66[00:57:35] <alexrelis[m]> I can't seem to find the
proper Debian package to install that corresponds to this library.
The best I can find is `libpng12-0`, but that is a packaged designed
for Jessie.. not Buster.
99[01:17:47] <EmptyMessage> Given that it is just 1 old library
I cant imagine that it would ruin things, I have done that in the
past, and so far it has never backfired
103[01:18:38] <EmptyMessage> its definately bad practice, but
there are times when I make the decision that it is fine to do. It
is up to your judgement
104[01:19:01] *** debhelper sets mode: +l 1114
105[01:19:23] <alexrelis[m]> I'm very picky with these
things. I'm the type of person where I strive for longevity
with my system and even if it's fine, it will almost usually
result in a crufty system because I always forget I installed it and
forget to delete.
106[01:20:07] <ryouma> should your bug be a known bug?
107[01:21:08] *** Quits: Tobbi (~Tobbi@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
108[01:21:12] <EmptyMessage> makes sense. Here is another
solution that is a bit jankier, but could be safer. Libpng16 should
be backwards compatabile with libpng12, so you could symlink
libpng16 to libpng12
110[01:21:48] <alexrelis[m]> ryouma: You talking to me? I
wouldn't say it's a bug, it's just a missing
dependency for an external program I'm trying to run for one of
my classes (Cisco Packet Tracer).
127[01:26:50] <ryouma> i find emulators and anything but a basic
chroot to be a bit cumbersome to understand and remember and use,
partly because of accessibilty in vms but also all sorts of options
you have to figure out
140[01:31:54] <ryouma> on oldstable, without upgrading
firefox-esr, it changed. it used to be that pressing and releasing
alt would bring up the main menu. this does not occur any longer.
193[02:30:18] <sponix2ipfw> Eryn_1983_FL: last time I checked
the some live Linux like Mint have the shred command
194[02:30:23] <ryouma> dd on devices is pretty good. there are
also fs-level tools like wipe or whatever, one of which has kind of
hilarious levels of paranoia-ish in the man page. some might work on
devices so you don't need to boot.
195[02:30:42] <ryouma> idk the options for dd though
196[02:31:09] <sponix2ipfw> Of course dd with urandom does well
enough for most things
204[02:40:32] <ryouma> it has been claimed that hdd are so dense
that you only need one pass. idk if that is true. they keep getting
denser which might imply otherwise.
205[02:40:40] *** Quits: black_ant (~antilope@replaced-ip) (Quit: simplicity does not kill)
261[04:13:04] <SofS> About secure erasing whole disks, first
there is the need to distinguish between SSDs and magnetic disks. On
SSDs the only sure way to delete anything is by using the
manufacturer provided secure erase tool as these will manage the
blocks in use transparently behind the OS to prevent wear. For
magnetic discs there are several tools, scrub (replaced-url
262[04:13:04] <SofS>
software/coreutils/manual/coreutils.html#shred-invocation) and the
secure-delete toolkit (sfill, sswap, sdmem).
263[04:16:07] <SofS> All of these have finer tuned and more
expansive options than plain dd and each has its use case, read the
respective manuals to be sure of what you need.
264[04:16:30] *** Quits: epony (epony@replaced-ip) (Remote host closed the connection)
525[10:38:24] <gzp> I would like to assign 3 different ip to one
interfice on debian10, but seems like the second and third ip
doesn't comes up. ifup says: ifup[320]: RTNETLINK answers: File
exists
581[11:30:02] <ksk> If it is protected DVD you will need to have
the library to decode them - be aware that just enabling non-free
and pulling a single package is not something bad per-se.
584[11:30:27] *** Quits: JohnML (~john1@replaced-ip) (Remote host closed the connection)
585[11:30:48] <oxek> this is for a kiosk that will be in a
school, where kids will be allowed to play DVDs
586[11:30:53] <oxek> so we don't want to break any licenses
587[11:30:58] <oxek> the DVDs are of course originals
588[11:31:14] <oxek> I know windows license allows playing DVDs
589[11:31:22] <oxek> (because it's included in the price)
590[11:32:21] <ksk> eh? It is not about "paying money"
or so. It is about Debian being "free software" - and the
software needed to decrypt DRM of such DVDs is not "free, as in
debian definition"
591[11:33:13] <oxek> and the software to play DVDs is also not
'free' as in beer, because a license needs to be paid to
decrypt DVDs, as far as I understand it
592[11:33:25] <oxek> I'm reading through the wiki page
though
593[11:33:43] <ksk> "enable non-free; apt install
$package" does not seem to invole "Put money here" :P
594[11:34:17] <ksk> It might be that you run into legal stuff by
using the library - but thats totally out of scope for #debian -
contact a lawyer.
647[12:03:13] <jelly> Oddmonger, there are two different
versions of "iptables" tools in Debian 10, one is legacy,
one is nft backed and creates nftables rules in background
649[12:04:02] <jelly> I don't think it will take anything
less than 10 years for "legacy" to disappear. They said
ifconfig was legacy back in 2000 with 2.4 kernel. It still exists.
650[12:04:56] <Oddmonger> well i've spent quite a lot of
time on iptables, now it works as i want to, so i'm not in a
hurry
739[12:58:12] <oxek> does debian provide a supported way of
running the longterm linux kernel version 5.4 as opposed to the
current version in debian-stable which is 4.19?
751[13:01:56] <oxek> I see there are packages in
buster-backports linux-image-5.4.0-0.bpo.2-amd64
linux-image-5.4.0-0.bpo.3-amd64 linux-image-5.4.0-0.bpo.4-amd64
752[13:02:05] <oxek> and I don't understand what that 2 3 4
means
754[13:03:43] <ratrace> oxek: debian uses LTS kernels in stable
releases. bullseye's kernel will be available through backports
on buster, but that won't be 5.4, but next LTS
755[13:04:55] <oxek> both 4.19 and 5.4 are LTS releases, so I
was wondering if the 5.4 LTS would be available in buster
781[13:11:54] <oxek> odd that those packages are still available
then
782[13:11:59] <ratrace> the packages you mention are no longer
offered in offician debian repos. I have no idea why you see them.
can you pastebin apt-cache policy linux-image-amd64 ?
785[13:12:40] <oxek> but I can still install them with apt
786[13:13:32] <oxek> probably because linux-image-amd64 does not
depend on those packages, but those packages can be installed
directly
787[13:14:40] <ratrace> linux-image-($uname a) is the actual
package name, the rest are metapackages, like linux-image-amd64
which will pull in specific package version
788[13:15:35] <oxek> linux-image-5.4.0-0.bpo.4-amd64 looks like
an actual package name
804[13:22:20] <ratrace> it shows the package. I never really
used list and I'm not sure what it's supposed to do
exactly.
805[13:22:49] <oxek> ok then 'apt show
linux-image-5.4.0-0.bpo.4-amd64' still shows the package
806[13:22:50] <ratrace> that package is maybe available in the
archives, but it's not referenced throught the metapackage
version, so shouldn't really be installed
807[13:22:57] <oxek> I can even 'apt -s install
linux-image-5.4.0-0.bpo.4-amd64' successfully
808[13:23:07] <ratrace> at any rate, it's not official for
buster, and won't be. bullseye/sid are moving on 'till the
next LTS
809[13:23:45] <ratrace> right if you apt policy _that_ specific
package name, it shows. it doesn't show the version for policy
of linux-image-amd64
909[13:50:55] <ratrace> wubi, that's the one, thanks oxek
910[13:51:25] <ratrace> meanwhile there's windd or google
for "dd for windows"
911[13:52:06] <oxek> between cygwin and WSL, there's little
need for dd.exe or wget.exe unless as part of some project
912[13:52:40] <ratrace> but dd.exe or whatsitcalled, ist just
something oyu click, install, run, and bake your iso. cygwin and wsl
are way more involved, right?
913[13:53:01] <oxek> oh, then I am thinking of a different
dd.exe
917[13:53:34] <ratrace> I had to use something like that last
year. my computer broke and the USB thingy with recovery ubuntu live
env was broken, so I had to bake a new one through friend's win
machine
918[13:53:36] <jelly> wsl or a VM (or wsl2, which is both) seem
to be the easy ways of getting linux to run if you're mostly
using windows
919[13:54:08] <H4ndy> for bootable ISO/thumb drivers there are
much better ways on Windows like that (usbit, Rufus, etcher, etc.)
920[13:54:15] <H4ndy> WSL is a godsend, it works so well
921[13:54:25] <algun> hey, here it is:
replaced-url
922[13:54:33] <oxek> etcher... using Electron framework to do dd
924[13:54:48] <H4ndy> Yeah don't go into specifics, I just
talk about user experience ;)
925[13:54:55] <H4ndy> Electron can die in a fire
926[13:54:56] <ratrace> wsl is a gateway to allow (linux baseD)
developers continue to use windows when MS switches to apple-lind of
closed garden with MS store only installables
927[13:55:27] <ratrace> apple-*kind
928[13:55:51] <H4ndy> they tried and failed, I doubt it will
happen anytime soon
929[13:56:00] <ratrace> meanwhile, I'm happy my windows
xperience is limited only to running it in a VM for the sole purpose
to play win-only games that won't even run on steam+proton
930[13:56:03] <oxek> I don't know how I feel about wsl yet.
It's nice that I can run debian userspace in windows.
931[13:56:07] <jelly> installing Debian from said store is still
fun.
932[13:56:38] <jelly> and very useful when you have too new
hardware that works like crap natively
933[13:57:02] <ratrace> there's so much betrayal in this
channel now, I feel nauseus
934[13:57:14] <algun> seems you're right jelly, I get an
error trying to run
replaced-url
935[13:57:39] <ratrace> praises to WSL..... ugh.... how can
you.... it's the tool in disguise for the final E
936[13:57:46] <H4ndy> "betrayel" ha. I use the tool
that I think is best for a task and this often involves Windows for
me 🤷
937[13:57:54] <oxek> talking of newer hardware in debian, how
much breakage is expected by running the backports kernel
(linux-image-amd64)?
947[14:03:34] <jelly> a rackmount server these days usually has
an internal usb port and/or an internal SD card place, so you can
get otherwise diskless servers booting without having to set up
netboot or boot from SAN
963[14:08:22] <oxek> I thought they are doing their own thing
964[14:08:23] <algun> yes
965[14:08:25] <oxek> proton or somesuch
966[14:08:28] <yanmaani> jelly: Well yes but with nice
performance, and WM integration
967[14:08:39] <hwm4rgs> with office suites being good enough in
the browser and supertuxkart having everything one needs for
multiplayer gaming, there's no reason left ;)
968[14:08:43] <yanmaani> Actually I think WM integration and
some hacks for the hadrware accep would be fine
969[14:08:53] <algun> it's a fork of wine
970[14:09:02] <yanmaani> why fork it?
971[14:09:15] <algun> because companies i guess
972[14:09:29] *** mirrorbird_ is now known as mirrorbird
973[14:10:05] <H4ndy> As someone involved with a OSS project, if
the maintainer is not able to access the mass of code influx or
wants the project in another direction, forks are just fine a
solution
974[14:10:31] <algun> this doesn't seem to be going in
another direction
975[14:10:38] <ratrace> and this particular fork is contributing
back
976[14:10:39] <algun> not that going in another direction is
always justified
977[14:10:54] <algun> or even most of the time imho
978[14:11:00] <ratrace> has to be a fork, just like to use
github code (with contrib in mind) you have to fork it too
992[14:13:28] <ratrace> checked steam? I found some rather old
games like Blood on steam
993[14:14:12] <oxek> I'm not a fan of steam. It works for
others, but I prefer fully offline games that I actually own as
opposed to license.
994[14:14:38] <ratrace> oxek: you never own a game even
"offline". you own the piece of plastic it came with, but
the software is licensed to you
995[14:14:39] <oxek> plus steam has weird packaging in debian
996[14:15:01] <oxek> ratrace: true. But at least I can't
have some 3rd party terminating my account.
997[14:15:05] <ratrace> dunno what you mean. apt install steam
worked fine here
998[14:15:24] <oxek> ratrace: yeah, but that does not install
steam, that installs the steam installer which then installs steam
999[14:15:34] <oxek> that's what I call weird packaging
1000[14:16:30] <ratrace> ah that. yes. well, that's because
only the installer is dfsg compliant, iirc. also, steam needs timed
and coordinated updates. debian maintainership process would nuke
that.
1001[14:16:31] <oxek> and it requires dpkg addarchitecture i386
1002[14:16:43] <ratrace> yea, the steam client is still 32-bit
1003[14:16:58] <ratrace> has to be, to support still-32-bit games
1004[14:17:21] <ratrace> (and by client I really mean the
ubuntu12.04 runtime it comes with)
1006[14:18:10] <ratrace> but all that is annoying a bit, I agree;
which is the reason I'm running Steam appArmored, and am
considering to switch to another linux VM entirely, since I have a
windows VM for games too
1007[14:18:32] <oxek> I see steam has a flatpak version
1008[14:18:41] <ratrace> I hear it's very well curated
1009[14:19:05] <ratrace> ironically, there's no snap of it.
and steam uses ubuntu runtime....
1010[14:19:39] <oxek> I probably remember it wrong then, I
thought steam's linux support was heavily based on debian -
they even gave all debian developers free games
1011[14:19:50] <ratrace> you're thinking of SteamOS
1012[14:20:00] <ratrace> the steam linux client always had ubuntu
runtime
1013[14:20:05] <oxek> probably, it's been a long time since
I played games
1022[14:23:09] <ratrace> I installed it with gpu passthrough few
weeks ago, and thought the VM was being sluggish and in need of
tuning until my windows using friend (also a windows admin) came and
told me that's normal behavior..........
1023[14:23:28] <ratrace> the games run fine tho, but otherwise,
the ui, everything, it's.... unbearable.
1024[14:23:54] <oxek> modern UI in any system is unbearable.
1025[14:24:18] <yanmaani> Protip: If you ever need to do anything
with Windows, use Windows Server or the Enterprise versions
1026[14:24:22] <ratrace> I've set up steam big picture mode
so I don't even see windows under it
1027[14:24:32] <yanmaani> much more sanity, less tracking, etc
1028[14:24:47] <oxek> and the licenses cost 10x more
1029[14:25:01] <oxek> (and you can't even buy enterprise
ones as a home user)
1030[14:25:02] <ratrace> well.... chances are you can find very
cheap ones on ebay
1031[14:25:12] <ratrace> I got mine for w10 home for like 5€
1032[14:25:19] <ratrace> and! it's all legal.
1033[14:25:35] <oxek> I don't know about the ebay ones.
I'd question the legality and I want all my stuff to be legal.
1034[14:25:42] <oxek> hence, debian.
1035[14:25:43] <yanmaani> Do people actually pay for Windows
licenses?
1036[14:25:54] <yanmaani> (I don't use Windows, to be clear
here)
1040[14:26:21] <ratrace> it's very much legal. the EULA says
you need an _invoice_ . I have an invoice, I bought it from a
company that has EU VAT ID. It is not my responsibility or legal
requirement, to check where _they_ got it from. I satisfied the
EULA.
1041[14:26:25] <oxek> and a lot of the times I could not avoid
it, e.g. when buying a laptop
1042[14:26:30] <ratrace> yanmaani: yes, I pay.
1043[14:26:40] <ratrace> I also donate to FOSSy software
1044[14:27:13] <yanmaani> it's not criminal to violate an
EULA, just breach of contract
1045[14:27:28] <ratrace> and the contract says nothing about
buying form third parties :)
1046[14:27:47] <ratrace> is that a loophole? dunno. but it's
not illegal, and not in violation of EULA.
1047[14:27:53] <oxek> ianal, so I'd rather be safe than in
jail
1054[14:29:01] <ratrace> the friend who recommended this
procedure, who is the said windows admin, told me he once bought a
license key from microsoft.com directly. turned out, microsoft.com
used a third-party vendor and later revoked the keys.
1055[14:29:02] *** debhelper sets mode: +l 1141
1056[14:29:15] <ratrace> so.... yeah... legal. even MS does it
through microsoft.com.
1057[14:30:11] <ratrace> if MS thinks the key is not legit,
they're free to revoke it, but I have an invoice of legal
purchase from a company with EU VAT ID. I'm within my consumer
rights.
1058[14:30:38] <oxek> yeah, but how do you get MS to fix it? They
are just like google - no way of contacting them.
1059[14:30:46] <ratrace> to fix what?
1060[14:31:02] <H4ndy> you contact your seller as they need to
fix the situation
1061[14:31:10] <H4ndy> at least under EU law
1062[14:31:11] <oxek> and the ebay seller is long gone
1063[14:31:22] <ratrace> also it's not true, unlike google,
microsoft has a 0800 number in most EU countries, and I called them
a number of times in the past
1068[14:32:05] <oxek> ratrace: with an actual human there? I only
ever called the 0800 to get the license codes and it was a bot
1069[14:32:09] <ratrace> if they revoke the key, I just get a new
one or reinstall anew. you can freely download the ISO from MS site,
and it's free to use without activatoin key for 30 days, after
which you just get a nag message about the key, but you can still
continue to use it.
1070[14:32:43] <oxek> using the software after 30 days is in
breach of license though. I even paid for winrar.
1071[14:32:44] <ratrace> oxek: yes, with an actual human. I used
that line to fix windows xp licenses activation, windows 7 and 8
activation. with 10 I don't need it any more, it's all
auto.
1072[14:33:01] <oxek> neat, so MS is better than google
1073[14:33:05] *** Quits: cupcake90 (~cupcake90@replaced-ip) (Remote host closed the connection)
1074[14:33:32] <ratrace> much better. I can get them to fix
deliverability issues to @hotmail and @outlook.com. I can't get
google, google doesn't have any email, contanct form, phone,
nothing.
1075[14:33:39] <ratrace> MS >>>> Google.
1076[14:33:44] <yanmaani> Google does, I've contacted them
1077[14:33:57] <yanmaani> they're not very helpful but it is
possible
1078[14:34:26] <ratrace> maybe if you called their switchboard in
moountain view
1079[14:34:29] *** Quits: endstille (~endstille@replaced-ip) (Quit: I'll be back.)
1080[14:34:38] <yanmaani> that, or send e-mail, or send letter
1096[14:40:59] <Oddmonger> «is your computer on ?»
«yes» / «have you reinstalled windows»
«HAHA it doesn't work from my phone too, find something
else !»
1160[15:56:22] <mandeep> i have an encrypted disk that im trying
to resize so i can create another partition. ive tried using google
resources to find a method but keep running into problems where the
free space on the drive isnt recognized
1161[15:58:19] <jelly> mandeep, what does your block device
hierarchy look like? Pastebin the output of "lsblk" and
"pvs" and "lvs" (run as root)
1162[16:00:10] <jelly> it's okay if you don't have pvs
and lvs commands available
1170[16:03:44] <jelly> mandeep, the LV and filesystem containing
your root filesystem is too big. To shrink ext4 it needs to be
unmounted and freshly fscked; so, if you're going to do it that
way, you need to boot some other linux, maybe a live usb or livecd
1171[16:04:22] <mandeep> yeah that's fine but since im not
too familiar with this, ive been looking for a step by step guide
but haven found anything that works
1172[16:05:11] <jelly> how much of your / is filled up?
1173[16:05:27] <jelly> df / or df -h / or just df
1174[16:05:35] <mandeep> 69%
1175[16:05:56] <jelly> and you need a new filesystem how big?
1176[16:06:06] <mandeep> i'd like to use 40gb
1177[16:06:13] <mandeep> so i guess 15-20%
1178[16:06:51] <jelly> if you have a second disk, make a backup
before doing anything
1179[16:06:58] <mandeep> no second disk :(
1180[16:07:27] <jelly> what do you need that 40GB for? Maybe you
can use an image file within the existing filesystem
1181[16:07:41] <mandeep> have to install ubuntu 18.04 for some
work software
1182[16:08:00] <jelly> ouch. Do it in a VM?
1183[16:08:11] <mandeep> thought about it but i need speed
1184[16:08:38] <jelly> depending on what software it is, a chroot
or systemd-nspawn container might work as well
1185[16:08:54] <jelly> you can use debootstrap to make a bionic
chroot
1186[16:09:00] <mandeep> hmm'
1187[16:09:18] <mandeep> i tried a docker container and it came
with its own set of problems. i wonder if a chroot would be similar
1188[16:09:23] <jelly> yes
1189[16:09:38] <mandeep> similar in that it has problems :(
1190[16:09:51] <jelly> but a different set of problems!
1191[16:10:16] <mandeep> haha
1192[16:10:18] <jelly> brb lunch
1193[16:10:21] <mandeep> i can give it a try atleast
1196[16:10:53] <mandeep> problem is that i need a gui, so i dont
think a container will work
1197[16:11:34] <jelly> sure it will work, just needs some
coercing
1198[16:12:33] <jelly> GUIs are not magical, they just need a few
resources from the X server or Wayland server accessible, and that
can be added to a chroot or a container
1199[16:12:44] *** Joins: magyar (~magyar@replaced-ip)
1203[16:15:42] <neoclust> i try to install virtualbox on debian
10 but it fails to start
1204[16:15:47] <neoclust> the error message is:
1205[16:15:50] <neoclust> where: supR3HardenedMainInitRuntime
what: 4 VERR_VM_DRIVER_VERSION_MISMATCH (-1912) - The installed
support driver doesn't match the version of the user.
1217[16:19:52] <mandeep> this seems like a lot more work than
partitioning and installing a new system
1218[16:20:13] <mandeep> especially for someone who is not
familiar with any of this
1219[16:21:01] <ratrace> mandeep: debootstrap effectively just
unpacks the base distribution into a directory. that means you need
to prepare the directoy with partitioning and other layers (lvm,
mdadm, luks, anything you might want)
1220[16:21:15] <mandeep> argh
1221[16:21:20] <ratrace> mandeep: then you need to setup the
resulting distribution with some basic configs like network,
root/user password
1222[16:21:26] *** Quits: aeWessan (~aeWessan@replaced-ip) (Remote host closed the connection)
1223[16:21:46] <ratrace> and in case of kernel/grub installation,
you need to chroot into the directory for grub/initramfs hooks to
autodetect and set themselves up
1226[16:22:11] <ratrace> mandeep: it's worth learning. I use
debootstrap only to install debians and ubuntus. primarily due to
zfs, btrfs and luks interacting.
1228[16:22:50] <mandeep> yeah i agree but unfortunately im on
limited time due to this being something needed for work and i
needed to be up and running by yesterday
1237[16:27:49] <amprxc> Hello, I want to transfer backup files
with rsync. They are ZFS snapshots encrypted and compressed as bz2
files. Then, it is worth using the --inplace option ? I plan to use
the options "-av --append" to resume stopped transfers. Is
it ok ? Thanks
1242[16:30:45] <ratrace> amprxc: shouldn't need any special
flags, rsync does delta transfers so in case of resuming, it'll
continue only chunks that are not there
1243[16:30:58] <ratrace> mandeep: oh so you were asking how to
debootstrap ubuntu...
1244[16:31:05] <mandeep> yeah
1245[16:31:26] <ratrace> well yeah maybe it's better to just
use the installer if you don't have experience with
deboostrapping.
1246[16:31:27] <mandeep> well i asked how to resize an encrypted
lvm drive, but then the deboostrap idea was given to me
1247[16:31:32] <amprxc> @ratr
1248[16:31:42] <amprxc> ratrace oh ok ! thanks !
1249[16:32:04] <ratrace> amprxc: btw... bad idea to use zfs snaps
as files for backups
1254[16:33:14] <amprxc> ratrace I don't understand what you
mean ? I already sent and receiced snaps successfuly
1255[16:33:55] <ratrace> amprxc: also, assuming you'll store
them on an fs that doesn't do integrity checks like zfs or
btrfs, then a bug mid-stream will basically destroy it, and you
_will_ need to apply them all in order to restore
1258[16:34:50] <ratrace> somehow you made the snaps into bz2
files, and you want to transfer the bz2 files.... where?
1259[16:35:25] <ratrace> assuming not to another zfs system (as
that would be very stupid), therefore I assumed you want to store
them as individual files until the time you need to restore from
them
1260[16:35:27] <amprxc> ratrace I did a ZFS send to a file, then
compressed it with bunzip, then encrypted it
1261[16:35:50] <amprxc> ratrace I transfer bz2 files on another
server
1262[16:36:01] <ratrace> amprxc: and this other server also has
zfs?
1269[16:39:06] <amprxc> ratrace I use Proxmox and a ZFS FS. I
export the VM ZFS FS with "zfs send", send it to a file,
then compress it and encrypt it. And I transfer the bz2 file on
another server. I also do that for ZFS snapshots... I don't see
the prob
1279[16:40:40] <ratrace> ALSO... if there's corruption on
the storage here you keep the zfs snaps, then the snaps are
worhtless, esp. if incremental
1280[16:40:49] <amprxc> ratrace that's why I ALWAYS keep a
FULL ZFS backup
1281[16:40:58] <amprxc> ratrace (at least one)
1282[16:41:08] <ratrace> which is worse than simply rsyncing at
the filesystem level and then losing one file, instead of whole
stream, in case of corruption, if you don't have something like
zfs or btrfs on the receiving end
1283[16:41:10] <amprxc> ratrace then I can restore the
incremental snaps
1342[17:47:43] <sgo11_> sney: I want to change the default for
terminal only. I am not familiar with these things. So I am not sure
what will happen if I use zsh for everything. I think the booting
process will need bash?
1343[17:48:21] <sney> you can change your user shell without
changing the system shell. run 'chsh' and then enter
/bin/zsh when it asks
1347[17:50:20] <sgo11_> sney: Will that affect anything except
terminal? I am afraind changing default user shell will affact some
user or system scripts? Thanks.
1354[17:51:26] <sney> plus, most shell scripts specify what shell
to use, so if you have a bash script and you execute it from zsh,
zsh will call bash to run the script
1356[17:52:03] <alexrelis[m]> Hey guys, so I'm putting a CD
in my server and when I try to access it in my file manager in an
X2go session, I am able to access the Audio CD layer but not the
CD-ROM layer.
1361[17:52:51] <sgo11_> sney: got it. thanks a lot. I think I
will need to re-define all environment variables in .zshrc again
right? I defined them in .bashrc before.
1376[17:55:56] *** dominic35 is now known as dominic34
1377[17:55:58] <sney> I don't know offhand but zsh and bash
do share a lot of features. look it up
1378[17:56:01] <sgo11_> If that is possible, I don't need to
swtich to zsh. multi-line paste is really dangerous. I made mistakes
sometimes. Ctrl+Shift+V to paste the wrong thing.
1379[17:56:54] <sgo11_> ok. thanks. I will google this. zsh
supports this out-of-box.
1393[18:02:11] <sgo11_> I found it. Bash supports it now. Just
need to put "set enable-bracketed-paste on" to
"~/.inputrc". Very cool. no need to switch to zsh. Not
very familiar with it.
1394[18:02:59] <jelly> sgo11_, you would typically set env. vars
in .zshenv if they needed to apply to all zsh instances, be they
interactive or not, login shells or not
1395[18:03:18] <sgo11_> GPU based terminal is very fast. I just
started to use it today.
1492[18:58:43] <petn-randall> Lady_Aleena: Your question is quite
vague. Are we talking packages? What do you mean with modified?
Security updates might comes pretty quick, feature updates however
only with a new Debian release.
1493[18:58:44] <Lady_Aleena> yanmaani, I'm on stable. So,
something I really really wanted added to something that was will
not be available to me for a while. ☹
1494[18:58:49] *** Quits: yans (~yans@replaced-ip) (Quit: chaos is the only true answer)
1495[18:59:24] <yanmaani> Lady_Aleena: Consider using backports,
maybe
1500[18:59:38] <dpkg> Insufficient Data For Meaningful Answer
1501[18:59:40] <yanmaani> or, depending on what it is, they might
have their own repo
1502[18:59:49] <Lady_Aleena> petn-randall, I asked for a new
option for the linux-util column. My wish was granted, but I wanted
to know how long it would take before I can use that neat new
option.
1504[19:00:48] <sney> I'd say look at tracker.debian.org to
see if the update is in sid yet, but tracker is down
1505[19:00:53] <petn-randall> Lady_Aleena: Probably next release
then, or you might get it via stable-backports.
1506[19:00:53] <jelly> hmm. Should update-pciids from pciutils
use dpkg-divert instead of just overwriting /usr/share/misc/pci.ids
1507[19:01:09] <jelly> it's a packaged file after all
1508[19:02:11] <Lady_Aleena> petn-randall, I will wait until it
is in stable. It's been so long since I installed anything from
backports I've forgotten how to do it.
1535[19:10:03] *** Quits: Dara (~Dara_Simc@replaced-ip) (Ping timeout: 240 seconds)
1536[19:10:14] <Lady_Aleena> jelly, I find that when I start
trying to write a Perl one-liner, I end up writing a full blown
script with more bells and whistles than I originally intended. The
linux utilities keep me from going down rabbit holes.
1597[19:56:06] <greycat> use "lspci -nn" to see what
your video devices are
1598[19:56:10] <sney> typically, install
'firmware-amd-graphics' and the kernel/xorg will handle
the rest automatically
1599[19:56:26] <diogenes_> there used to be ati, fglrx.
1600[19:56:41] <Eryn_1983_FL> 0300 amd
1601[19:56:51] <Eryn_1983_FL> apd family 15h
1602[19:57:08] <Eryn_1983_FL> models 60h-6fh io memory management
unit
1603[19:57:21] <Eryn_1983_FL> [12033:2576]
1604[19:57:33] <Eryn_1983_FL> 1022:1576
1605[19:57:41] <sney> ,pciid 1022:1576
1606[19:57:42] <judd> [1022:1576] is 'Family 15h (Models
60h-6fh) Processor Root Complex' from 'Advanced Micro
Devices, Inc. [AMD]' with kernel modules 'ahci',
'sdhci-pci', 'snd-hda-intel' in stretch. See
also
replaced-url
1607[19:57:52] <sney> welp
1608[19:57:58] <Eryn_1983_FL> sorry
1609[19:58:13] <Eryn_1983_FL> 1002:98e4
1610[19:58:17] <Eryn_1983_FL> rec c0
1611[19:58:19] <Eryn_1983_FL> rev
1612[19:58:27] <sney> Eryn_1983_FL: install
'firmware-amd-graphics' from non-free and reboot. if
you're still having video problems after that, provide details
1613[19:58:49] <Eryn_1983_FL> kk
1614[19:58:57] <Eryn_1983_FL> ty man i knew it was amd something
1615[19:59:01] <Eryn_1983_FL> installing now
1616[19:59:50] <greycat> Drivers and firmware are different
things. Drivers are loaded into the kernel and run on your CPU.
Firmware is loaded into the miniature computer on the device, and
runs on the device.
1621[20:04:33] <Eryn_1983_FL> how do i get a uuid lsblk?
1622[20:04:39] <sney> blkid
1623[20:05:28] <Eryn_1983_FL> not installed
1624[20:05:53] <greycat> there's also lsblk -o +UUID
1625[20:05:58] <Eryn_1983_FL> was not in path lol
1626[20:06:04] <sney> !sbin
1627[20:06:05] <dpkg> Some binaries, particularly system
utilities are installed to /sbin or /usr/sbin rather than /bin or
/usr/bin. The former directories are not in a standard user's
PATH. Try calling them directly, e.g. /sbin/ifconfig, or running
them as root.
1632[20:10:17] <jhutchins> Eryn_1983_FL: When working with
programs that affect the whole system rather than just one user, you
generally need to be root.
1635[20:10:40] <Eryn_1983_FL> /usr/sbin/ was not in path like he
said
1636[20:10:44] <sney> !buster su
1637[20:10:44] <Eryn_1983_FL> didnt know that
1638[20:10:44] <dpkg> In buster, su no longer overrides PATH by
default, requiring that you use "su -" or "su
-l" for login shells (which is not really a new thing at
all...). To approximate the previous behaviour, put
"ALWAYS_SET_PATH yes" in /etc/login.defs. See
replaced-url
1639[20:10:57] <Eryn_1983_FL> ok
1640[20:11:12] * sney resets the "days since someone admitted to
using su wrong all this time" counter
1642[20:12:16] <greycat> They're not using it wrong. Debian
broke it and refused to consider how much pain they were causing,
even after we told them, over and over.
1645[20:12:54] <ndegruchy> I use su and sudo wrong all the time.
But I can remember stupid `tar` arguments, no problem
1646[20:13:06] <greycat> Being Like Red Hat was more important to
them than supporting their users, or being backward compatible with
themselves.
1647[20:13:32] <jhutchins> greycat: What other distributions
(besides ubuntu) had the cludge?
1648[20:13:47] <sney> I've always been in mixed environments
so I've always used su the non-debian way then, I guess.
1649[20:13:48] <greycat> It was not a kludge! It was the way su
was SUPPOSED to work!
1650[20:14:14] <jhutchins> greycat: So who else did it that way?
1651[20:14:16] <greycat> Debian used a different source code for
su before buster. One that worked the way most commercial Unixes
work. One that works sensibly.
1657[20:18:16] <greycat> The fact that the workaround is to edit
some other program's config file, which causes the OTHER
program to emit a warning, is also reprehensible.
1658[20:19:31] <SpeedyG> hi, i've got two usb sticks which
both seem to have a similar issue so i'm not sure if its the
ubs sticks or the usb port.... when using dd, it says it wrote
3+GB/s to it and is done in less than a second but the partition
table of the usb stick is all messed up... is there a way to fix it?
1664[20:22:44] <SpeedyG> doesnt do a thing unfortunately...
system still thinks there's 3 partitions on there
1665[20:22:45] <sney> iirc there's a dd option that writes
directly without caching, which makes this unneccessary, but I
don't remember what it is and I always just sync after
1669[20:24:31] <SpeedyG> dd'ing /dev/zero to the device also
does not 'clean' anything... just finishes in a
millisecond and doesnt change a thing on it...
1670[20:25:02] <sney> that sounds like bad flash then.
1671[20:25:17] <greycat> or you're using the wrong device
1672[20:25:37] <sney> or that. or some kind of read-only switch
that's not visible in software?
1673[20:26:04] <greycat> hardware write-protect should give you
an error, though
1674[20:26:06] <SpeedyG> hm, my guess would be bad flash then...
1751[21:07:18] <oxek> I have a mystery to solve. I am on debian
stable, and run the backports 5.7 kernel. When I look at the package
'wireguard', it depends on wireguard-dkms. But when I
install wireguard, it does not install the wireguard-dkms
dependency.
1775[21:17:08] <SpeedyG> jhutchins: windows to the rescue.. after
inserting the usb stick into a windows laptop, it started yapping
about unreadable disks... reformatted it to fat32 7.8G and, after
another reboot of my debian box, I was finally able to dd the image
and use the usb stick as bootable debian stick
1777[21:17:49] <ratrace> oxek: I think that's a bit of a red
herring. there are package that won't automatically install
their dkms members. zfs or nvidia, you need to explicitly pull in
the dkms packages too
1778[21:18:22] *** Quits: Jerrynicki (~niklas@replaced-ip) (Remote host closed the connection)
1779[21:18:28] <oxek> ratrace: I don't have dkms installed
though.
1780[21:18:43] <oxek> and the installation was very quick, there
was no building kernel module step
1781[21:18:48] <ratrace> well that's my point, packages with
-dkms need to have them installed explicitly
1782[21:19:09] <oxek> but when I try that on a debian-stable with
the usual kernel, it pulls in those dkms dependencies
1783[21:21:08] <ratrace> are you sure it doesn't do that
with the bpo'd kernel too? does here
1784[21:21:32] <oxek> I'm absolutely sure, no dkms
installed.
1785[21:21:37] <ratrace> apt install wireguard pulls in
wireguard, wireguard-tools and wireguard-dkms
1788[21:24:16] <ratrace> huh wait, I'm on stable kernel, I
thought I was no bpo'd
1789[21:24:38] <oxek> I have a theory
1790[21:25:08] <oxek> If the backports kernel had a note
'Provides: wireguard-modules' then it would make sense
1791[21:25:13] <oxek> but I don't see such a note
1792[21:25:26] <ratrace> but that just can't be. assuming
you're installing wireguard-1.0.20200513-1~bpo10+1, the
dependency on -dkms is ther expressed in the package,
unconditionally
1793[21:25:39] <ratrace> you sure you don't have some other
version there?
1794[21:25:46] <ratrace> from bullseye or sid that maybe altered
behavior?
1796[21:26:19] <ratrace> (actually dependency is on
wireguard-tools which in turn wants wireguard-dkms)
1797[21:26:25] <oxek> this is a fresh installation of debian
stable, where I then installed the backports kernel using 'apt
-t buster-backports install linux-image-amd64' and then
installed wireguard
1798[21:26:58] <ratrace> oxek: did you reboot after installing
bpo'd kernel?
1842[21:42:03] <greycat> "Transfer a file to someone who has
a smart phone" doesn't necessarily say that the file must
be *stored* on the phone in the end. You could just hand the person
a CD with the file burned on it. Or USB stick for you younguns.
1843[21:43:01] <ratrace> seriously greycat, a CD. those things
went extinct even before we sow our lawns.
1844[21:43:08] <Tyszka> ratrace, its a friends phone who is
remote country
1845[21:43:24] <oxek> I made the right assumption then
1846[21:43:31] <jhutchins> Tyszka: email usually works.
1847[21:43:39] <oxek> not for 100MB
1848[21:43:44] *** Quits: Vizva (~Vizva@replaced-ip) (Remote host closed the connection)
1849[21:43:50] <ratrace> email _should_ work for 100MB in 2020
1850[21:43:56] <ratrace> esp. if freemail
1851[21:44:03] <oxek> should, but I'd not rely on it
1852[21:44:09] *** Quits: yuta (~pi@replaced-ip) (Quit: WeeChat 2.9)
1853[21:44:12] <Tyszka> doesn't gmail allow 4gb ?
1854[21:44:13] <jhutchins> Compressed.
1855[21:44:18] <oxek> in theory, there's no limit to
filesize in the RFCs, is there?
1856[21:44:19] <Tyszka> sorry to ask
1857[21:44:20] <greycat> ratrace: I still have quite a few left
on a few different spindles. Sadly, the burner in my PC stopped
working a little while back, but I do have a USB CD burner I can
hook up in a pinch.
1858[21:45:05] <ratrace> I diched mine with the 2015 computer
upgrade. first one without a cdrom or floppy
1859[21:45:12] <jhutchins> oxek: AFIK there isn't, but in
practicality you're going to hit limits on relay boxes.
1860[21:45:28] <Tyszka> oh gm is up to 50M
1861[21:45:28] <jhutchins> oxek: I think I've managed 20M.
1862[21:45:34] <oxek> jhutchins: yeah, and those relay boxes can
have a limit as small as 20MB
1863[21:45:44] <jhutchins> oxek: I've seen 10M.
1864[21:45:59] <oxek> I remember when freemail had 10MB free
1865[21:46:21] <oxek> which was like thousands of emails, so who
would ever need more?
1878[21:49:18] <oxek> Tyszka: what's preventing you from
using
replaced-url
1879[21:49:36] <oxek> no ads, no captchas, no accounts, 512MiB
limit
1880[21:49:45] <Tyszka> oxek, link looks scary
1881[21:49:58] <oxek> isn't it a null byte?
1882[21:51:15] *** Quits: dreamon (~dreamon@replaced-ip) (Remote host closed the connection)
1883[21:51:34] <ratrace> oxek: okay I think I found it
1884[21:51:49] <ratrace> first, indeed, doesn't pull in
-dkms with 5.7.0 bpo'd kernel
1885[21:52:11] <ratrace> second, I think that's because it
depends on wireguard-modules OR wireguard-dkms, BUT the 5.7.0 kernel
_provides_ wireguard-modules virtual
1888[21:53:36] <ratrace> so the only magick here is that the 5.7.
kernel is providing the virtual that wireguard depends upon
conditionally in "OR" relationship with wireguard-dkms
1889[21:53:54] <ratrace> in other words, the kernel carries the
modules so dkms is not needed. which isn't the case with 4.19
stable kernel
1902[21:56:35] <cVsup> somebody can say what this error? squid3:
symbol lookup error: squid3: undefined symbol:
_ZN7libecap4NameC1ERKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE
1903[21:56:45] <oxek> maybe I should run all output through
lolcat, to give it some color distinguishing
1914[22:00:33] <cVsup> depend libecap2 and libecap3
1915[22:01:00] <greycat> I still suspect you've somehow
mixed up packages from other releases, or upstream binaries.
1916[22:01:39] <greycat> Where do you see the error? How do you
reproduce it? How does squid attempt to start on your system? Are
there multiple installations of squid?
2012[23:23:55] <tomreyn> an old php web mail application
2013[23:24:35] <UncleCidd> It's taken me a good while for me
to find this solution... is this the only workaround the people here
are aware of off the top of your head?