3[00:04:23] <CyberManifest> foul_owl_: your gnome-clocks
reminded me of the slashdot quote of something similar to:
"I'd rather imagine little elves inside doing all the
work."
16[00:10:33] <oxek> realistically, what is the point of using
'most' as a pager instead of 'less'? Less is at
least being updated whereas most seems dead.
17[00:12:53] <tomreyn> was the response to this question when
you asked it on ##linux not satifying?
18[00:13:27] *** Quits: tijara (~tijara@replaced-ip) (Disconnected by services)
19[00:14:02] <tomreyn> generally, people have different
preferences, and alternatives are always good to have available.
38[00:25:05] <sponix> oxek: having both "more" and
"less" wasn't enough - so they added "most"
?
39[00:26:25] *** Quits: dionysus69 (~Thunderbi@replaced-ip) (Remote host closed the connection)
40[00:26:57] <oxek> yeah, but at least I get that
'more' is stripped down to occupy as little space as
possible (for embedded, etc.), and 'less' has all the
features possible.
41[00:27:03] <oxek> I don't see where 'most'
fits in.
42[00:27:17] <sponix> oxek: other than a bad joke ?
43[00:27:33] <oxek> I do see the joke value
44[00:27:46] <sponix> On Ubuntu they gave into the newbs and
created a "rename" command instead of just using mv from
one name to another :P
45[00:27:54] <oxek> I like how it colors manpages when used as
a pager for man, but I can probably get that done with
'less' too.
46[00:27:57] <sponix> maybe most falls in this category
47[00:28:53] <oxek> 'rename' (both the perl version,
and the version from util-linux) both have more functionality than
'mv' though.
48[00:29:25] <oxek> unless ubuntu made something that is
literally an alias to mv
49[00:30:02] <sponix> not sure, never cared to research it
further. just had some new user swear it existed, so I verified that
it does
151[02:13:20] <Case_Of> hi, what package could contain perl
Mozilla::CA library?
152[02:13:23] *** Quits: Dara (~Dara_Simc@replaced-ip) (Ping timeout: 240 seconds)
153[02:15:02] <alexrelis[m]> I'm having a strange problem
on my sister's laptop. So I replaced the keyboard and fan
recently, and for whatever reason on Debian, the system shuts down a
few seconds to a minute after boot. This doesn't happen on
Windows. And it doesn't happen when I leave the BIOS settings
running.
154[02:15:17] <alexrelis[m]> And it's not an abrupt
shutdown. It's simply a graceful shutdown, almost as if I
pressed the power button.
160[02:18:39] <alexrelis[m]> Well, sometimes it doesn't
turn off to the point where I can type in the username. But then it
shuts off soon after. And then there are times where it shuts off
when systemd boots.
161[02:19:42] <short-bike> Anything logged in dmesg ?
216[03:15:55] <sney> got the virtualbox factoid updated at
least. the old one read too much like an endorsement.
217[03:15:57] <sney> !virtualbox
218[03:15:58] <dpkg> Oracle VM VirtualBox is a software virtual
machine solution. Not in buster and unlikely to be in any future
debian stable releases due to #794466. 3rd party packages are
available from Oracle, see
replaced-url
255[03:46:39] <sney> the main point is that if someone wants a
point-and-clicky VM manager in debian, virt-manager should be their
first choice, but if they need virtualbox-brand virtualbox, they can
go to that wiki link.
256[03:47:08] <sney> they don't need a oneliner kludge from
dpkg explaining how to add a 3rd party repository.
257[03:48:21] <somiaj> Yup, I just found it strange when I read
it, but from someone who just wants an easy to install vm system it
is the way to go, though isn't gnome-boxes becomming more
popular too?
278[03:58:58] <sney> in all except hyper-v, really
279[03:59:16] <ryouma> the main thing i was doing was trying to
install debian while running debian. using a pass through something
or other to disk. it was able to do that. and i was able to get it
to hav its menus and so on use large fonts.
280[03:59:20] <sney> not specifically to do with font size, but
some display issues, and disk or network i/o as well
281[03:59:59] <Casper26> I installed firefox 80 in /opt/firefox
how can i get it to launch in terminal by just typing firefox? I
already removed firefox-esr.
282[04:00:07] *** Quits: magic_ninja (~sparkie1@replaced-ip) (Remote host closed the connection)
283[04:00:36] <sney> symlink the firefox binary to
/usr/local/bin, is probably what I would do
284[04:01:42] <sney> mozilla may have an official recommendation
on their website.
287[04:03:10] <ryouma> (installing debian from cd is a burden
for me because of the need to physically put the cd in and the
bright blue and bright white and the small fonts and the fact that i
can't do other things at the same time and so on; i was able
to, althought perhaps slighly brokenly, get that to work. it was
good, but a lot of effort.)
316[04:39:50] <sney> if you're having trouble with a
missing dependency, check out equivs. otherwise, actually fix the
problem.
317[04:40:23] <GaiusCaesar> yeah, I kludge most things, afai
recall --nodepends was a way to install something and not get all
the extra libs or so, eg kde app on buster
318[04:40:27] <GaiusCaesar> kalarm
319[04:41:35] <GaiusCaesar> it wants 64MB
320[04:42:22] <somiaj> you want --no-install-recommendeds, but
expect some things to not work as a result, there is also an
apt.conf option to disable recommends by default (not recommended)
324[04:54:50] <ryouma> there is also the option of periodically
purging stuff like htat. here is my amateur script: apt-get -o
"Apt::AutoRemove::SuggestsImportant=false" purge `aptitude
search ~c | db wsv2o` `deborphan ${1:+--guess-all}`
325[04:55:04] <ryouma> plus autoremove which workssometimes
326[04:55:39] <somiaj> I think there is some packages that use
recommended when it should be suggested, but in general by
defaintion most will want recommended packages.
332[05:00:48] <fuxxy> I got strace installed on my openwrt
device, but there's a perl project to help decode strace. Am I
correct in the assumption that I can pipe the strace output to my
debian box to the perl script?
333[05:02:12] <sponix> I didn't know strace was encoded
334[05:03:38] <fuxxy> it's "encoded" as in
escaped C, from my understanding
335[05:04:23] <sponix> you have something crashing ?
336[05:06:01] <fuxxy> It's not crashing. It's a closed
source program that's hammering my NTP server. I'll keep
reading on my own.
337[05:06:46] <sponix> fuxxy: interesting
338[05:11:39] *** Quits: Tom01 (~tom@replaced-ip) (Remote host closed the connection)
349[05:14:08] <somiaj> fling: sounds like an XY problem, what
issue are you having. If running debian stable things should just
work. If you are having trouble with apt, please provide the
following information
350[05:14:12] <somiaj> !bat
351[05:14:12] <dpkg> [Basic Apt* Troubleshooting]. To diagnose
your problem, we need ALL OF THE FOLLOWING information: 1. complete
output of your apt-get/apt/aptitude run (including the command used)
2. output from "apt-cache policy pkg1 pkg2..." for ALL
packages mentioned ANYWHERE in the problem, and 3. "apt-cache
policy". Use
replaced-url
352[05:14:13] <gry> fling: are you speaking with me?
354[05:14:25] <fling> gry: yes, I suggest trying lxd instead
355[05:15:13] <fling> somiaj: I want to have few config files
mounted readonly and I don't want apt to worry about not being
able to rewrite them or to make it to not even try writing
360[05:16:25] <sponix> gry: pretty sure on my rig, putting my
user in the libvirt and libvirt-qemu groups helped me do things
361[05:16:42] <somiaj> fling: first dpkg (since that is the
actual programing installing) won't overwrite config files
unless you give it permission. So if you make modifications to
configuration files, dpkg will leave them alone. What config files
do you have mounted read only?
362[05:16:49] <gry> sponix: ok :)
363[05:16:49] <sponix> fling: a config file can be made to be
just read only, doesn't need to be mounted on a read only file
system for that
364[05:17:32] *** Quits: gry (~test@replaced-ip) (Remote host closed the connection)
365[05:17:33] <fling> sponix: I know about chattr +i, but the
config files will be mounted from another host
366[05:17:40] <fling> sponix: thanks for the info!
367[05:17:45] <somiaj> gry: also be careful, are your vms in the
user or system uri (there is qemu://user and qemu://system, and
virt-manager will use system by default and thus as sponix pointed
out you need to be in the proper groups)
368[05:17:46] <jmcnaught> You only need to be in the libvirt
group.
369[05:18:16] <somiaj> sory qemu:///session (Which is for the
user)
371[05:18:59] <somiaj> I personally have to set export
LIBVIRT_DEFAULT_URI="qemu:///system" to make sure virsh by
default access the system (which is what virt-manager uses by
default)
372[05:19:42] <sponix> somiaj: is that better ?
373[05:19:48] <somiaj> fling: can you give an example of the
configuration file a package is trying to overwrite, (just want to
make sure it is a config file regerested with debconf)
374[05:20:38] <fling> somiaj: don't have one yet, still
planning.
375[05:20:46] <fling> I will tell you more if/when I hit an
issue.
376[05:20:52] <somiaj> sponix: My understanding is it is more a
management decision. When I first started using vms, I installed
with root, so had to change the URI so my user could also access
them.
379[05:22:03] <somiaj> fling: in general actual config files (if
you modify them, so make sure they are different even if just you
just put in a comment), dpkg will not overwrite them once the
package is installed. It will aks you what to do, and then create a
configfile.dpkg (or something like that) with the system one and
leave your user one alone.
380[05:22:22] <somiaj> so just mount the config files after the
install of the package (there maybe ways to tell debconf about the
config ahead of time, but not certain here)
381[05:23:02] <sponix> somiaj: hmm, I've always been under
the impression from a security standpoint doing things with the
least privileged account possible was better. And normally things I
put in VMs are to keep them contained. Thus running them as a user
in user mode vs system/root
382[05:23:57] <somiaj> sponix: I think if either root or the
user does the work, in the end it is libvirt user's permissions
the vm has. It is just a matter of which vms virsh can see.
383[05:25:05] <fling> somiaj: ok got it
384[05:25:23] <jmcnaught> libvirt runs the VMs in qemu:///system
as the user & group libvirt-qemu, so it gets reduced privileges,
but you don't want your user to be in the libvirt-qemu group,
only the libvirt group is needed to control libvirt.
385[05:25:45] <sponix> somiaj: interesting
386[05:26:01] <somiaj> guess I was mistaken.
replaced-url
387[05:26:24] <somiaj> seems virt-manager defaults to
qemu:///system just because that is what it was doing before
qemusessions were worked out.
388[05:26:57] <somiaj> I was more just pointing out that if the
vms are in qemu:///session then virt-manager won't see them by
default (though wasn't very clear about that)
390[05:29:49] <somiaj> though at least on debian, you don't
need the root password to access qemu:///system, being in the
correct groups is fine.
391[05:32:32] <somiaj> (reading that over, I think I ended up
using qemu:///system due to networking and setting up the bridge, as
that was the easiest way I found at the time)
392[05:32:58] <jmcnaught> Yeah you're limited to user mode
networking with qemu:///session
393[05:34:02] <sponix> jmcnaught: what "limits" does
that impose ?
394[05:34:06] <sponix> vs system
395[05:35:46] <jmcnaught> sponix: with user mode networking you
will not be able to access the guest from the host. The guest will
be able to access the host and the internet, but this wouldn't
work for something like learning apache2 in a VM.
396[05:35:57] <somiaj> sponix: the limits seem to be that you
end up having to become root to configure the networking, then the
session can access it.
431[06:25:07] *** Quits: dez (uid92154@replaced-ip) (Quit: Connection closed for inactivity)
432[06:26:09] <somiaj> okay, I think I am just in that group as
I fumbled around figuring out what groups I needed, and then have
just coppied it over. Thanks
433[06:26:48] <jmcnaught> On your own system there's
probably little risk, my point was more that only libvirt group
membership is needed.
437[06:29:21] <somiaj> Understood, but would rather limit the
groups I'm in if it isn't needed. I have just been using
the groups I found that work without to much tought, which is
probably poor decision on my part.
635[09:53:23] <Dungeon> [01:56] <Dungeon> are there any
hardware techs in the house?
636[09:53:23] <Dungeon> [01:56] <Dungeon> i got a 1tb
drive that is detecting in the bios and is in deed spinning up but
it comes back with 0 size within the drive policy.
637[09:53:23] <Dungeon> [01:57] <Dungeon> however it
passes the smart check and quick and long checks. it just wont even
start to write to the drive itself
642[09:56:26] <kale> hi guys, would you happen to know if mbr
supports partitioning for a 4TB disk?
643[09:56:52] <klys> no, that goes up to 2 TiB.
644[09:57:07] <klys> hybrid isn't an option either if you
use windows
645[09:57:14] <kale> klys: ok thanks. is the viable alternative
gpt then?
646[09:57:31] <klys> kale, yes, depending on how viable you need
647[09:59:02] <kale> klys: i think i will gradually switch to
gpt for partitioning, many years ago i saw i bios that could boot
mbr, but not gpt, so i wanted to wait some years before switching. i
do not think problems like that happens on newer systems.
648[09:59:13] *** _Posterdati_ is now known as Posterdati
649[10:00:58] <kale> what partition type do i choose for luks
encryption? "linux-filesystem"?
650[10:02:41] *** Quits: maggotbrain (~maggotbra@replaced-ip) (Remote host closed the connection)
651[10:02:45] <wyoung> kale: Neither, you setup the encryption,
a new device is creted under /dev then you create a lvm volume on it
or a single file system.
652[10:03:23] <wyoung> kale: There is a special partition type
for luks you can use but you dont have to.
653[10:03:32] <gry> how do i enable webcam and mic access in
virt-manager?
654[10:03:53] <kale> wyoung: well, i will have to set up the
encryption one something. here that is a partition in gpt. and for
that partition i need to choose the type. are you saying that i will
choose the type for whatever is encrypted?
655[10:04:59] <wyoung> kale: There is a type for encryption or
use linuc, 83 or 82 what ever the hex code for it is
656[10:06:46] *** Joins: mezzo (~mezzo@replaced-ip)
657[10:07:00] <kale> the closest thing i can find in cfdisk on a
gpt partition is "linux server data" or "linux
filesystem"
660[10:10:12] <ratrace> kale: it really doesn't matter,
luks has its own identification headers
661[10:11:09] <kale> ratrace: ok i'll stick to "linux
filesystem" then. i just thought that this being the new
partitioning scheme, would have such type.
662[10:11:52] <ratrace> kale: well there is, e8 is for LUKS, and
83 is generic linux filesystem.
692[10:29:16] <ratrace> kale: well I'd argue that the
"correct" thing here is that in general those guids are
rarely used, and one shouldn't bother with distinction within
"linux filesystem" group. there are few exceptions like
efi and bios_boot being marked specifically.
693[10:29:34] <ratrace> eg, I have yet to see a linux distro
installer set these other than generic linux filesystem
695[10:31:08] <kale> ratrace: ah ok, so this might also affect
applications that does not know about the luks partition type, so
they will be unable to handle it
696[10:31:21] <kale> like cfdisk does not have that type
697[10:32:24] <ratrace> kale: not sure what you mean by
"not being able to handle it", unless you just mean set or
change the partition type GUID
698[10:33:10] <ratrace> but if you really wanna bother, you can
set the type guids with gdisk. list of guids:
ttps://wiki.archlinux.org/index.php/GPT_fdisk#Partition_type
703[10:35:22] <ratrace> the type for bios boot efi are important
so the firmware and boot loader can directly see it without needing
to carry complex autodetection code. but once /boot is found, linux
doesn't care any more, it uses filesystems, lvm headers, mdadm
headers, luks headers, to autodetect. kinda like it doesn't
care about filename extension but uses mimetype magic
706[10:36:12] <kale> yes, i really hate that efi stuff
707[10:38:04] <ratrace> although with luks specifically, the
userland needs to be told what to do (eg, you need to specifically
mark and state the rootfs luks container), because it can't
assume contents without unlocking
708[10:38:27] <ratrace> this btw is done automatically by grub
and initramfs hooks
709[10:38:41] *** Quits: Kuttenbrunzer (~Kuttenbru@replaced-ip) (Remote host closed the connection)
710[10:39:29] <ratrace> you only need /etc/crypttab entries to
tell initramfs what to unlock on boot
804[12:55:51] <leafwiz> I found this on an Arch wiki "To
start the daemon for a specific interface alone, start/enable the
template unit dhcpcd@interface.service, where interface can be found
with Network configuration#Listing network interfaces."
805[12:56:31] <leafwiz> i tried systemctl enable
dhcpcd@wlan0.service
830[13:23:47] <oxek> Can somone help me reproduce a bug in tmux?
I found a way of making tmux stop responding, occupy 100% of one CPU
core, and only be able to be killed by kill -9, not by SIGTERM.
831[13:23:50] <oxek> These are the rough steps:
replaced-url
842[13:39:50] *** Quits: mezzo (~mezzo@replaced-ip) (Quit: leaving)
843[13:42:25] <mi11k1> nebody up to date with android boxes? I
have old Jynxboxlive. On the chip is printed "Rockchip
RK3066A" . I would like to flash it with Debian
844[13:43:02] <mi11k1> do I just need custom recovery to move
ahead?
896[14:43:11] <ksk> Have you looked at
replaced-url
897[14:44:12] <leafwiz> Yes, I'm just trying to do it the
systemd way. But after being guided by #systemd it does not seem to
work.
898[14:44:14] <ksk> see the Remark "Bridging with a
wireless NIC" also, still not 100% sure what you are trying to
do here..
899[14:44:24] <ksk> !does not work
900[14:44:25] <dpkg> "Doesn't work" is a vague
statement. Does it sit on the couch all day long? Does it
procrastinate doing the dishes? Does it beg on the street for
change? Please be specific! Define 'it' and what it
isn't doing. Give us more details so we can help you without
needing to ask basic questions like "what's the error
message?". Ask me about <smart questions>, <sicco>
and <errors>.
902[14:44:45] <leafwiz> Okay, maybe that is not a good idea
903[14:44:53] <leafwiz> to bridge with a wlan int
904[14:45:00] * zerocode is away: (automatically dead ) [BX-MsgLog On]
905[14:45:00] * zerocode is idle, automatically dead [bX(l/on p/off)]
906[14:45:03] <leafwiz> Just thoguht that it would be the same
lan
907[14:45:13] <leafwiz> for my dhcpd
908[14:46:15] <leafwiz> But I have decided after three hours
fiddling with systemd and debian that I will just install devuan and
do it the classic way.
927[14:55:13] <tomreyn> this is usually due to an invalid (at
the time) network configuration, such as depending on other network
devices which are not yet available or configured, or involving
hooks which aren't available at this boot stage
1041[16:12:04] <Eryn_1983_FL> is this somebody trying to break
in?
1042[16:12:06] <Eryn_1983_FL> solan sshd[20242]: Unable to
negotiate with 103.133.110.117 port 38776:
1043[16:12:08] <Eryn_1983_FL> no matching key exchange method
found. Their offer: diffie-hellman-group1-sha1 [preauth]
1044[16:14:08] <cluonbeam> Eryn_1983_FL: Looks like they're
using deprecated crypto settings. PS: If you have a system exposed
to the internet, people are always trying to break in. Use secure
passwords and don't worry about them. Or, better yet, set up
multi-factor authentication, or prohibit password logins entirely
and use only SSH keys.
1045[16:15:19] <Eryn_1983_FL> i use keys
1046[16:15:24] <Eryn_1983_FL> and i upgrade regular
1047[16:15:30] <Eryn_1983_FL> and i got fail2ban
1048[16:15:38] <Eryn_1983_FL> not sure why i got like 10 of those
though
1049[16:17:21] <cluonbeam> It'll happen. Even fail2ban, in
my opinion, is of minimal value.
1050[16:17:48] <oxek> SanchoPensa: espeak-ng or festival
1051[16:17:50] <Eryn_1983_FL> yeah they keep hopping ips on their
global botnets..
1052[16:17:54] <oxek> SanchoPensa: both are pretty bad though
1053[16:18:01] <cluonbeam> Eryn_1983_FL: That and why care about
failed logins?
1054[16:18:03] <Eryn_1983_FL> wish i had a global botnet, for
good..
1055[16:18:08] <Eryn_1983_FL> i was curious
1056[16:18:13] <SanchoPensa> oxek: heh, thanks
1057[16:18:13] <Eryn_1983_FL> i was looking at the log
1058[16:18:25] <cluonbeam> Eryn_1983_FL: Oh, sorry, that's
not what I meant. Why take any action on failed logins?
1059[16:18:25] <SanchoPensa> is there really nothing, that is
actually usable?
1060[16:18:29] <cluonbeam> e.g. fail2ban
1061[16:18:44] <Eryn_1983_FL> to keep em from trying something
else.
1062[16:18:55] <Eryn_1983_FL> so maybe they start with ssh and
then try to go attack http https
1063[16:19:04] <Eryn_1983_FL> or another one of my ports,
1064[16:19:06] <cluonbeam> That's not really how that works.
1065[16:19:28] <Eryn_1983_FL> then how does it work cluonbeam
1066[16:19:55] <cluonbeam> They're broad sweeps across the
internet, looking for weak SSH credentials. They're looking for
low-hanging fruit. They're automated in that search. No one is
going to expend the energy trying to break into YOUR web server if
basic SSH attacks don't work.
1067[16:20:12] <cluonbeam> You'll see that in the way they
try common usernames.
1068[16:20:16] <Eryn_1983_FL> well if their pissed off at me they
might
1069[16:20:35] <cluonbeam> A random person in
China/Russia/Korea/insert other country doesn't even know who
you are.
1070[16:20:42] <Eryn_1983_FL> my so tends to piss off people on
the net and our cable modem has gone out several times...
1071[16:20:48] <cluonbeam> And you can't really piss off a
botnet unless we've made GREAT strides in AI development.
1072[16:21:04] <cluonbeam> Eryn_1983_FL: Again, no one is going
to expend the energy. You are not a valuable target.
1073[16:21:11] <Eryn_1983_FL> well 9/10 are botnets
1095[16:53:18] <oxek> sometimes, when a manpage for a command is
not available, I am told "See 'man 7 undocumented'
for help when manual pages are not available."
1096[16:53:48] <oxek> but 'man 7 undocumented' says
"No manual entry for undocumented in section 7"
1097[16:54:21] <oxek> what am I supposed to do?
1098[16:54:40] <ratrace> "exception X raised. in addition,
execption Y raised when handling the X exception"
1099[16:55:16] <ratrace> oxek: btw, which manpage is missing,
aside from undocumented(7)?
1106[17:01:15] <ratrace> oxek: first of, it's firefox-esr
manpage (`man firefox-esr`)
1107[17:01:17] <oxek> (I know there's man firefox-esr,
I'm using this as a demonstration that 'man 7
undocumented' appearently leads nowhere)
1108[17:01:50] <ratrace> right. and frankly this is the frist
time I see that comment, about undocumented(7). eg `man
dfjghdfklghdfklgfj` doesn't show it
1109[17:02:18] <oxek> yeah I tried `man adsdfasdfasdf` and it did
not show either
1114[17:04:07] <ratrace> (though it's in section 1)
1115[17:04:41] <oxek> yeah, that's weird too
1116[17:04:59] <ratrace> oxek: you seem to be hitting all these
weird edge cases lately :) unlisted but installable packages,
missing manpages singled out with undocumented(7) suggestions, ....
:)
1117[17:05:09] <oxek> I know :(
1118[17:05:47] <oxek> put me in front of a working computer and
within 30 minutes I'll find a bug
1126[17:17:25] <kittonian> need some serious help really fast
please. i'm freaking out. i was backing up a file into a
tarball and typed the command backwards and now the file I needed to
backup is like 45 bytes instead of over 4gb
1127[17:17:35] <oxek> #506653 #39830 #32019 related
1134[17:19:07] <oxek> kittonian: do you have any previous
backups?
1135[17:19:34] <kittonian> obviously not for a while
1136[17:19:48] <kittonian> that's what I found today and why
I was making the backup
1137[17:20:02] <oxek> honestly, if it is something of
irreplaceable value, contact one of those data recovery services and
expect to pay $$$$
1138[17:20:11] <oxek> there's little you can do on your own
1139[17:20:32] <oxek> and any attempts to recover the data can
make stuff even worse, like playing around with testdisk
1140[17:21:58] <oxek> sorry for your loss
1141[17:24:07] <ratrace> The making of a data backup proponent,
in action :)
1142[17:25:19] <ratrace> for your next backup system setup, I
recommend a ZFS mirror. or you can use one disk but then I'd
recommend copies=2. doesn't help if whole disk fails, but does
offer redundancy in case of corruption
1194[18:36:15] <sney> it's not clear why 3.0 hasn't
been packaged yet, but they are apparently 2 separate libraries that
can be installed simultaneously, according to this wishlist bug:
replaced-url
1202[18:38:49] *** Quits: usr1987 (~usr1987@replaced-ip) (Remote host closed the connection)
1203[18:39:41] <sney> you can go to #debian-mentors on OFTC if
you're interested in packaging it yourself. I know it's
not usually recommended for a library to be someone's first
package, but if you're familiar with it, that might not be an
issue
1248[18:54:54] <sney> you could send the authorities a
screenshot. hard to know the jurisdiction of random stuff on the
internet but maybe interpol is a good place to start?
1249[18:55:34] <cluonbeam> If the service is hosted in the US,
the FBI/NCMEC (national center for missing/exploited children) has
an online portal for these things.
1250[18:55:50] <coconut> sounds well
1251[18:56:09] <sney> I'm pretty sure paste is hosted in
germany
1256[18:57:38] <cluonbeam> I wouldn't be surprised if the
German government has some similar facility. Interpol might not be a
bad idea either, as sney said.
1258[18:59:08] <kittonian> another, ext3. sorry for the delay but
i'm freaking out and in a bunch of different conversations. the
best I can think of at this point is that I rm -r -f a file called
Data.fs.old and if I can somehow recover that, I will be ok
1263[18:59:57] <cluonbeam> kittonian: You've overwritten the
beginning of the file. The rest of it is just orphaned blocks, and
there's no longer any structure to find.
1264[18:59:59] <kittonian> just don't know how else to track
this down since I've seen a lot of people saying that
extundelete doesn't find all the files
1265[19:00:20] <cluonbeam> kittonian: If you need to recover this
data, you should seek professional services and stop futzing with it
before you fully destroy it.
1266[19:01:03] <kittonian> btw, i appreciate any assistance but i
do not need anyone telling me how stupid this was. i already know
and it's not a joke. this is the last 6 months of work and yes,
it should've been being backed up, and when I noticed it
hadn't today, i immediately went to back it up and typed the
tar command backwards, thus overwriting the file.
1268[19:01:43] <cluonbeam> kittonian: Not trying to badger you.
Just trying to save you from further damage.
1269[19:01:46] <kittonian> i cannot seek professional services,
though i'm happy to pay whatever it costs to get this back,
because this is a Linode and I spoke to them and they told me, maybe
i could find someone who could do this without accessing the disk
1270[19:02:01] <kittonian> i would be more than happy to pay
someone to get this back right now
1271[19:02:09] <kittonian> and i really don't care what it
costs
1272[19:02:10] <cluonbeam> kittonian: Did you have their backup
service enabled?
1273[19:02:14] <kittonian> i did not
1274[19:02:17] <sney> photorec is another utility used for
recovering deleted files, but if you overwrote the file that you
need, it's likely that it's just gone
1275[19:02:23] <kittonian> or else i wouldn't be having this
issue
1276[19:02:24] <cluonbeam> Then I don't believe that you
have any recourse.
1277[19:02:52] <kittonian> i literally deleted the file and
didn't do anything else besides run the bad tar command before
rebooting into rescue mode
1278[19:02:54] <another> did you copy it anywhere else? git repo?
1280[19:03:03] <kittonian> there's no way i could have
overwritten it
1281[19:03:11] <kittonian> it was almost immediate
1282[19:03:21] *** Quits: Tobbi (~Tobbi@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
1283[19:03:22] <kittonian> like within 10 minutes or so of me
deleting it
1284[19:03:34] <kittonian> another, believe me, if I had it
anywhere else I would be fine
1285[19:03:41] <cluonbeam> kittonian: But you did. You overwrote
the first block of the file, separating it from any of the rest of
the on-disk structure. It's not like just deleting it.
1286[19:03:43] <kittonian> this is not something I'm
forgetting is there
1287[19:04:04] <kittonian> cluonbeam, i did?
1288[19:04:45] <cluonbeam> Yes. You output data to an existing
file, and that involved a truncation. There is no longer any
association between that block of the file and any of what else is
on the disk.
1289[19:04:53] <kittonian> no
1290[19:05:00] <kittonian> i had a copy of that file that I
deleted
1291[19:05:05] <kittonian> that's what I'm trying to
recover
1292[19:05:23] <kittonian> the Data.fs file was overwritten. the
Data.fs.old file is what I rm -r -f
1293[19:05:32] <kittonian> i'm looking for Data.fs.old
1294[19:05:36] *** Quits: Deknos (~deknos@replaced-ip) (Remote host closed the connection)
1336[19:44:05] <cluonbeam> kittonian: Files that no longer exist
no longer have a name. You're searching for the content, not
the name.
1337[19:44:49] <kittonian> ok, so since I am looking for an FS
file, and I specified that in the file list in photorec, it
doesn't matter what the file is named
1363[20:04:41] <oxek> truly, contact a professional
1364[20:05:34] <sney> most professional file recovery is based on
having the physical disk, and in this case the disk is an image in a
linode-owned array in a datacenter somewhere
1377[20:16:15] <kittonian> but it didn't recover the file
1378[20:16:31] <moldy> ah, ok
1379[20:16:49] <kittonian> so now i'm trying ext3grep
1380[20:17:03] <moldy> in that case the next option i see is to
get someone with knowledge about the exact format used by zope and
see if they can come up with a custom solution maybe
1381[20:17:25] <kittonian> unfortunately, i know one of the
initial devs and he already tried helping and couldn't
1382[20:17:41] <kittonian> basically i have to recover this file
somehow or i've set multiple companies back 6 mo
1383[20:18:50] <moldy> the initial dev doesn't necessarily
know the specifics of the internal database format
1384[20:18:57] <moldy> i'd talk to a plone / zope expert
1386[20:20:12] <moldy> i know that zope uses some kind of custom
version of python's pickle format, but i don't know the
exact details
1387[20:20:54] <moldy> like, which pickle protocol version, and
what the exact customizations are. but some of the plone / zope core
devs can probably tell you, or maybe you can research it yourself
1388[20:21:30] <oxek> kittonian: are you absolutely sure you are
the only one who has had that single file in the last 6 months?
1389[20:21:55] <moldy> kittonian: oh and also, i'd contact
linode. maybe they do have some kind of unofficial backup they can
give you
1390[20:22:35] <kittonian> am i sure? pretty sure because
it's my server and i'm the only one with access
1391[20:22:52] <kittonian> already contacted linode. no help
there
1392[20:23:01] *** Quits: gelignite (~gelignite@replaced-ip) (Quit: Stay safe! Stay at home! Stop the chain reaction!)
1393[20:23:08] <oxek> contact them again
1394[20:23:40] <moldy> i probably don't have to tell you
that next time, you should setup proper incremental backups and
store them on a separate system ;)
1399[20:26:01] <kittonian> moldy, that is so unhelpful and if you
read above you will see i already addressed that
1400[20:26:07] <kittonian> this is not funny
1401[20:26:11] <kittonian> and is not time for lectures
1402[20:26:20] <kittonian> this is not a game. these are
people's lives
1403[20:27:14] <moldy> it wasn't meant to be unhelpful
1404[20:27:25] <et09> kittonian: as he said it sounds like your
only chance is doing a physical backup of the driver and a
byte-by-byte scan for what you're looking for
1405[20:27:29] <moldy> i just wanted to make sure you know how to
avoid this fromhappening again
1406[20:27:29] <et09> drive *
1407[20:27:42] <moldy> et09: yeah, but i wouldn't be
surprised if they also have some internal backups that they could
theoretically make available
1408[20:27:47] <et09> but as a rule, backups are an essential
failsafe and often you only learn that by a horrifying mishap
1409[20:27:48] <kittonian> et09, i cannot gain physical access to
the drive
1410[20:27:55] <kittonian> i can't keep repeating myself
here
1411[20:28:06] <kittonian> linodes are not separate physical
drives for each user
1412[20:28:15] <et09> the storage is on a physical drive
1413[20:28:19] <et09> virtualized or not
1414[20:28:28] <kittonian> and they are no help in getting this
resolved. i offered them as much as they wanted to charge me and
they still said it was outside of their scope to assist
1415[20:28:33] <moldy> thought it would probably take more than
contacting their usual first-level support to get them to do so
1421[20:29:23] <et09> the larger the file, and the longer since
it's been erased, the less likely it's recoverable
1422[20:29:39] <kittonian> et09, it was erased 10 min before I
rebooted into rescue mode to save it
1423[20:29:44] <kittonian> and that's where I'm at
1424[20:29:47] <kittonian> nothing else was done
1425[20:30:09] <et09> think of what the file is on disk.
it's a sequence of bytes of some given format. it's most
identifiable by the header of the file, but in theory any part of
the file should be identifiable, provided that the reads those tools
are doing are scanning the same locations on disc that were there
originally
1426[20:30:28] <kittonian> ok, so?
1427[20:30:33] <kittonian> how does that help me?
1428[20:30:47] <et09> so to recover any part of it you have to
back up the contents of the physical disk and scan them
1429[20:30:54] <et09> through any means
1430[20:30:55] <kittonian> the question was whether these
utilities has a file size cutoff
1431[20:30:59] <moldy> you could search for certain contents of
the file on the raw disk image
1432[20:31:15] <et09> i doubt it
1433[20:31:22] <moldy> but for that, you first have to know the
file format. the first thing i'd find out is whether this is
using the ascii or the binary pickle format
1434[20:31:22] <et09> but
1435[20:31:24] <kittonian> they are certainly finding much older
files that were deleted
1436[20:31:27] <kittonian> just not this one
1437[20:31:41] <et09> to recover the whole file, if it was a
large file, it's less likely it was stored contiguously
1438[20:31:48] <kittonian> in fact, photorec recovered thousands
of old files, to the point it ran out of disk space to continue
1439[20:31:53] <et09> and since the fs headers describing the
location of the file are gone, that's a problem
1440[20:32:01] <moldy> this is probably documented somewhere at
zope. or ask an expert.
1441[20:32:35] <kittonian> moldy, zope doesn't exist anymore
and i already said, a friend of mine is an original dev and he
couldn't help either after trying a bunch of things
1442[20:32:42] <kittonian> my only option is to recover this file
1443[20:32:44] <kittonian> that's it
1444[20:32:47] <kittonian> there is no other option
1445[20:33:06] <moldy> kittonian: the zope docs do exist, as do
zope experts
1446[20:33:13] <kittonian> moldy, stop
1447[20:33:16] <kittonian> seriously, stop
1448[20:33:28] <moldy> yes, i am trying to tell you *how* to
recover the file...
1449[20:33:32] <kittonian> no you are not
1450[20:33:36] <kittonian> because you are not listening
1451[20:33:38] <oxek> kittonian: post your problem somewhere on
reddit or hackernews or some other place like that, with a bounty.
1452[20:33:40] <moldy> ok. good luck.
1453[20:33:51] <kittonian> my friend is one of the original zope
developers. he literally created zope
1454[20:33:52] <et09> yeah upwork maybe
1455[20:34:02] <kittonian> he is the one who tried to assist
before anyone else and couldn't
1456[20:34:08] <oxek> kittonian: there's nothing we can do
beyond all the help we've already provided
1457[20:34:29] <et09> kittonian: if you can't recover the
whole file, you have to recover it piece by piece. this is your only
available game plan
1458[20:34:34] <kittonian> then linode support really screwed me
because they had me reboot into rescue mode, which means all of the
zope db info that I could access, is now gone because after a reboot
the memory is cleared
1459[20:34:38] <moldy> if he created zope he can probably
instantly answer which pickle protocol this is based upon, which is
likely crucial information for any further recovery attempts.
1461[20:35:08] <kittonian> i am trying every file recover option
i can get my hands on
1462[20:35:11] <oxek> If I lost some data, I'd contact a
data recovery expert and ask them if they have any experience with
linode. Some company will have experience, and will even tell you
they have a contract with them, etc.
1463[20:35:25] <kittonian> i would love other suggestions if
possible. already tried extundelete, ext3grep, photorec, and
testdisk
1464[20:35:54] <moldy> scalpel. but if those all failed, it will
likely also fail. but hey, worth a try.
1465[20:36:17] <oxek> kittonian: that's why we initially
told you not to reboot, and contact an expert.
1477[20:39:28] <alex11> just type apt show, type a few letters
and let bash autocomplete stuff for you
1478[20:39:39] <jim> ,i v4l2loopback-utils
1479[20:39:41] <judd> Package v4l2loopback-utils (graphics,
optional) in buster/amd64: Commandline utilities for the for the
v4l2-loopback module. Version: 0.12.1-1; Size: 19.5k; Installed:
37k; Homepage:
replaced-url
1480[20:40:02] <jim> for the for the :)
1481[20:41:32] <jim> I'll do some more judd queries in
private... thanks a lot
1485[20:44:39] <kittonian> moldy, what does that have to do with
getting the file back?
1486[20:44:57] <kittonian> it's zope 2.13.29
1487[20:45:15] <moldy> kittonian: as i told a couple of times
now, it is absolutely crucial to get the file back if the
"obvious" approach using photorec et al. doesn't work
1488[20:45:36] <kittonian> ?
1489[20:45:56] <moldy> kittonian: you can use this information to
patch photorec for example, or (probably easier), you know what
values to pass to ext3grep --search
1490[20:46:01] <et09> kittonian: here's my file,
ZOPE__{a:25}{b:30} with pickle v1, [a:25],[b:30] with pickle v2 (not
really but pretend)
1492[20:46:17] <et09> if your disk looks like
sdoghasdgoiashgoasidhgaosdighZOPE__{a:25}{b:30}dofighsogihsdfogisdhfgoisdfg
1493[20:46:22] <et09> how you gonna find that in there
1494[20:46:32] <kittonian> i have no idea
1495[20:46:39] <et09> that is the problem before you
1496[20:46:52] <et09> it's possible the ZOPE part got
deleted
1497[20:46:58] <et09> so all you can look for is the {a:25}{b:30}
1498[20:46:59] <moldy> iirc zodb format is append-only, so if you
can find the start of the file, your chances of restoring most of it
even if a part is lost might be pretty high
1499[20:47:24] <kittonian> ok, where do I even start with that
1500[20:47:39] <et09> well, if you make a file backup of the
disk, you scan the backup for what you're looking for
1501[20:47:47] <et09> it's slow but that's what's
in front of yo
1502[20:47:48] <et09> you
1503[20:47:58] <kittonian> tell me what to do and I'll do it
1504[20:48:07] <et09> i don't know what's in the files
1505[20:48:08] <moldy> kittonian: 1) check which pickle format it
is (replaced-url
1506[20:48:09] <et09> but you have to find it
1507[20:48:24] <et09> i'm just explaining the outline of the
process to you
1508[20:49:19] <kittonian> moldy, it's 2.13.19, older than
anything on that page
1509[20:49:24] <kittonian> 2.13.29
1510[20:49:29] *** Quits: Tom01 (~tom@replaced-ip) (Read error: No route to host)
1513[20:52:23] <moldy> kittonian: so, reading
replaced-url
1514[20:52:55] <moldy> experiment with it or ask someonw who
knows, or dig through the zope history
1515[20:53:40] <alexrelis[m]> Hi, does anyone know what package
contains `libjpeg.so.8`?
1516[20:54:12] <alexrelis[m]> I'm trying to get this
proprietary program working by turning it into an AppImage and
I'm manually copying dependencies over so that it'll run.
1517[20:54:28] <alexrelis[m]> (doing this because I have to for
one of my classes)
1518[20:55:22] <moldy> kittonian: i guess that's a python2
project? then it's apparently either protocol 0 or 1
1531[20:56:54] <judd> No package named 'libjpeg8' was
found in amd64.
1532[20:57:52] <sney> you can *probably* dig the .so out of an
old .deb from snapshot, but that might cause new problems. how about
installing libjpeg9 and symlinking libjpeg.so.9 to .so.8?
1534[21:00:22] <alexrelis[m]> sney: Just to clarify, I will not
being installing these dependencies to my system, I'm just
adding them to an AppImage. You think it would still cause problems
if I manually added the exact one?
1543[21:04:01] <kikuchi> having some issues with debian 10. All
of the sudden when i start my computer and boot, it doesnt start in
GUI mode, it boots to the console
1544[21:04:23] <kikuchi> not a huge problem since I can fix it by
just logging in there and running startx
1571[21:27:42] <somiaj> could the default target have changed to
multi-user.target instead of graphical.target? Also check the
contents of /etc/X11/default-display-manager
1594[21:57:19] <jim> would that show me stuff I haven't
installed?
1595[21:57:37] <petn-randall> jim: That will list any packages
available. If you only want to list packages on the system
(installed or recently uninstalled), you can use `dpkg -l obs*`.
1666[23:16:05] <March15> looks like it was my socks5 proxy
1667[23:16:09] <somiaj> mirko: daily builds? You might want to
check a more appropriate channel (this is a stable support channel).
Also what is your goal? You can install testing using the stable
isntaller.
1668[23:16:19] <mirko> since 'daily' is linking to the
latest, guides like
replaced-url
1671[23:17:42] <mirko> somiaj: i'm trying to point out that
builds are daily builds missing, resulting in official tutorials
guiding the user to do things which won't do at the moment
1673[23:18:49] <somiaj> mirko: and I'm saying this
isn't the place for that, if it is a bug, report it to the bts
against the debian-installer, you could also try #debian-installer
on irc.oftc.net and maybe letting someone there know would be
useful.
1674[23:19:28] *** Quits: tommaso (~tommaso@replaced-ip) (Remote host closed the connection)
1675[23:20:07] <somiaj> I also wouldn't call the wiki
offical since anyone can edit it, but I think the users in
#debian-installer maybe more likely to spread light on why the daily
builds are missing.
1676[23:20:54] <mirko> somiaj: thanks, dropped it there
1700[23:53:58] <tabakhase> hm, and installer without internet
also leaves you without any debian repos... -> trying to setup
mirror asks for proxy "empty to skip" - this empty to skip
seems ignored, i get an error cant connect - no way to bypass other
than "not adding repos"...
1701[23:54:21] <tabakhase> guess its not rly expected to run the
netinstal image without internet ;-) but hey :P