Just installed Debian btw
524[10:20:49] <oxek> yeah and I would not be surprised if sid
woke me up at 3am, spit on me and call me a dirty girl, and right
after that start crying and say sorry. It really is unstable.
525[10:21:45] <oxek> carloratm: I'm so sorry if this is
your first impression of #debian
656[11:40:17] <quadrathoch2> rue_mohr if you have a xeon from 20
years ago, no it's not an amd64 version
657[11:40:20] <quadrathoch2> but otherwise yes
658[11:40:27] <Lope> tomreyn, `dd if=/dev/md127 of=/dev/null
bs=1M` completes without error.
659[11:40:38] <rue_mohr> and it shouldnt be called i386 if an
i386 cant boot it (it cant, you guys put instructions in that kernel
that are not 386)
660[11:40:49] <azeem> rue_mohr: are you trolling?
661[11:41:06] <rue_mohr> no I'm trying to install debian on
the cool new dell server I was just given
662[11:41:17] <azeem> did you try a usb stick?
663[11:41:19] <rue_mohr> 24G of ram :)
664[11:41:28] <azeem> that doesn't sound like a new server
665[11:41:34] <azeem> unless you meant 24T
666[11:41:34] <rue_mohr> I do not think it can boot from a stick
667[11:41:35] <tomreyn> Lope: file -s /dev/md127
668[11:41:41] <quadrathoch2> rue_mohr so make sure that the
downloaded amd64 iso is fine per SHA512SUM, and verify that the way
you burn the disk is also good
669[11:41:41] <rue_mohr> its only option seems to be cdrom
670[11:41:48] <rue_mohr> stupid thing doesn't even have pxe
671[11:42:04] <quadrathoch2> so sounds like a very old system
672[11:42:08] <quadrathoch2> so maybe it's not amd64
673[11:42:23] <quadrathoch2> rue_mohr could you verify which
cpus you got in that device? (or which dell server)
674[11:42:40] <rue_mohr> Dell poweredge T610
675[11:42:46] <rue_mohr> anyone know?
676[11:43:55] <Lope> tomreyn, unfortunately I didn't
include file in my initramfs
677[11:44:11] <azeem> looks like it's from 2012
678[11:44:15] <azeem> but should be amd64
679[11:44:19] <Lope> tomreyn, I'll try chroot
680[11:44:24] <tomreyn> Lope: lsblk?
681[11:44:34] <quadrathoch2> rue_mohr so, as the cpu does have
amd64 instructions, it's still on your end, why the disc
doesn't boot
682[11:44:34] <tomreyn> i mean blkid
683[11:44:45] <Lope> tomreyn, `dd if=/dev/md127 of=/dev/null
bs=1M` completes without error.
684[11:44:48] <azeem> rue_mohr: that's the error you get
when trying to boot that?
685[11:44:58] <Lope> tomreyn, sorry, bad clipboard.
686[11:45:01] <Lope> tomreyn, /dev/md127: block special (9/127)
687[11:45:07] <Lope> that's the output of file
688[11:45:08] <rue_mohr> no bootable disk found
689[11:45:17] <rue_mohr> press F1 to retry
690[11:45:18] <rue_mohr> :)
691[11:45:37] <quadrathoch2> rue_mohr sounds like the cd was
burned wrong
701[11:46:54] <muffindrake> Heyas. I have a fresh debian sid,
and have run into a problem with the current obs-studio package. The
UI appears corrupted, with text and elements not rendering
correctly.
702[11:47:00] <rue_mohr> I'm not using up the last writable
cd's in the world am I?
703[11:47:12] <rue_mohr> hmm
704[11:47:33] <quadrathoch2> honestly, rue_mohr, it's
really weird, that that server can't boot off of usb :/
708[11:48:04] <tomreyn> Lope: hmm, this looks much bettzer,
maybe the 'file' command just didnt have a proper magic
bytes DB at hand.
709[11:48:14] <jmcnaught> Lope: are you trying to put your ESP
on an mdadm array?
710[11:48:31] <rue_mohr> its too bad it cant do pxe, I'm
set up!
711[11:48:31] <Lope> jmcnaught, EFI on mdadm raid 1
712[11:48:38] <Lope> jmcnaught, is that a no go?
713[11:48:43] <tomreyn> that makes not much sense
714[11:48:55] <jmcnaught> Lope: I do not see why that would
work.
715[11:49:07] <Lope> Well, it's just a fat filesystem on
mdadm
716[11:49:45] <Lope> so should I do the EFI naked, no mdadm?
717[11:49:58] <Lope> I'm trying to set it up so that either
SSD can fail, and it'll boot.
718[11:50:15] <jmcnaught> I would not expect the computer's
firmware to be able to understand an mdadm array.
719[11:50:17] <rue_mohr> I'll try the 386 image,
720[11:50:18] <tomreyn> well, it can work, metadata resides at
the end of the device, so firmware might just see the vfat file
system. but it's a recipe for breaking things, especially if
you let the firmware write to it.
758[12:17:55] <sfx2496> Does anyone known how to spoof
custom/enhanced build of ffmpeg in the linux(Debian/Mint20)package
manager, so it won't install the default compiled ffmpeg when
other programs want to install ffmpeg as req. dependency?
759[12:18:50] <tomreyn> there is no
"linux(Debian/Mint20)package manager"
760[12:19:13] <sfx2496> I just tried to make a deb package, it
installed, but does not show up as installed in the package manager
(synaptic show gray i icon before it, and the software manager of
Mint does not show it as proper installed either)
775[12:37:39] <shtrb> sfx2496, after building the deb file , did
you run dpkg -i filename.deb ?
776[12:37:43] <noisemaker> When I install some packages in one
machine, can I make that machine a local mirror for other debian
machines but only with the packages installed ?
777[12:37:50] <shtrb> Lope, sorry I was out
778[12:38:01] *** debhelper sets mode: +l 1200
779[12:39:23] <shtrb> Lope, are you sure that you are booted in
an EFI enabled mode ? [ -d /sys/firmware/efi ] && echo uefi
|| echo oldschool
782[12:44:06] <deq2496> shtrb: I just did, makes no difference
for the installation state, is it because of conflict with default
package?(I named it the same with same version in "contol"
file)
784[12:45:37] <shtrb> give us please the output of dpkg -i
yourpackaname.deb , also you should have a higher version than the
regular one so your version would not be overwritten
817[13:02:03] <dpkg> Ubuntu is based on Debian, but it is not
Debian. Only Debian is supported on #debian. Use #ubuntu on
chat.freenode.net instead. Even if the channel happens to be less
helpful, support for distributions other than Debian is offtopic on
#debian. See also <based on debian> and <ubuntuirc>.
876[14:26:33] *** Quits: lesless (~lessless@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
877[14:30:54] <ratrace> So I'm reading the Debian New
Maintainer's Guide, cover to cover. I wonder, why does it say
the package "should not" be a daemon? Who is supposed to
pacakge up daemons then?
880[14:35:37] <themill> Someone who is not a new maintainer.
881[14:37:28] <themill> It's just substantially harder and
adding all that is required to the new maintainer guide would
unnecessarily complicate the guide. The guide is about making a
simple package.
882[14:38:24] <themill> (the very next sentence even says as
much)
893[14:55:00] <shtrb> new mainter guide is actually written very
good, even outsiders (like me ) found it easy to package and almost
no lintian warnings , getting stuff into debian later might be a
problem
911[15:15:40] <gorillapatch> Hello everyone, I am experiencing
strange things with my Samba installation on Debian Buster. I have
added a usershare and even if I login with another username, I
cannot access and see this in my logfile "session closed for
user nobody"
948[16:19:55] <lf94> Hey, I'm trying to install debian on a
HP EliteDesk 800. What I've done is dd the iso onto an external
SSD, but I cannot boot from it.
949[16:20:17] <lf94> ...I'm out of ideas. I've gone
into bios and turned off secure boot, enabled legacy boot, tried a
combination of things.
950[16:20:34] <dvs> lf94: Do you intend to install Debian on
that same SSD?
951[16:20:41] <lf94> yeah
952[16:20:47] <dvs> that won't work.
953[16:20:50] <robobox> that doesn't work
954[16:20:56] <lf94> yeah, just realizing as you say it
955[16:20:59] <lf94> :D
956[16:21:08] <robobox> try a USB drive
957[16:21:15] <lf94> I've also tried dd'ing my current
install
971[16:26:35] <ratrace> if not, one should be very careful when
working with both copies on the same system. if yes, one should be
careful to update the clone's fstab and other configs for the
new UUIDs, if they were used.
972[16:26:38] <lf94> So for example, when I plug it into my
laptop (which I'm talking to you with), it appears as /dev/sda
973[16:27:01] <lf94> So technically it's been seen a usb
drive.
1033[16:38:57] <Mister00X> but I meant the original iso, did you
comared it against a hash
1034[16:39:03] <lf94> ahhh
1035[16:39:08] <lf94> no, lemme do it
1036[16:39:15] <ratrace> should be possible to md5sum the ISO and
then md5sum from the USB disk, from byte 0 to exactly the number of
bytes the ISO is
1037[16:39:18] <gorillapatch> pasiz: Sounds strange, as my client
is presented a login dialog
1038[16:39:37] <jhutchins> lf94: The purpos of secure boot is to
prevent you from installing linux, so you want that disabled. Any
quick/fast boot feature should also be disabled.
1039[16:40:02] <lf94> jhutchins: that's what I thought too,
and I"m just trying anything at this point.
1166[18:09:27] <cole_phelps> i have a problem with python3_numpy
can someone help me?
1167[18:09:46] <sney> !anyone
1168[18:09:47] <dpkg> Please do not ask if anyone can help you,
knows 'something' or uses 'some_program'.
Instead, ask your real question. (If the real question _was_
"does anyone use 'some_program'?" ask me about
<popcon> instead.) See <ask> <ask to ask>
<polls> <search> <sicco> <smart questions>.
1174[18:10:41] <Mister00X> alexrelis[m]: yes apt-listchanges I
think
1175[18:11:39] <alexrelis[m]> Mister00X: `apt-listchanges`
doesn't have a changelog for every package, so I'm
guessing it's only for some packages then right?
1186[18:15:52] <cole_phelps> s me an error of input i try to
apt-get upgrade on kali and python3_numpy is not installing because
of\some input/output error which i think causes a dependency problem
and causes the packages to not upgrade i've been stuck w/ it
for two weeks now
1188[18:16:48] <alexrelis[m]> shtrb: This is epic! You learn
something new every day. Thanks guys!
1189[18:16:48] <sney> why do kali noobs all think #debian will
support them
1190[18:16:50] <ratrace> cole_phelps: nothing we can help with in
#debian, tho.
1191[18:16:55] <sney> !kali-magic
1192[18:16:56] <dpkg> Kali Linux is just an operating system. It
does not magically turn you into a hacker. It is also offtopic in
#debian. Please fulfill your fantasies in #kali-linux on
chat.freenode.net. See also <based on debian> and <they
don't know>.
1193[18:16:58] <cole_phelps> fuck you
1194[18:17:12] <azeem> cole_phelps: please keep it on topic
1195[18:17:14] <cole_phelps> at least answer with something
usefull
1196[18:17:15] <alexrelis[m]> cole_phelps: Don't be mean
1197[18:17:34] <ratrace> !tell cole_phelps about kali
1198[18:17:38] <shtrb> sney, because kali is in fact a rebardning
of unstable
1199[18:17:43] <alexrelis[m]> You can ask your question in the
#kali-linux chat.
1213[18:21:40] <sney> shtrb: it's not a rebrand, it's a
derivative like all of the other derivatives. Mint isn't a
rebrand, neither is parrot, mxlinux, etc. And Kali's website
has a *lot* of documentation on it including links to community
support.
1217[18:23:42] <shtrb> sney, kali is advertised^(tm) as a pure
bland of debian, while in fact there are some differences. Mint on
the other hand is not a pure bland
1219[18:24:14] *** Quits: Makerblaker (~Makerblak@replaced-ip) (Remote host closed the connection)
1220[18:24:45] <azeem> (it's blend)
1221[18:25:16] <sney> there is no mention of that on
replaced-url
1222[18:25:38] <EdePopede> that's how it *looks* like >
replaced-url
1223[18:25:52] <shtrb> azeem, thanks for the correction
1224[18:26:18] <EdePopede> found it 2 days ago. and btw i'd
really like to have this kind of dependency graphs instead just dot
;)
1225[18:26:18] <cole_phelps> back from the kali chan...nobody
answered
1226[18:26:24] <sney> !they don't know
1227[18:26:24] <dpkg> We're sorry your distro's channel
isn't being helpful, but that doesn't make it appropriate
to use #debian for non-Debian questions. Please go back to your
channel and wait patiently for better help, or install Debian and
party with us.
1228[18:27:19] <shtrb> sney, thanks , perhas I got the wrong
impression all toghther , I sure remembering such a claim somewhere
, thank you for correcting me
1229[18:27:37] <azeem> huh, so kali linux is a knoppix fork?
1230[18:27:40] <sney> regardless, #debian has never supported
those pentesting distros and we aren't about to start now
1231[18:27:49] <azeem> I guess they rebased it since
1235[18:28:41] <EdePopede> azeem: would be nice if distrowatch
was up-to-date and they'd autocreate such graphs :)
1236[18:28:43] <sney> one of kali's core developers is a
debian developer, he also co-authored the debian
administrator's handbook. the connection is there. and they
*could* have made a blend, but they didn't.
1237[18:28:59] <sney> I suspect bundling non-DFSG stuff is part
of that.
1238[18:28:59] *** Quits: Ericounet (~Eric@replaced-ip) (Quit: Je m'en vais ...)
1254[18:36:23] <shtrb> ratrace, they didn't do anything yet,
but if they put their code where their words are we should expect
mozilla products, I would expect them to enable kinds of censership
and to prevent some people from using their tools (which would be a
DFSG problem )
1255[18:36:34] <EdePopede> wouldn't this be just the
newsstream on the new tabs?
1256[18:36:57] <ratrace> shtrb: that's a very vague and
nonspecific and speculative super-maybe-someday-something-something
:)
1257[18:37:03] <shtrb> EdePopede, first they need to commit, and
push, now that is just too vauge
1258[18:37:07] <EdePopede> or do you expect sth like
apple/google/.... did these days?
1259[18:37:08] <azeem> yeah, can we get back to topic
1430[21:44:34] <gholinbrown> a clock shouldn't need that
kind of memory. it's also questionable why the non-virtual
memory is 32mb for a clock makes no sense either. it's a clock.
1431[21:45:56] <gholinbrown> but let's focus on the 300+MB.
as virtual memory my understanding is that's how much got
DMA'd?
1432[21:46:37] <AimHere> Are you taking into account that most
memory-reading tools in Linux count every shared library as part of
the memory total, even if they're used by something else?
1433[21:47:35] *** Quits: pvdp (~pvdp@replaced-ip) (Remote host closed the connection)
1434[21:47:39] <gholinbrown> shared memory says 25 MB. that
accounts for the non-virtual memory i guess. but not the virtual
1446[21:56:53] <ratrace> gholinbrown: so that's somethign
specific to the MATE desktop? and how are you measuring that
"applet's" specific memory usage?
1475[22:04:29] <ratrace> gholinbrown: yeah it's pretty much
all memory, including shared libs
1476[22:04:57] <gholinbrown> is it literally using that memory?
1477[22:05:06] <ratrace> so that's not memory exclusive to
it. large desktop environments use tons of libs that interlink and
you only have to link with one of them to get the whole tree into
your virtual memory
1478[22:06:56] <gholinbrown> it's giving me totals like
these are statically linked
1479[22:07:31] <shtrb> gholinbrown, lot's and lot's of
fancy libs to do fancy animations ICU is used for
internationalization (globalization) , ibexpat is travel set ,
libXext is actually X related , libpng you get what it does
1480[22:07:32] <ratrace> wait until you learn about electron apps
carrying an entire copy of chromium.... each :)
1481[22:07:46] <EdePopede> what are these anon entries at the
end? in my case 2 of them are at 65M and some at 8M which is a big
part already of the 300 total.
1482[22:08:23] <shtrb> EdePopede, this could be a mmapped file or
true malloc
1483[22:08:30] <ratrace> EdePopede: memory without a file backing
it
1493[22:14:50] *** Quits: sparky4 (~sparky4@replaced-ip) (Remote host closed the connection)
1494[22:14:55] <gholinbrown> why would crypto be linked
1495[22:15:54] <shtrb> EdePopede, If he has some app or config
that need Thai , or if the applet support the Thai language it would
be understanble to have it
1496[22:16:28] <EdePopede> ,info ibus-libthai
1497[22:16:31] <judd> Package ibus-libthai (utils, optional) in
buster/amd64: Thai input method engine for IBus based on LibThai.
Version: 0.1.4-4; Size: 19.2k; Installed: 75k; Homepage:
replaced-url
1498[22:16:39] <sney> sometimes thai gets pulled in with
multi-language font support for emoji purposes
1499[22:16:55] <EdePopede> i had ibus installed, removed it
since. didn't check further. who knows.
1500[22:17:03] <EdePopede> oh. ah.
1501[22:17:32] <sney> pango has some thai dependencies as well,
at least here on bullseye
1502[22:17:57] <EdePopede> yep, i've seen pango. one of the
names i recognized at least.
1508[22:22:29] <gholinbrown> thank goodness. so it's just an
implied minimum of need libraries (once) but also the app.
1509[22:22:56] <freem> so basically, kernel only have one copy of
each in memory, and only duplicates the data sections (I think) when
they are writen into
1510[22:24:43] <shtrb> freem, and then we have guix
1511[22:24:59] <freem> well... yeah
1512[22:25:12] <shtrb> I wonder if we will have guix in debian
1513[22:25:47] <shtrb> I'm talking about the multi
versioning option for packages
1514[22:26:05] <freem> I wonder how hard it would be to write a
dpkg that would install things into $HOME instead of $ROOTFS
1515[22:26:25] <shtrb> not that it would be fun to have packages
written in scheme
1516[22:26:54] <freem> I know guix/fooix/whateveuix offers other
guarantees, ofc, but this clearly seems their main sell point
1517[22:26:58] *** Quits: yans (~yans@replaced-ip) (Remote host closed the connection)
1525[22:33:32] <jhutchins> alexrelis[m]: The problem is that we
rely on the packagers and developers - which are often different
teams to include proper changelogs, and a lot of times they
don't.
1526[22:34:07] <shtrb> jhutchins, apt-get changelog give some
data (*) not the best one but some
1527[22:34:59] <freem> each time I looked at this, I had only
packaging changes, I think
1528[22:35:38] <shtrb> well yes, that's the idea
1529[22:35:55] <shtrb> bugfixes, solutions , new options etc
1530[22:36:38] <freem> well, this is my point, actually. Only
_packaging_ changes. Like, "bump version", "new
maintainer", etc
1532[22:37:20] <freem> but I admit I don't look at it often,
maybe I didn't paid enough attention, especially since debian
normally does not add new features while in stable
1533[22:38:14] <shtrb> So what kind of a changelog would you
expect to have ? the actual software changelog ?
1534[22:39:04] <jmcnaught> A package's
/usr/share/doc/PACKAGE/ directory will have a changelog.Debian.gz
(packaging changes) and a changelog.gz from upstream.
1535[22:39:06] <freem> yes, but I presume it's simply in
/usr/share/doc/$package when it exists?
1536[22:39:28] <freem> but that imply installing/upgrading before
reading it
1538[22:39:52] <black_13> can packages be installed with debug
symbols
1539[22:40:02] <sney> !dbgsym
1540[22:40:03] <dpkg> Packages that end in '*-dbgsym'
contain the symbols required for debugging executables and
libraries. The dbgsym packages are automatically generated packages
that are in a separate archive; add a line like "deb
replaced-url
1541[22:40:10] <shtrb> black_13, you have special packages for
that called dbgsym ones
1562[22:49:22] <unixbsd> With Libreoffice or similar for ODT
document (writer/opensource doc), how to export FONTs into the
exported PDF document? there is no infos about that.
1563[22:50:19] <sney> some "no infos" from the first
google result:
replaced-url
1604[23:12:03] <EdePopede> maybe one of them is down
1605[23:12:09] <tomreyn> security.debian.org resolves to 12
ipv5/6 addresses
1606[23:12:19] <tomreyn> *ipv4/6 :)
1607[23:13:03] <sney> if you want to use a mirror that is *a*
mirror instead of a cdn, try ftp.(country code).debian.org, or see
the 'secondary' entries at
replaced-url
1608[23:15:05] <tomreyn> seanthegeek is ignoring us anyway,
though.
1609[23:15:38] <unixbsd> how to send to 30 persons with
thunderbird an email? We need to add it one after one into the 30
fields: TO ???
1623[23:20:12] <ratrace> had to termporarily block clients email
accounts because the jokers would put hundreds of addresses into a
single TO field .....
1624[23:20:49] <ratrace> (during xmass, because all the greetings
and nonsense)
1627[23:23:07] <EdePopede> no way to do such a thing with some
addon at least? thunderbird is gui, so it should be normal workflow
to open a dialog with a list and just click the names
1655[23:33:08] <sney> I've seen the grub step fail on
install a few times over the years, though it usually has the
courtesy to display a big error message about it. maybe it was
accidentally installed to the wrong volume.
1656[23:33:54] <dvs> maybe the MBR version was installed instead
of EFI?
1657[23:34:37] <ddsys> he installed to the EFI partition
1658[23:35:06] <dvs> but did he use the EFI install option?
1659[23:35:20] <ddsys> hmm
1660[23:35:25] <ddsys> not sure
1661[23:35:31] <frazzydee> dvs: I remember checking for that, and
thought it was the EFI version. Where would that option be?
1662[23:36:01] <dvs> frazzydee: right at the beginning with the
menu that selects the different install opptions
1677[23:44:16] <ddsys> frazzydee: is dual-booting windows. It
already had an EFI partition. doesnt an EFI partion mean its in uefi
mode and the installer should pick up on that?
1678[23:44:59] <ddsys> dvs ^ ?
1679[23:45:12] <sponix> ddsys: only if it is continued to be
booted that way. If you are EFI in Windows, and don't choose
EFI boot in Linux, it will just do Legacy and not see the Windows
(and Grub will not either)
1682[23:46:14] <dvs> ddsys: No, it depends on how the USB
installer was booted. On my machine if I have an install ISO, I have
two boot options for it: one with just the name and another with
"UEFI:" in front of the name.
1683[23:46:18] <sponix> These messes are hard to straighten out
1684[23:47:09] <frazzydee> I recreated the install media and
booted with it. it shows "UEFI installer menu"
1685[23:47:47] <ddsys> dvs: i could swear my installer usb only
shows uefi if my disk is gpt
1688[23:48:07] <frazzydee> I think I checked to make sure it was
in UEFI mode and not BIOS mode previously. I've been trying to
get this working for a few days and have been on this channel
before, someone else asked me to check that
1689[23:48:19] <frazzydee> Yes, there isn't a choice of BIOS
or UEFI mode for me either
1690[23:48:37] <frazzydee> Anyways I'm glad it works now,
and if someone else has the same issue they can try refind also
1691[23:49:03] <sponix> frazzydee: you got it working ?
1692[23:49:22] <frazzydee> sponix: Yes! You were the someone else
haha ;)
1693[23:49:25] <dvs> ddsys: it totally depends on how you boot
the USB installer. It will show either BIOS or UEFI, but not both.
1694[23:49:35] <sponix> frazzydee: I have memory loss :(