this is #debianan IRC-Channel at freenode
(freenode IRC service closed
2021-06-01)
0[00:00:11] <rgr> why would I need to set my PATH? because I
want to run things from scripts that assume things are picked up on
a path. The same reason everyone needs a path.
1[00:00:30] <petn-randall> rgr: Why not set PATH in the script
then?
8[00:04:10] <rgr> sure I could move my path to zshrc but thats
then shell specific. I could source .profile from .zsrhc. Calling
the script from key bindings. OK, maybe its not so dire as my
original "oh my god" reaction implied. If I shebang the
scripts and then full path them from the key bindings I guess they
pick up the path. I guess. I shall try.
16[00:11:41] <ws2k3> i have a weird issue with iptables.
traffic arrives at my box but it does not arrive at the
destination.what is a good approach to find out what happens to the
packets?
125[01:57:22] <terr__> does anyone know how the boot in linux
_actually_ takes place? like I'm using MBR right now. These
systems do not support EFI. and my disks are <= 2 tb. I will have
a look at the partition table but some of the information is printed
anyways. my question. when its going to Stage 2 and actually loading
an OS - how does stage 1 know where Stage 2 lives... Note - I have
not actually dumped the MBR yet. I jsut want to know if anyone knows
and if this is
126[01:57:22] <terr__> documented. Documentation is good because
then I know what to look for.
193[03:01:02] <terr__> I set a partition label and I forget how.
Its a BIG partition which is just going to be my backup pool. Thing
is I see Stretch is using this to create /media/terr/'backup
pool' I hate having to use quotes in the filename so I want to
change this to 'backup-pool'. I think fdisk should do this
but I do not see how.
194[03:01:19] <mason> terr__: GPT or MBR?
195[03:01:40] <mason> terr__: I don't remember if Debian
ships a new enough fdisk, but gdisk can set labels, as can parted.
199[03:04:05] <terr__> mason, I will read the docs. I probably
will not look at the grub source (now) but I may end up editing the
partition table with my own code. I know a little about grub but I
also know I do NOT have the proper config files set up. What
I'm doing is pretty easy. Lift (via ddrecover) a complete
bootable partition and set the proper addresses so my backup copy
(in of course 'backup pool') can be booted if necessary
200[03:04:11] <terr__> mason MBR
201[03:04:32] <mason> terr__: MBR doesn't use labels the
way GPT does
210[03:06:29] <terr__> mason, this machine is old enough it
can't use GPT. so far all my drives are < 2TB. So for my
laptops (I have 3) MBR is okay. I also have a pair of 64 bit I5
towers and they will do GPT (UEFI) and I'll figure that out
once I get this done.
215[03:07:11] *** phatcow6 is now known as phatcow
216[03:07:13] <mason> terr__: GPT is just what's on the
disk. You should be able to use it anywhere.
217[03:07:43] <mason> terr__: The difference is that you use a
gpt_bios partition on non-UEFI systems using GPT.
218[03:07:48] <terr__> mason, I am VERY pedantic. I want backups
where I can use ddrecover to a fresh drive - copy the partition and
set the boot sequence and that machins is RUNNING regardless of the
OS
219[03:08:37] <terr__> mason, this machine does not have the
right bios. And I don't need it for this machine anyways. THe
towers are a totally different matter
222[03:10:11] <terr__> mason, are you saying "use" as
in when I'm using UEFI partitions on a machine with a newer
BIOS - or- repurpose the existing GPT partition and turn it into a
boot partition?
223[03:10:26] <mason> There are GPT partitions, not UEFI
partitions.
226[03:10:58] <terr__> mason, presently I'm working on
windows 7 machines. THere are a few hidden parts and I can't
find any documentation and have no idea even if they are used.
228[03:11:46] <terr__> mason, there should be a dummy MBR which
spans the whole drive - then the GPT is mapped somehow inside.
somehow = I don't knwo yet
236[03:18:40] <terr__> back. I lost my iced tea. Last week I
lost my Raspberry pi and it took me 3 hours to find it. If I had a
wife I would have someone to blame.
238[03:20:06] <terr__> mason, I'll find out. BUt maybe it
doesn't matter. I can copy them sector by sector and as long as
it works. These machines are getting FROZEN and they will think they
are at the bottom of Lake Vostok in Antarctica
239[03:20:57] <terr__> Wives are good for a few things.
SOmetimes they are pretty good company when we take them for dinner.
243[03:26:57] <terr__> How is this possible? I have an external
HDD connected. I see it in /media/terr and its 3 partitions formated
out of 5. ls shows me the contents of partition 5. I tried to umount
it - says its not mounted. I tried umount /dev/sde5 - it says it is
not mounted. How can ls see it if its not mounted?
286[03:54:10] <Kinaree> when i boot buster, i see volume group
not found for my Luks login , I can login and get to the lightdm
screen but is this normal?
287[03:54:38] *** Quits: Organizm (~irssi@replaced-ip) (Quit: Lost terminal)
289[03:55:35] <WoC> Somehow, i have a feeling there is a easier
way to set vi as editor other than purging my system from all other
editors
290[03:55:43] <terr__> I'm running parted. How do I show a
partition name and how do I know if this is a PC98 disklabel? How do
I find out the label type?
291[03:56:12] <terr__> WoC, I just type vi... what's yer
issue?
336[04:08:27] *** phatcow2 is now known as phatcow
337[04:08:42] <TravisMosley9> on Debian how do you make a slow
running tar backup job instantly complete?
338[04:09:10] <terr__> TravisMosley9, that is a decent way -
I'm changing this and putting /home in the boot partition... if
I have an extra partition I can always mount it over top - and ues
either one.
341[04:10:01] <TravisMosley9> ok...but that isn't useful
342[04:10:28] <TravisMosley9> all backups are at like 10MB/s max
343[04:10:36] <TravisMosley9> and they won't go any faster
344[04:10:51] <terr__> windows "linux compatibility"
has to be a virtual machine. I don't think they can do it any
other way. but - they can run linux software if the have the proper
libraries.
345[04:11:19] <terr__> how many are you running and what kind of
interface?
373[04:19:18] <terr__> TravisMosley9, what kind of USB?
374[04:19:25] <TravisMosley9> it's an SD card
375[04:19:29] <oiaohm> terr__: windows/Microsoft tried with WSL1
to wrap Linux syscalls to windows and that did not work out well.
WSL2 is a modified Linux kernel running in a cut down version of
hyper-v.
396[04:45:02] <terr__> TravisMosley9, okay - that is what I
wanted to confirm. Your SD _should_ be running USB 3.0 and perhaps
its not. It might be running 2.0... I wonder how you can find out.
397[04:45:21] <TravisMosley9> um what kernel module it uses
403[04:47:17] <terr__> TravisMosley9, its likely the same driver
- the question is how the SD controller is interfaced... I do not
know how SD ties into USB... maybe its the same interface? somehow I
doubt it.
404[04:49:10] <TravisMosley9> lol it's a USB 2.0 card
reader from Realtek
451[05:31:02] *** Quits: fas3r (~fas3r@replaced-ip) (Remote host closed the connection)
452[05:31:03] <tomreyn> jim: i have to admit i'm mostly
into ubuntu so far, not yet sure which state debian is in in this
regard.
453[05:31:11] <nyov> yeah, for systemd stuff, ask tomreyn. Maybe
it even magically makes you some swap ;) Otherwise, swapspace should
be equal the size of your ram
454[05:31:45] <jim> as you can see I got double swap right now
457[05:32:39] <jim> how do I boot a suspended machine?
458[05:33:18] <nyov> linux kernel handles that. So long as it
knows the resume partition on the boot commandline
459[05:33:53] <nyov> though if you want to work with encrypted
suspend images it might be more trickier
460[05:34:26] <jim> ok, so I'd use the same boot choice as
I did to boot the thing up in the first place?
461[05:34:52] <nyov> pretty much
462[05:35:18] <jim> the bottom line right now... do I need to do
anything to get it to suspend to disk, and to get it to boot?
463[05:35:21] <tomreyn> if you use systemctl hibernate (invoking
which *may* also be possible from a GUI) then hopefully ram should
get dumped to disk and restored from disk after resume from
hibernation
515[05:47:56] <tomreyn> with hibernation, i think it's the
initrd that restores that has the kernel restore the memory dump.
but i'm not certain how it works exactly right now
516[05:48:05] <AngelKde> hello everyone someone having problems
with gdebi on debian 10?
526[05:51:17] <AngelKde> jim, yes i have problems with gdebi,
trying to install deb packages tells me that i don't have root
permissions
527[05:51:29] <tomreyn> jim: so suspend-to-disk "does not
work"? if so, where does it not work, how can you tell, is
RESUME= set and did you run update-initramfs and update-grub
afterwards?
528[05:53:48] <jim> I had problems with it months ago (or a
couple 3 years)... how can I test it?
529[05:54:13] <tomreyn> <tomreyn> if you use systemctl
hibernate (invoking which *may* also be possible from a GUI) then
hopefully ram should get dumped to disk and restored from disk after
resume from hibernation
580[06:31:21] <jim> ok, it suspended, and it resumed... but
either ardour, jack or the usb alsa device I had an alsa_in
invocation running, didn't like being suspended... it looked
like jack was started and it looked like ardour was running... but
I'm not sure what state they were in, and I'd have to look
into that whole thing to possibly get hib to work properly
789[09:14:08] <yibotg> Hi I asked the question about the fcitx
can't be triggered on for Chromium,Now I want to see what
happen about this,is there anyone can help me?I find it is not the
key binding problem for if I config the share status between
windows,the fcitx can't run also.and I launch the chromium from
xterm,the console output some error message but the chromium can
run,I think maybe this is the reason.
790[09:15:36] <yibotg> the error message is ERROR:bus.cc(393)
Failed to connect to the bus:Could not parse server address:Unknown
address type(exmaples of valid types are "tcp" and on UNIX
"unix"
791[09:15:52] *** Quits: r1nt3c (~r1nt3c@replaced-ip) (Remote host closed the connection)
796[09:18:41] <yibotg> and I think if I can see the log when I
press ctrl+space in chromium for what error happen leads the fcitx
can't not run may give me some information.How can I see the
log of Chromium,gtk,X and the kernel? thanks for attention.
895[10:00:26] <ratrace> well then, use Debian Stable kernel.
896[10:00:28] <rmrfchik> With kernel 4.19 all works fine
897[10:00:29] *** Quits: r1nt3c (~r1nt3c@replaced-ip) (Remote host closed the connection)
898[10:00:31] <ratrace> !sns
899[10:00:31] <dpkg> Shiny New Shit Syndrome is a serious
disorder, which usually breaks out into an epidemic every time
something new is released. If you have SNS, ask me about
<backports> and <ssb>; these are better options than
upgrading to <testing> because it is a <moving target>.
900[10:00:49] <rmrfchik> ratrace: I thought this regression
should be noted
901[10:00:57] <ratrace> not here tho. you can file a bug report.
904[10:01:31] <ratrace> or in #debian-next on OFTC, that's
about running (and) testing Debian.
905[10:01:54] <jm_> someone complained about poor interactivity
with 5.x kernel here, I also saw a post to LKML where that person
linked it to swap, if they disabled swap, everything was fine, with
just minimal swapping → boom
906[10:01:56] <ice9> there is "./cron" process taking
800% how can I know whats going on and why this is happening?
907[10:03:39] <jm_> ice9: you could use strace to see what
it's doing or gdb to collect backtraces if it shows something
sensible
908[10:04:25] <ice9> jm_, strace is showing lots of
"lock_gettime(CLOCK_REALTIME," messages happening
continuously
975[10:28:00] <ice9> ratrace, sorry, i will check first if they
the guys here knows that IP before i paste it
976[10:28:04] <ratrace> and uh yes... virus IS a latin word,
meaning poison.
977[10:28:07] <Nako> I am looking for a step by step
guidе/documentation on how to configure a highly available NFS
cluster, with DRBD9 and Pacemaker, on Debian 10?
978[10:28:08] <ratrace> oops, wrong chan
979[10:28:21] <ice9> ratrace, may i know what you gonna with the
ip?
982[10:28:54] <ratrace> ice9: well I don't carea bout it.
you can `whois` it, or use `dig -x <that ip>` to see if it
rings a bell or something. however the whole thing looks very
suspicious.
983[10:29:01] *** debhelper sets mode: +l 1549
984[10:29:14] <ratrace> a "cron" binary in home dir
with a remote http con. "Seems legit".
985[10:29:20] <jm_> :)
986[10:29:25] <ratrace> ice9: oh btw, is it a binary? use
"file" command on it to see what it is
1082[11:32:44] *** Quits: Eurisko (~thee-anar@replaced-ip) (Quit: Of all tyrannies, a tyranny sincerely exercised for the
good of its victim may be the most oppressive.)
1118[11:58:02] <molluskempire> Hmm, I've got iwlwifi
installed but my laptop is still having issues actually recognizing
the wifi card as an option...any ideas?
1149[12:20:53] <petn-randall> nickgaw: Sure, I'd check
/var/log/syslog or kern.log. But from your description it sounds
like a hardware issue or power surge if it powers off when you plug
it in.
1153[12:23:10] <nickgaw> It will be running fine then for the
last few days I have been coming back to it and it has been
completely powered off and I have checked all connections and
everything looks to be ok if I sometimes get core temperature above
threshold when the system is running doing nothing what could this
mean?
1178[12:32:52] <nickgaw> When I look at /var/log/kern.log I find
loads of the temperature above threshold messages Could it be the
battery is warn out even when I do mainly always use it plugd in and
if the power management hardware is broken can it be fixed?
1182[12:34:56] <nickgaw> acpi -t -f says thermal 0 ok 85 F
thermal 1 active 199 F what does this mean?
1183[12:35:36] <nickgaw> It also says battery 0 full %100 what
does this mean?
1184[12:37:15] <tomreyn> the former probably means that the
thermal sensor currently measures a temperature of 85F, and would
consider 199F to be critical. and the latter probably means that the
single battery you have installed is 100% full.
1188[12:38:22] <tomreyn> seeing tmeprature above threshold
messages logged and concluding that it is battery reklated seems
like jumping to conclusions to me.
1190[12:39:30] <nickgaw> How long are the log files stored before
they are deleted?
1191[12:39:47] <tomreyn> i suggest you post some of the messages
you get, and install lm-sensors, run sensors-detect and post the
output of "sensors", too
1192[12:40:35] <tomreyn> how long log files are stored before
they are deleted is defined in the logrotate configuration
1193[12:41:02] *** Quits: bvs_ (~beavis@replaced-ip) (Remote host closed the connection)
1194[12:41:04] <tomreyn> you will likel yhave multiple kern.log*
files in /var/log
1195[12:41:44] <tomreyn> if you're running a current debian
version you should also have a systemd journal, which may be easier
to work with, using the "journalctl" command.
1221[12:52:44] <LtL> nickgaw: as tomreyn asked, post the output
of the 'sensors -f' command on a sane pastebin.
1222[12:53:23] <LtL> if you prefer celsius, drop the -f
1223[12:54:50] <nickgaw> What package should I install for
pastebin to work and what command should I run after the package is
installed to send the output to pastebin from sensors -f?
1224[12:55:24] <LtL> nickgaw: nc or pastebinit
1225[12:56:40] <nickgaw> It is now installed.
1226[12:57:10] <LtL> nickgaw: 'pastebinit sensors -f'
or sensors -f | nc termbin.com 9999' no qoutes, show us the url
it gives you
1261[13:08:20] <nickgaw> I bought this laptop from a friend of
mine so can not send it back to Samsung to get it fixed so would
taking it to someone local to have them take it apart be the best
option then?
1289[13:12:53] <autopsy> Take the fan off if yoiu see a grey
layer off filth and hair and dust wipe that stuff off with a cloth
and a toothbrush. blow it out.
1290[13:13:11] <LtL> nickgaw: of course, yes
1291[13:13:22] <autopsy> Yes you can order like 12 DC brushless 3
pin fans from ebay.com for like $15
1292[13:13:24] <nickgaw> I will first clean the existing fans.
1293[13:13:52] <autopsy> I used to have my harddrive on top of
one, and one on top of it.
1347[13:48:44] <autopsy> I have this stereo its a JVC UX-G50 is
there anyone who can lend me some basic rtoubleshooting tips when it
comes to electronics because when I plug in the left channel
speakers, any time I start playback of a source the component just
shuts off as in it is in standby mode.
1389[14:09:27] <rafalcpp> on related topic, is there some tool to
uninstall (and kill running processes if any) packages/programs that
are known to have (yet unfixed in debian) vulns (CVEs or debian
security adv)
1410[14:12:34] <xgpt> is there a way for me to set a root
password that's blank? i.e. no password, just type root into
the username and forget about entering a password? (I need it
temporarily for a poorly written script)
1420[14:14:51] <joepublic> well, xgpt, as root you could passwd
-d
1421[14:14:53] <nickgaw> Setting a blank root password would not
be a bad thing can I even go a step further and just have console 1
automatically login as root?
1422[14:15:15] <joepublic> hi nickgaw
1423[14:15:19] <xgpt> nickgaw: I need the password to log in
without a password
1424[14:15:52] <xgpt> i mean i need root to log in without a
password, but I am curious how to get TTY1 to log in as root without
a password, it doesn't solve my problem as interesting as that
is :(
1429[14:16:58] <xgpt> no, I'd have to rewrite like a dozen
interlocking bash scripts
1430[14:17:14] <joepublic> if only there was some command that
did what you want
1431[14:17:28] <xgpt> they're designed to get a system set
up from a debian netinstall image where the instruction is
"don't enter a root password, just hit enter and ignore
it"
1432[14:17:33] <ratrace> xgpt: so there's no one script you
call?
1438[14:18:16] *** Quits: il (~il@replaced-ip) (Ping timeout: 265 seconds)
1439[14:18:26] <xgpt> I know about sudo, I've been a debian
user for about 11 years, but I've never thought to strip the
root user of a password and I've never attempted to have
passwordless logins on my machines.
1440[14:18:39] <ratrace> xgpt: well I'm not sure I
understand your use case. If you want <something> to call
<a script> as root, you can easily do so as unprivileged user
set up for that script for passwordless sudo
1441[14:18:44] <xgpt> shoot....more like 12 years now...of good
security practices!
1443[14:19:08] <ratrace> what I fail to understand is why you
have to log in as root at all
1444[14:19:32] <xgpt> yeah, I'll just try it with sudo with
my user and set up sudo to let me call sudo without a password
1445[14:19:42] <xgpt> ratrace: one second:
replaced-url
1446[14:20:04] <ratrace> you're not setting up "call
sudo without password", you're setting up sudo(er) user to
call a script with no password prompt, aka. "passwordless
sudo"
1452[14:21:31] <ratrace> xgpt: you can also set up _one_ script
(called via sudo passwordlessly) that then cascades to call whatever
other scripts and commands it needs, they'll all run as root
due to their caller being run as root (unless they explicitely drop
priv)
1453[14:21:35] *** Quits: nickgaw (~nick@replaced-ip) (Quit: ircII EPIC5-2.0.1 -- Are we there yet?)
1475[14:30:21] <torchinz> xgpt, as in you could not reproduce the
error?
1476[14:30:33] <xgpt> torchinz: try running "sudo apt-get
update && sudo apt-get upgrade --dry-run"real quick,
let me know if there's a pending upgrade for deluge
1477[14:30:43] *** Guest15365 is now known as zodd_
1478[14:30:45] *** Quits: root____1 (~root@replaced-ip) (Quit: Lost terminal)
1479[14:31:13] <xgpt> torchinz: I can't even install deluged
right now, idk wtf is up with it, very strange
1480[14:31:48] <torchinz> I was having the biggest pain in
installing deluge plugins, last night
1481[14:32:56] *** Quits: ajazdzewski (~quassel@replaced-ip) (Quit: No Ping reply in 180 seconds.)
1482[14:33:02] <torchinz> the dry run option gave me two options
to autoremove primus-libs socat. Removed them
1484[14:33:10] <xgpt> torchinz: there's soemthing funky with
my mailutils-common files? looks like a dependency mismatch with
some crypto libraries, can you give it a half a day and come back to
it? try update/upgrade again, might be a lag in the repos freaking
out
1488[14:33:39] <xgpt> oh, yeah those aren't dependencies of
deluged so they aren't likely to be part of the issue, just
extra packages from you installing/uninstalling software on your box
1507[14:42:07] <xgpt> torchinz: torchinz: hold on, just rebooted
and it fixed my dependency weirdness, which is weird in and of
itself, but whatever, installing now
1508[14:42:26] <torchinz> Thanks for the follow-up xgpt :)
1509[14:42:43] <xgpt> torchinz: i'm waiting on other stuff,
nbd :)
1510[14:44:20] <xgpt> torchinz: working fine for me, try apt-get
purge deluge* && apt-get install deluged and see if
that'll fix it, if you don't have configs you can't
bear to lose in the process
1511[14:44:24] <xgpt> wait
1512[14:44:50] *** Quits: scholl (~petrilla@replaced-ip) (Remote host closed the connection)
1513[14:44:54] <torchinz> This was a completely fresh install.
Nothing to lose. I have tried the purge but it has not worked. I can
try that again
1538[14:49:50] <xgpt> torchinz: check /var/log/apt files for
python updates that might have come through recently
1539[14:50:06] <xgpt> also give your machine a reboot because why
not
1540[14:50:08] <nevivurn> torchinz: also try purging your user
pip-installed stuff, maybe try running deluged as a fresh new user,
see if that works
1541[14:50:12] <nevivurn> otherwise it's a system-wide issue
1542[14:50:36] <xgpt> torchinz: what nevivurn said . much better
first-next-step for the deluged issue
1543[14:50:54] <nevivurn> Although the fact that your errors say
/usr/lib/* suggets it may be a system-wide issue, which may be
trickier D:
1544[14:50:57] <torchinz> nevivurn and xgpt I have rebooted the
machine twice as of now. I just installed pip to just see if I can
install deluge via that.
1551[14:52:23] <xgpt> torchinz: try a reboot first before you
muck around with too many more packages, you're dealing with
environmental variables at this point so it can't hurt
1552[14:52:26] *** Quits: ich (~ich@replaced-ip) (Ping timeout: 240 seconds)
1553[14:52:27] <torchinz> Actually, I think I messed up when I
was trying to make my python 3 version as the reboot
1561[14:53:49] <torchinz> I tried messing with the bashrc file
and alias python=python3 and stuff like that
1562[14:54:02] <nevivurn> Oh, deluged on buster depends on
python2, you're definitely doing something wrong if it's
calling python3. At this point I would normally abandon all hope and
reinstall...
1563[14:54:22] <torchinz> Reinstall what?
1564[14:54:35] <xgpt> your life :P jk jk jk, dump that alias
1565[14:54:44] <xgpt> just dump that alias, you can't do
that with python
1568[14:55:02] <nevivurn> your system (depends on what you did to
make it call py3). I may have an answer on py3 deluge plugins on
buster, one sec
1569[14:55:06] <xgpt> it's not backwards compat
1570[14:55:19] <torchinz> I removed that alias from the bashrc
file but now the default python IS 3.7.3. I think, that's what
is creating this shitshow
1577[14:56:31] <nevivurn> re trying ltconfig, I suggest you just
go for the older version...)
1578[14:57:00] <torchinz> Should I just purge both the python
versions and reinstall them after a reboot? Would that help?
1579[14:57:26] <xgpt> I highly suggest you undo whatever python
stuff you changed recently back to stock. (you are running ZFS/BTRFS
snapshots and can just "revert" right?) I kid....I kid
1580[14:57:31] <xgpt> sorry mate
1581[14:57:33] <nevivurn> If it was just the bashrc alias,
logging out completely and re-logging in will fix it.
1582[14:57:59] <nevivurn> (if you're using a display
manager, may need to reboot again)
1583[14:57:59] <torchinz> nevivurn, I have rebooted at least a
couple of times for that. Nothing is happening there.
1584[14:58:26] *** Quits: xcm (~xcm@replaced-ip) (Remote host closed the connection)
1586[14:58:50] <torchinz> xgpt, I would not have been worried if
it was just deluge that was not working. But when pip didn't
work, that was a little harsh for me
1587[14:58:58] <torchinz> nevivurn, sorry but a display manager?
1588[14:59:16] <nevivurn> In that case, if you don't
remember what you did to make `python` point to `python3`, I'm
not sure reinstalling will help, short of reinstalling the entire
system fresh.
1589[14:59:19] <torchinz> Are you talking about gnome or
cinnamon?
1591[14:59:45] <nevivurn> lightdm or gdm, but if you're
using gnome or cinnamon you're most probably using a display
manager.
1592[14:59:49] <torchinz> when you say "entire system"
you don't mean the entire debian buster, do you?
1593[14:59:56] <nevivurn> I do mean debian buster.
1594[14:59:58] <torchinz> I am on cinnamon
1595[15:00:04] <torchinz> WHAT?!
1596[15:00:21] <torchinz> Fuck, that would be insane and painful
:(
1597[15:00:55] <nevivurn> I mean, I am confused as to what you
did to make it so `python` is `python3`, so no idea what to do to
revert it / whether purging and reinstalling python (which will be
painful) will fix it.
1598[15:02:05] <nevivurn> If it really was just a shell alias,
check the output of `alias -p`to make sure it's gone. If
it's still there, delete whatever is setting it and reboot,
repeat until it goes away.
1606[15:03:58] <nevivurn> The stuff in /boot or /root probably
isn't the thing that is making python==python3, it will
disappear as soon as whatever you did to make it so goes away.
1607[15:04:19] <nevivurn> As a temporary and horrible fix, you
may be able to alias python to python2 explicitly, I guess..?
1608[15:04:25] <torchinz> let me try that
1609[15:04:41] <ayekat> stop panicking please
1610[15:04:52] <ayekat> a shell alias won't change what
python scripts will actually invoke
1611[15:05:08] <ayekat> (unless you launch python scripts
explicitly from your shell as `python somefile.py`)
1618[15:06:11] <ayekat> yeah, that advice there is pretty...
limited
1619[15:06:21] <torchinz> ayekat, what do you suggest? I am
checking my "python --version" and it is showing me 3.7.3.
I mean, python3 is there, isn't it?
1623[15:06:57] <ayekat> torchinz: sure, but if you invoke any
utility with a shebang of #!/usr/bin/python or #!/usr/bin/env
python, that shell alias has absolutely no effect
1624[15:06:58] <Fox> use update-alternatives to change that, but
first check python2 is still there
1625[15:07:04] <nevivurn> Oh my, yeah the shebang on deluged
points to /usr/bin/python, jeez
1629[15:07:43] <ayekat> torchinz: if you want to *really* make
`python` be `python3`, you'd have to change the symlink, but
I'm going to tell you right away that that's a VERY bad
idea
1630[15:07:59] <torchinz> ayekat, didn't get you. Fox how do
I use update-alternatives?
1631[15:08:08] <torchinz> ayekat, I just want my old system back
:D
1632[15:08:16] <torchinz> No more experiments with python :D
1633[15:08:25] <Fox> first, what versions of python do you have,
both 2 and 3, please
1634[15:08:27] <ayekat> torchinz: if the alias is all you've
set, just remove that alias
1635[15:08:37] <ayekat> torchinz: I don't expect anything to
be really broken
1651[15:10:29] <themill> Debian doesn't use alternatives for
python…
1652[15:10:41] <nevivurn> themill: exactly... what's going
on here??
1653[15:10:42] <xgpt> wait waht? it doesn't use
alternatives?
1654[15:10:51] <nevivurn> At least not on any of my systems
1655[15:10:52] <torchinz> sorry, got kicked and didn't get
to see what you guys wrote
1656[15:10:52] <xgpt> wait, do you have python2 installed
anymore/
1657[15:10:55] <_alx_> I need to install libicu but specifically
the version that has the shared object libicui18n.so.56.1, the one
in apt is libicu63 for the latest debian. Can someone help me with
getting the older one please?
1658[15:11:23] <themill> xgpt: of course not. alternatives are
for different and compatible implementations of exactly the same
thing
1662[15:11:49] <SnoopJeDi> _alx_, doesn't look like
there's a package for that version (56), so you'll have to
get it directly from ICU I think:
replaced-url
1667[15:12:34] <xgpt> themill: fair enough, I just figured it
might, the whole python2 vs python3 thing is weird AF to me and
I'm amazed it's been allowed to go unfixed for so long in
the FOSS world personally
1668[15:12:41] *** Quits: ob-sed (~obesd@replaced-ip) (Remote host closed the connection)
1669[15:12:42] <xgpt> unfixed being the deprecation of an old
language
1670[15:12:51] <_alx_> SnoopJeDi: oh hey man! didn't know
you where in this channel too. I never saw that site, very helpful
thank you!
1671[15:12:55] <xgpt> because it's basically forked into a
different thing
1673[15:13:11] <SnoopJeDi> _alx_, you probably want the 56.1 tag
on GitHub:
replaced-url
1674[15:13:14] <BCMM> torchinz: in general, `python` should
execute python2.
1675[15:13:21] <nevivurn> xgpt: it's one of the main
examples of what *not* to do in case of major breaking changes now,
so hopefully people will avoid these kinds of dumpster fires in the
future...
1678[15:13:42] <BCMM> torchinz: this is because nothing should
actual call the `python` command - scripts should have #!s that
refer to the actual version they need, for example
1679[15:13:59] <themill> xgpt: deprecation of python2 will not
make it sensible to change /usr/bin/python to point to something
else
1680[15:14:10] <BCMM> torchinz: and when somebody *does* ask for
just "python", it *almost* always means it's an old
script that just assumes python2
1681[15:14:27] <_alx_> SnoopJeDi: affirmative, i'm going to
build it from source :)
1682[15:14:28] <ayekat> or it is compatible with both versions
:-P
1683[15:14:32] <torchinz> I am learning this the hard way -_-
1684[15:14:49] <ayekat> but really, the distro package
maintainers will usually fix that sort of thing
1687[15:15:08] <ayekat> in *theory*, you could change the symlink
without much breakage - but it's not expected on debian, so no
guarantee
1688[15:15:11] <b1ack0p> what is the default network/wifi
manager?
1689[15:15:26] <torchinz> I tried to change the alias in the
bashrc file to alias python='/usr/bin/python3'
1690[15:15:27] <xgpt> nevivurn: I mean python is great, but just
been impossible for me to get into when it comes to learning a
language, seriously people tout it as the end-all-be-all of beginner
languages but the python crew is so freaking burnt on what they see
as obvious fixes for v2 vs v3 incompatibilities they know how to fix
super easily...they're super mean to people and tell them to
RTFM all the time. Idk, I had a little
1691[15:15:29] <ayekat> if you're on one of the few distros
where they have actually changed the symlinks, a bit more testing
has gone into that
1692[15:15:33] <xgpt> more fun with Ruby and frankly I've
learned how to do a ton of stuff with just basic bash scripting and
unix utils with a few pieces of choice unixy-philosophy software.
it's sad that scripting is more stable than an interpreted
language
1693[15:15:35] <b1ack0p> sometime ago i installed debian without
any desktop and configured wifi over command line
1694[15:15:38] * xgpt steps off python hate soapbox
1695[15:15:49] <b1ack0p> now i installed xfce via tasksel but
there is no network manager
1696[15:15:49] <BCMM> b1ack0p: "default" is a bit of a
vague concept in debian, but if you installed a desktop task, you
probably have networkmanager
1698[15:15:58] <nevivurn> Ohhh, did you do
replaced-url
1699[15:16:02] <nevivurn> That is horrifying.
1700[15:16:04] <b1ack0p> but i still connect to wfi
1701[15:16:17] <BCMM> b1ack0p: if you didn't install a
desktop task, you're probably just using ifupdown (the
traditional /etc/network/interfaces system)
1710[15:17:21] <xgpt> how'd you get the system to call
python3 when you call python? would a "which python"
command provide any insight? I'm frankly fascinated to see what
you did
1713[15:17:34] <nevivurn> torchinz: did you do this:
replaced-url
1714[15:17:41] <BCMM> b1ack0p: task-xfcfe-desktop recommends
network-manager-gnome. are you sure that isn't already
installed?
1715[15:17:43] <xgpt> torchinz: what did you do? what guides did
you follow that led you down this path to re-link python to python3?
1716[15:18:16] <b1ack0p> BCMM: if it was installed it suppose to
show in panel right?
1717[15:18:22] <torchinz> nevivurn, not that answer. I tried the
first one.
1718[15:18:26] <b1ack0p> in notification area
1719[15:18:37] <themill> nevivurn: eek. that needs to be
downvoted to oblivion.... :(
1720[15:18:40] <xgpt> torchinz: frankly I did stuff like that
back in the day when I was learning too, and I learned not to tinker
with stuff in /etc/ /bin/ or to run anything that requires sudo
without making a backup first. reinstalling might just be easier if
you don't know what you did
1721[15:18:41] <BCMM> b1ack0p: i'm afraid you *may* need to
actually use package manager to find out, instead of just looking at
teh tray
1722[15:18:42] <torchinz> xgpt, I used the first answer in
nevivurn's link
1723[15:18:56] <xgpt> torchinz: he wasn't telling you to run
the update-alternatives program on python
1724[15:19:00] <b1ack0p> how can i look?
1725[15:19:03] <themill> torchinz: no, you've done much more
than that.
1726[15:19:17] <torchinz> damn -_-
1727[15:19:21] <ayekat> torchinz: the first answer is simply to
set the shell alias (which is almost perfectly useless)
1728[15:19:32] <torchinz> ayekat, how do I do that?
1729[15:19:33] <ayekat> but your alternatives output above
indicates that you have done more than just that
1730[15:19:36] <xgpt> torchinz: can you PM me your .bashrc
.bash_aliases and any other .bash files in your home directory other
than your history? (if there's nothing sensitive in them)
I'd prefer it if you'd pastebin them
1731[15:19:39] <nevivurn> You need to tell us what you've
done to your system, since it's impossible to have arrived
where you are right now just from a bash alias (the first answer)
1732[15:19:41] <xgpt> actually can you pastebin them to the
channel
1733[15:19:57] <torchinz> xgpt, let me try
1734[15:19:58] <BCMM> b1ack0p: `dpkg -l network-manager-gnome`
(or check in whatever graphical frontend you use)
1735[15:20:07] <xgpt> he mentioned bash aliases, I think a
pastebin of his ~/.bash* files would help us debug it
1736[15:20:07] <b1ack0p> network-manager-gnome is already the
newest version (1.8.20-1.1).
1737[15:20:07] <b1ack0p> network-manager-gnome set to manually
installed.
1738[15:20:32] <nevivurn> The fact that python is managed by
update-alternatives is baffling, and wrong.
1739[15:20:54] <SnoopJeDi> The preferred way to request a
specific python interpreter is usually to point to the interpreter,
which might be in a venv if needs be.
1740[15:20:55] <b1ack0p> but why cant i see in the panel
available networks?
1750[15:24:03] <BCMM> b1ack0p: i was suggesting that you might
check that. `pgrep nm-applet`, or use whatever task manager xfce
comes with
1751[15:24:27] <b1ack0p> $ pgrep nm-applet
1752[15:24:28] <b1ack0p> 1088
1753[15:25:04] *** Quits: we6jbo (~we6jbo@replaced-ip) (Remote host closed the connection)
1754[15:25:07] <BCMM> b1ack0p: ok, so it *is* running. the
question becomes "why can't you see it in the notification
area", i guess
1755[15:25:12] <xgpt> torchinz: just copy the files to a pastebin
service, please make sure to title the pastes with the filenames :)
1756[15:25:15] <torchinz> unable to copy the bashrc file. CTRL-C
is not working in the command line. What's "M-6"?
Sorry for this insanely stupid question
1757[15:25:17] <b1ack0p> no idea
1758[15:25:20] <BCMM> b1ack0p: do you have anything in there?
1759[15:25:22] <torchinz> xgpt, sure
1760[15:25:32] <b1ack0p> there is an icon looks like ethernet
1761[15:25:32] <xgpt> ctrl-c is an interrupt signal, it tells the
running process to stop
1762[15:25:36] <b1ack0p> not wifi icon
1763[15:25:48] <xgpt> torchinz: can you use leafpad/gedit a gui
editor?
1764[15:25:49] <BCMM> b1ack0p: oh, right. well that's your
networkmanager frontend
1778[15:28:06] <BCMM> b1ack0p: ok, so networkmanager is working,
and your frontend for it is working in xfce too. but, for some
reason, networkmanager does not think it is supposed to handle your
wifi interface.
1788[15:28:42] <xgpt> torchinz: let's just try something
real quick before you go any further
1789[15:28:48] <torchinz> sure
1790[15:28:53] <BCMM> b1ack0p: well, have a look in
/etc/network/interfaces
1791[15:28:57] <b1ack0p> i read and asked here to setup wifi
connection via command line before DE installed
1792[15:29:10] <xgpt> run this command, we're going to move
your bash config files to an archive folder and then you're
going to reboot and run another command
1793[15:29:23] <xgpt> torchinz: is this IRC connection on the
computer you're debugging or another device?
1794[15:29:25] <BCMM> b1ack0p: if you used that file to set up
ifupdown to manage your wifi, networkmanager (on debian) will
respect that, and leave the wifi alone
1803[15:30:57] <BCMM> b1ack0p: it sounds like you've
successfully used the command line before. consider using it now.
1804[15:31:20] <BCMM> b1ack0p: are you saying the
`/etc/network/interfaces` file does not exist, or that some gui app
you tried to use to open it doesn't like its format?
1807[15:31:40] *** Quits: dvs (~hibbard@replaced-ip) (Remote host closed the connection)
1808[15:31:44] <terr__> I just lifted the MBR off a couple of my
disks.hexdump does not show text. Usually I would see two columns
and one will be hex and the other ascii.. is there a utility that
will do this?
1824[15:34:00] <xgpt> petn-randall: because he's horribly
borked his system and I've been debugging his system now for an
hour
1825[15:34:09] <xgpt> mrrobot: are you torchinz ?
1826[15:34:14] <BCMM> b1ack0p: ok. you need to delete the parts
of that file that refer to your wireless interface. you can't
manage wifi with ifupdown and networkmanager at the same time.
1827[15:34:20] <mrrobot> yes xgpt
1828[15:34:31] <petn-randall> xgpt: It would be enought to
restart the shell, no need to reboot.
1829[15:34:37] <xgpt> mrrobot: run "python --version"
and tell me what's up
1831[15:34:56] *** Quits: torchinz (~torchinz@replaced-ip) (Remote host closed the connection)
1832[15:35:03] <b1ack0p> delete parts under #my wifi device ?
1833[15:35:11] <mrrobot> xgpt, restarted
1834[15:35:17] <xgpt> petn-randall: you're right, but I have
no idea what else he's done and frankly a reboot is a real good
way to get things back to a baseline, it also took about two minutes
of his time vs. more than that if he did something else
1835[15:35:17] <b1ack0p> iface wlp2s2 inet dhcp
1836[15:35:21] <BCMM> b1ack0p: look, i don't know what
comments you left in that file or what they mean
1837[15:35:28] <b1ack0p> wpa-ssid
1838[15:35:32] <b1ack0p> wpa-psk
1839[15:36:02] <BCMM> b1ack0p: you can delete your PSK and
pastebin it, or consult `man interfaces` and work it out for
yourself.
1840[15:36:09] <xgpt> mrrobot: run python --version and tell me
what you gegt
1856[15:40:32] <terr__> here is what I find. On a new HDD where I
do NOT expect an MBR and I expect it to NOT be bootable... in fact
there is an MBR. I need to read the partition tables out of the
file. Can this be done? Reading form sector zero (0) is not an
issue. I want to parse the file copy of sector zero... and later
change it.
1863[15:41:58] <terr__> I did use fdisk to make a partition as
bootable and it is reasonable that Fdisk may have built an MBR to
point the boot loader into the said bootable partition. I am trying
to find out what is actually going on here.
1885[15:45:45] <ayekat> though wait... it appears `ln -sf` is not
actually atomic
1886[15:45:49] <jelly> hah, /me slow
1887[15:45:50] <ayekat> but certainly still more compact
1888[15:45:52] *** Quits: netmk (~alex@replaced-ip) (Remote host closed the connection)
1889[15:45:54] <xgpt> jelly: I wanted to have him learn more
about unlink and link
1890[15:46:03] <torchinz> xgpt, it opened up a manual kind of
thing
1891[15:46:12] *** CrystalMath is now known as KittenMath
1892[15:46:16] <torchinz> "Summary of Less Commands"
1893[15:46:17] * xgpt tells ayekat trying to turn it into a learning
opportunity for torchinz
1894[15:46:29] <xgpt> oh torchinz after you're done with
that run "man man"
1895[15:46:34] <xgpt> also "man info"
1896[15:46:38] <xgpt> also "info info"
1897[15:46:45] <ayekat> xgpt: if they really do a step by step
thing and pause between `unlink` and `ln`, they'll end up with
a system without /usr/bin/python
1898[15:46:48] <torchinz> sorry, I didn't get what we did
with man update
1909[15:48:55] <xgpt> running it between the quotes removes the
link, and then if successful moves immediately to the next command
which is to run another link command ayekat
1910[15:49:06] <torchinz> deluge is up and so is pip!
1912[15:49:30] <torchinz> cannot thank xgpt and ayekat enough and
all the rest who helped me!
1913[15:49:36] <xgpt> ayekat: so if he ran the entire quoted
command string it's basically the same, unless he's
somehow borked ln and unlink...which I suppose you very well could
do but is unlikely
1914[15:49:45] <ayekat> xgpt: I'm aware of what it does -
I'm more thinking of the scenario where they do it step by step
instead of glued together like that
1916[15:49:52] <xgpt> np torchinz go forth and don't beak
stuff like that
1917[15:50:09] <torchinz> xgpt, what in the god's name did I
do? and what was this solution of yours?
1918[15:50:27] <xgpt> ayekat: fair enough, he'd already been
running stuff in quotes verbatim so I wrote commands to do exactly
what I wanted based on the info I had
1923[15:51:59] <xgpt> torchinz: you used update-alternatives to
link and unlink programs, what that program/script does is link
commands as typed from one to the other. for instance, there's
an old program called "awk" that searches for text inside
of files, and it's not the program that's shipped with
debian anymore, so they've gone and linked "awk" to
"gawk" which is what's shipped. gawk remains
compatible with awk, so any
1924[15:52:05] <xgpt> script that calls awk should be good with
gawk running it instead
1925[15:52:07] <xgpt> that's what that's there for
1930[15:52:39] *** Quits: jerry (~jerry@replaced-ip) (Read error: Connection reset by peer)
1931[15:52:39] <xgpt> 09:50 -!- Shahnaz_
[~Shahnaz@2a00:11c0:1337:b0f4:8d68:5a0f:8d21:ed87] has quit [Remote
host closed the connection]
1932[15:52:42] <xgpt> 09:50 -!- dfstorm
[~dfstorm@unaffiliated/dfstorm] has quit [Quit: dfstorm]
1933[15:52:45] <xgpt> 09:51 -!- Shahnaz_
[~Shahnaz@2a00:11c0:1337:b0f4:8d68:5a0f:8d21:ed87] has joined
#debian
1934[15:52:48] <xgpt> 09:51 < xgpt> torchinz: you used
update-alternatives to link and unlink programs, what that
program/script does is link commands as typed from one to the other.
for instance, there's an old program called "awk"
that searches for text inside of files, and it's not the
program that's shipped with debian anymore, so they've
gone and linked "awk" to "gawk" which is
what's
1935[15:52:51] *** chinztor is now known as mrroboto
1936[15:52:53] <xgpt> shipped. gawk remains compatible with awk,
so any
1937[15:52:56] <xgpt> 09:51 < xgpt> script that calls awk
should be good with gawk running it instead
1938[15:52:58] <xgpt> 09:52 < xgpt> that's what
that's there for
1939[15:52:59] *** mrroboto is now known as mrroboto2458
1940[15:53:00] <xgpt> 09:52 -!- mrrobot_
[~mrrobot@123.201.65.173] has joined #debian
1941[15:53:03] <xgpt> 09:52 < xgpt> or for calling
"sensible-editor" to call your specific text editor of
choice
1942[15:53:03] *** xgpt was kicked by debhelper (flood)
1978[15:58:55] <torchinz> I will be out of here, so that others
can be helped :) Good day, gentlepeople! ;)
1979[15:59:17] <xgpt> torchinz: just keep reading dude, people
are rarely this happy to help but I figured you didn't know
where to start poking around. keep learning how all the pieces fit
together, the source code is all there and the documentation is kind
of spotty and out of date but it's there if you learn how to
read between the lines
1980[15:59:21] *** Quits: ihristov (55c48532@replaced-ip) (Remote host closed the connection)
1981[15:59:23] <xgpt> bye torchinz
1982[15:59:55] <torchinz> thanks for that sane advice, xgpt. Will
do! :)
1992[16:04:25] <xgpt> jelly: good to know. I'm actually more
versed w/ ubuntu, jumped off that ship years ago and have been a
debian loyalist since, except it's generally SBCs, so
armbian/raspbian for the most part.
2109[17:05:35] <entropon893> IDK if anyone here is familiar with
kali installs, but if there's non-free software
(drivers&&firmware) req'd for install is there a quick,
dirty, effective way to include them in the installation media?
2123[17:09:19] <ksk> on the other hand, you might want to check
out dockers ressource managment and use that, and not just nice the
whole docker thingy ( do you actually know that nicing "that
part" of docker will lead to its containers using less CPU? :P
)
2124[17:09:24] <ratrace> actually it'll start an empty
"drop-in" file, so you just need to add the Nice entry to
it (in its appropriate [Group])
2125[17:10:05] <jelly> the idea of drop-in files is great
2129[17:11:12] <terr__> I asked before. I didn't see any
response so I'll ask again. I lifted the MBR (system will not
support EFI) on two drives - 1) the original windows 7 drive (sda)
and a backup external hard drive (sde). What I see in the two MBRs
has me confused:
2130[17:11:22] <e01> ksk: well I have more than 30 running
containers, and some times they freeze the whole system when doing
something in background, and because it's aws instance I
don't want go and reboot machine after each freeze
2131[17:11:23] <terr__> here is what I find. On a new HDD where I
do NOT expect an MBR and I expect it to NOT be bootable... in fact
there is an MBR. I need to read the partition tables out of the
file. Can this be done? Reading form sector zero (0) is not an
issue. I want to parse the file copy of sector zero... and later
change it.
2132[17:11:23] *** Lord_of_Life_ is now known as Lord_of_Life
2134[17:11:50] <terr__> I did use fdisk to make a partition as
bootable and it is reasonable that Fdisk may have built an MBR to
point the boot loader into the said bootable partition. I am trying
to find out what is actually going on here.
2147[17:14:02] <e01> ksk: I am not sure, but may be it worth to
try, I tried to set cpu limit, but not get anything from it, also I
found the documentation of containers niceness a bit complex so I
set values with the power of magic :D
2148[17:15:01] <terr__> well it seems no one is taking on the MBR
issue for now. I'll ask later. Meanwhile I'm looking for
articles and I might find something
2173[17:23:48] <ratrace> e01: you can try even smaller shares
(default is 1024 for every process/cgroup on the machine) ; but
ideally, you would look into the process table and find out what
those containers are doing. are they cpu bound? is there IO?
2179[17:24:57] <flakinson> I have: Depends: mypackage (=1.2) but
2.0 is to be installed
2180[17:25:04] *** Quits: Krennic (~Krennic@replaced-ip) (Quit: Lost terminal)
2181[17:25:15] <flakinson> in my repo I have both, but I want to
install 1.2.!
2182[17:25:22] <jelly> !bat
2183[17:25:22] <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
2195[17:27:37] <ratrace> e01: well if you have more than one
mysql on that machine, things are gonna get ugly. databases are
really huge consumers of IO and ideally you run a single server that
then does the scheduling
2198[17:29:07] <ratrace> like, single master mysql server can
easily do scheduling for hundreds of concurrent clients, but two (or
more) masters have no coordination between themselves and they both
want the kernel to do its work ASAP. especially with small pages
like DBs use, things quickly escalate.
2213[17:38:42] <ctcx> I used groupdel to delete some groups. But
I didn't know that the deleted groups were the secondary groups
of some users. When id -Gn those users I noticed the deleted groups
were no longer there.
2214[17:38:44] <repo> hi there! i'm having trouble setting
up X on a fresh debian installation (bare minimum).
2215[17:38:56] <repo> here's my Xorg.0.log output:
replaced-url
2216[17:39:14] <ctcx> Is this standard behavior on Linux? Or did
I do something very wrong by not deleting users from groups first?
2217[17:39:45] *** Quits: earthundead (~earthunde@replaced-ip) (Remote host closed the connection)
2219[17:40:26] <greycat> ctcx: if fred is a member of
(supplementary) group "bedrockers", and you delete group
"bedrockers", what did you expect to happen?
2220[17:41:25] *** Quits: oaudry (~oaudry@replaced-ip) (Remote host closed the connection)
2223[17:42:01] *** Quits: oaudry (~oaudry@replaced-ip) (Remote host closed the connection)
2224[17:43:15] <ctcx> greycat: could I do some kind of harm to
the system by deleting existing supplementary groups? Or should I
just not care about it?
2225[17:43:19] *** Quits: citypw (~citypw@replaced-ip) (Remote host closed the connection)
2226[17:43:35] <greycat> It's not harmful. I'm just
confused about what you think the problem is.
2227[17:43:48] <greycat> Did you expect an error message or
something?
2228[17:43:57] *** Quits: mortderire (~mortderir@replaced-ip) (Remote host closed the connection)
2231[17:45:15] <ctcx> Yes, like what happens when deleting a
group which still owns some files/directories. When ls -lah those
files group column displays only the GID
2232[17:45:30] <ctcx> But, this also doesn't harm system,
right?
2237[17:46:15] <ctcx> I thought I was inquiring in potential
security issues or the like...
2238[17:46:20] <greycat> In the case of supplementary groups, the
implementation is simply that there is a line in the /etc/group file
that says "bedrockers:1020:fred" or similar. When you
delete that line, fred won't be in that group next time he logs
in. There's nothing else to it.
2239[17:46:27] <ctcx> But hopefully it's just my mind...
2241[17:46:50] <ctcx> I thought I was inquiring in potential
security issues or the like
2242[17:47:14] <greycat> No. You've deleted the group.
Exiting processes that have that group in their set of privileges
will continue to have it, but NEW logins will not acquire it.
2254[17:52:40] <Hack5190> G'day. I've installed the
latest Kali Linux (2020.1) on an Acer Chromebook. Everything works
except audio (dummy output). Since Kali is based on Debian I found a
thread (on the Debian forums) from a user with the same processor
& audio hardware. Unfortunately there was no resolution posted.
Hoping someone here can guide me through resolving this issue.
Thanks. Here is the aforementioned thread ->
replaced-url
2266[17:56:21] <greycat> the Kali people will most likely tell
you that Kali isn't *meant* to be used as a primary desktop
operating system, and sound, what the hell good is sound when
you're pen-testing...
2267[17:57:09] <Haohmaru> gotta play some dubstep to feel moar
epic
2278[17:58:37] <ioricloud> any here use vivaldi browser
2279[17:58:50] <Haohmaru> vi-wat browser o_O
2280[17:58:51] *** Quits: fnaticrisk (~fnaticris@replaced-ip) (Remote host closed the connection)
2281[17:58:59] <greycat> !anyone
2282[17:59:00] <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>.
2283[17:59:02] <jelly> !vivaldi
2284[17:59:10] <Haohmaru> <crickets.wav>
2285[17:59:17] <jelly> fork off old opera code iirc
2286[17:59:25] *** Parts: Hack5190 (Hack5190@replaced-ip) ("Hack5190 has gone...")
2287[17:59:27] <greycat> oh, ... that makes some sense
2288[17:59:29] <dob1> well opera, vivaldi makes sense
2384[18:15:51] <judd> [1002:15d8] is 'Picasso' from
'Advanced Micro Devices, Inc. [AMD/ATI]' with kernel
module 'snd-hda-intel' in stretch. See also
replaced-url
2388[18:16:14] <ioricloud> @jelly I'll send a pastebin
2389[18:16:15] <greycat> jelly is very fast
2390[18:16:24] <jelly> predictive irc.
2391[18:16:57] <ioricloud> ls -ld /opt/vivaldi/vivaldi ->
permission denied
2392[18:17:07] <jelly> yeah we saw that
2393[18:17:57] <jelly> repo: I do not know whether AMD GPUs need
firmware to work AT ALL these days
2394[18:18:52] <jelly> repo: but you can certainly try enabling
non-free sources and installing firmware-amd-graphics and rebooting
to see if it makes any difference
2395[18:18:58] <jelly> !non-free sources
2396[18:18:58] <dpkg> Edit /etc/apt/sources.list, ensure that the
two main Debian mirror lines end with "main contrib
non-free" rather than just "main", then
«apt-get update». But bear in mind that you'll be
installing <non-free> software. These may have onerous terms;
check the licenses. See also <sources.list>.
2489[18:57:57] <terr__> I asked before. I didn't see any
response so I'll ask again. I lifted the MBR (system will not
support EFI) on two drives - 1) the original windows 7 drive (sda)
and a backup external hard drive (sde). What I see in the two MBRs
has me confused:
2490[18:58:08] *** Quits: grobi (~rtng@replaced-ip) (Remote host closed the connection)
2491[18:58:36] <terr__> here is what I find. On a new HDD where I
do NOT expect an MBR and I expect it to NOT be bootable... in fact
there is an MBR. I need to read the partition tables out of the
file. Can this be done? Reading from sector zero (0) is not an
issue. I want to parse the file copy of sector zero... and later
change it.
2493[18:58:58] <terr__> I did use fdisk to make a partition as
bootable and it is reasonable that Fdisk may have built an MBR to
point the boot loader into the said bootable partition. I am trying
to find out what is actually going on here.
2495[18:59:10] <terr__> I did use fdisk to make a partition as
bootable and it is reasonable that Fdisk may have built an MBR to
point the boot loader into the said bootable partition. I am trying
to find out what is actually going on here.
2496[19:01:41] <valdyn> terr__: the bootable flag is just a flag,
its useless
2497[19:02:17] <jelly> terr__: what's your actual goal,
after you read the partition table?
2498[19:02:20] <valdyn> terr__: it only makes sense with a boot
loader that looks at that flag
2499[19:02:29] <terr__> valdyn, I kinda thought so... and
I've asked in ##bsd also.
2507[19:03:19] <jelly> terr__: did you at any point write to it
or kept the disk connected to a booted OS?
2508[19:03:26] <jelly> like windows?
2509[19:04:00] <terr__> My goal? get Grub installed (config file
issues I think) and be able to move partitions from disk to disk
with a bootable image so as to have backup and recover and fail safe
2510[19:04:44] <greycat> You said "I did use fdisk to make a
partition ..." so the disk *was* "formatted" in the
sense that you laid down a disk label (partition table).
2511[19:04:46] <valdyn> terr__: grub needs files on disk in some
cases, its to large for the MBR
2514[19:05:06] <terr__> jelly - I had it in a window 7 systems
and partitioned it there. BUT - I did not set any of those
partitions as bootable and I did not even copy in a file... At least
I don't recall doing it... I used Stretch for that.
2519[19:06:05] <jelly> terr__: what kind of filesystems are there
and do you plan to keep any windows bootable?
2520[19:06:20] <terr__> valdyn, ya I know that - has to be in a
bootable partition. But this SHOULD mean it lives in a file... and
it becomes the 2nd stage boot. How does that sector address get
populated adn its it relative to the parition start or absolute?
2526[19:07:53] <terr__> I do intend to have a bootable windows
partition on this drive. It is a backup drive and has an image of
sda1. (windows 7). But it also has a bottable windows partition and
a bootable Linux partition... and it likely will spend most of its
life in a bank saftey deposit box.
2527[19:08:21] <valdyn> terr__: partitions are not
'bootable' - that flag is useless
2528[19:08:25] <jelly> terr__: usually grub-pc keeps a tiny piece
of code in mbr, and a larger piece in the unparititioned space
between sector 0 and first partition.
2529[19:08:41] <jelly> grub or linux do not care about the
bootable flag.
2530[19:08:54] <terr__> jelly, that DOES make sense. and if so
then I _should_ be able to look at the parition addresses... they
have to be there - and now I run into I don't know - and that
is how does Stage 1 boot jump into stage 2?
2551[19:12:19] <terr__> its been a few days since I tried
grub-install. I started looing into how grub actually works... and I
do think grub-install will do the whole thing... once I find out
what its missing. The directories and files refered to in the docs
are NOT present in the buster.iso
2554[19:13:13] <jelly> terr__: grub-install needs to be from
inside the installed OS.
2555[19:13:17] <terr__> lets see. There is a utility in Stretch
that will record the screen... I do not recall at this point... I
saw it a few days ago.
2556[19:13:26] <valdyn> terr__: it can load files as soon as the
filesystem driver is loaded - and it can also put plenty stuff into
the unused area
2557[19:13:27] <jelly> terr__: it does not matter what's on
install media.
2563[19:14:13] <jelly> terr__: you mount your linux, prepare
(bind-mount) /dev /sys and /proc, chroot into your linux install,
and run it there
2564[19:14:38] <terr__> here is why. The machine is booted adn
ruinning Stretch. I am converting to Buster. I have a boot diectory
set up and I did chroot. Grub says missing something... I can do it
again.
2566[19:15:35] <greycat> converting to buster shouldn't
require you to chroot into anything, or install grub manually... it
should just continue using your existing grub stuff
2567[19:15:35] <jelly> terr__: this buster you're converting
to is a completely separate installation of Debian on a different
filesystem, right?
2576[19:16:46] <terr__> I might be missing some pieces... I do
have the .iso... and the issue its the default and to use it I will
always have to have a separate boot media - which is pretty easy. I
want this backup HDD to be totally self contained and to be able to
rescue if necessary and install if necessary - both windows and
Linux and in fact ANY partition - bootable or not!
2579[19:17:13] <dpkg> [stable] The status of a Debian release
when no packages will be added or version-bumped, and changes will
only fix security issues and critical bugs. Packages can be removed
in rare circumstances. The current stable version of Debian is
Buster (10.x); ask me about <releases>. Security bugs are
fixed in stable by backporting the fix to the stable version (ask me
about <security backports>).
replaced-url
2580[19:17:42] <terr__> jelly - yes. Stretch lives on a 32 GB USB
thumb. Buster will live on a 1 TB external HDD and it has a nice big
partition all for itself.
2581[19:17:54] <greycat> oh, you're trying to *install*
buster, not upgrade to buster
2582[19:17:55] <jelly> terr__: is a completely separate
installation of Debian on a different filesystem and a different
disk, and you intent to make the disk bootable even if it's
standalone and the only disk in the system?
2585[19:18:38] <terr__> Windows is on a 3rd HDD (internal) and I
have buster.iso on an SD plus I loop back mounted it and I have
copies pretty much everywhere - not that it matters.
2586[19:18:47] <jelly> terr__: if you have a second, already
running linux, that's enough
2588[19:19:14] <terr__> yep. That HDD has to be totally stand
alone... one that I can plug into pretty much any old machine.
2589[19:19:36] <jelly> terr__: you mount your new linux
installation somewhere, bind-mount /dev /sys and /proc, chroot into
your linux install, and prepare the boot loader there
2598[19:20:48] <dpkg> To reinstall <GRUB> boot to your
Debian install disk/live CD, switch to the other console (Alt-F2),
mount your root filesystem (mount -t ext4 /dev/whatever /target ;
mount --bind /dev /target/dev ; mount -t proc none /target/proc ;
mount -t sysfs none /target/sys), chroot into it (chroot /target),
run "mount /boot/efi" on EFI and "update-grub
&& grub-install /dev/whatever". See also <rescue
mode>, <dual boot guide>, <supergrub>.
2599[19:21:09] <jelly> terr__: start from step "mount your
root filesystem ..."
2600[19:21:39] <terr__> okay... let me give it a go.
2601[19:21:46] <jelly> ignore the efi bits if not using efi
2602[19:22:05] <terr__> jelly - also I am nursing a mild flu and
sometimes I need to take breaks... it comes and goes.
2603[19:22:28] *** Quits: Grldfrdom (uid391113@replaced-ip) (Quit: Connection closed for inactivity)
2604[19:22:49] <terr__> I won't need efi on this disk. It
has to run on 32 bit - but I will have 63 bit installed on it. For
Linux - I think its automatic.
2625[19:28:37] <terr__> jelly, it was a typo. But I do recall IBM
had an issue with extended addressing in some of their machines and
I think it had to do with somehow they didn't have a full
address.
2626[19:28:54] <terr__> maybe it was a 31 bit address - it was a
long time ago...
2631[19:30:27] <terr__> jelly, do you know if the boot addresses
are relative or absolute? They start out in sector (0) and have to
be absolute. But once a partition table is loaded they can be
relative.
2636[19:30:59] <greycat> if you want to understand the GORY
DETAILS of boot loaders, this may not be the best channel...
there's #debian-boot on OFTC. Maybe some grub channels here.
2638[19:31:28] <terr__> In 1983, IBM introduced 31-bit addressing
in the System/370-XA mainframe architecture as an upgrade to the
24-bit physical and virtual
2667[19:41:45] <PaddyF> i installed another distro in /dev/sdb3
bios boot and /dev/sdb4 root and the update-grub of debian (in
/dev/sda5) detects this but i get the grub error: no such device
2668[19:42:00] <PaddyF> "can not get C/H/S values; you need
to load the kernel first"
2669[19:42:24] <PaddyF> what can i do about it?
2670[19:42:44] <PaddyF> the partition is there and the UUID is
correct
2709[19:50:45] <dpkg> To provide command output in English
instead of your native language, set your locale to an English one
(e.g. C) prior to running the command, e.g. "LC_ALL=C apt-get
-f install".
2716[19:53:41] <tomg> I have Debian 10 on a VPS/cloud computer,
but letsencrypt doesn't work. It says "Certbot
doesn't know how to automatically configure the web server on
this system.". I have Apache installed
2787[20:27:06] <savantgarde> Is anyone here familiar with Docker
image vulnerability scanners that work well with Debian? I tried
trivy and it found lots of vulns (about 59) in Debian stable, and
none in testing, so I'm wondering if it's missing
backported fixes
2806[20:29:46] <savantgarde> So in general Debian stable should
have all the security fixes from testing?
2807[20:29:53] <jelly> no.
2808[20:30:03] <skandalist> Is there any manual about managing
LD_PRELOAD? (add, remove, reset,)
2809[20:30:07] <jelly> stable has MORE security fixes than
testing
2810[20:30:18] <jelly> testing is the worst option to use
2811[20:30:44] *** Quits: TESTED123 (~abc@replaced-ip) (Remote host closed the connection)
2812[20:30:58] <greycat> A few years ago, some people tried to
set up a "testing security" thing. It didn't catch
on. It still exists, in theory, but it's empty.
2813[20:31:15] <jelly> !testing security
2814[20:31:16] <dpkg> Security updates in testing are delayed by
the normal testing migration *and* may be further delayed by missing
dependencies, etc. See
replaced-url
2815[20:31:20] <jelly> it went nowhere slow
2816[20:31:37] <jelly> LTS works better
2817[20:32:21] <jelly> companies have actually been paying for
security fixes for old-old-stable enough to support about 1-2 people
working full time worth of
2819[20:33:08] <de-facto> Hello, I am using acmetool for lets
encrypt: just got an email that ACMEv1 will be disabled on
June/2020. Are there any "debian" package directories for
acmetool-0.2.1 (ACMEv2) as the current debian package will become
obsolete/usable with ACMEv1?
2832[20:36:42] <de-facto> just got an email from
noreply@letsencrypt.org "Update your client software to
continue using Let's Encrypt" stating they will deactivate
ACMEv1 protocol on June 1st 2020.
2833[20:36:49] <jelly> oh, added around 2018. Hopefully the
version in stretch or buster does it.
2834[20:37:20] <knives> jelly: I wonder what whip cream made of
camel milk taste like.
2835[20:37:20] <de-facto> not really, even sid will be obsolete
replaced-url
2836[20:37:36] <jelly> I mean the other LE client, dehydrated,
not yours
2837[20:37:44] <greycat> jelly: does what?
2838[20:37:55] *** Quits: section1 (~section1@replaced-ip) (Remote host closed the connection)
2873[20:52:38] *** Quits: mandeep (uid394387@replaced-ip) (Quit: Connection closed for inactivity)
2874[20:52:40] <teclo-> heh there's some package called
dehydrated
2875[20:52:43] <teclo-> didin't know
2876[20:52:57] <ratrace> dehydrated is awesome, I'm using
it.
2877[20:53:02] <de-facto> so how about acmetool? any ideas?
2878[20:53:15] <greycat> it's the one jelly and I are
using... but my web server is just *barely* running jessie, and even
then I have to shut down the wiki to run apt-get or my weekly
backups, because it runs out of memory
2904[20:59:24] <greycat> E: Build-Depends dependency for
dehydrated cannot be satisfied because candidate version of package
debhelper can't satisfy version requirements
2905[20:59:58] <greycat> JUST UPGRADE YOUR CLIENT TO THE LATEST
VERSION HOW BAD CAN IT BE RIGHT?!?!
2913[21:02:23] *** phatcow2 is now known as phatcow
2914[21:02:32] <greycat> what the fuck, there's only FIVE
FILES in the source dir, not counting debian/ and docs/ WHY THE HELL
IS IT SO HARD TO BACKPORT
2919[21:04:53] <thenori> Hey everybody. I have the package
libxcb-util0 but pkgconfig can't find the module xcb-util. What
do I need to do to get that package accessible by pkgconfig?
2944[21:09:45] <greycat> Someone convince me that it's worth
bashing my head against a wall repeatedly to try to build a backport
that inexplicably needs fancy buster bullshit debhelper options to
"build" a package that is ONE DAMNED SHELL SCRIPT.
2953[21:11:18] *** Quits: Adbray (~Adbray@replaced-ip) (Quit: Ah! By Brain!)
2954[21:11:53] <greycat> I think my best option at this point is
to mark the package as "hold" and then replace
/usr/bin/dehydrated with the dehydrated file from the source
directory.
2984[21:18:07] <jelly> things I stopped caring about: being able
to build the software on the same release it's going to be run
on, if the runtime deps are ok (and for a shell script, they usually
are), I grab .deb on stretch and copy it into my jessie repos.
2996[21:20:35] <jelly> I found a lenny in production last week.
2997[21:20:37] <ratrace> ehh why does dehydrated need apache2-dev
2998[21:20:45] <jelly> ratrace: for apache integration.
2999[21:20:46] <greycat> it has stupid build-deps because the
source package apparently produces multiple binary packages, one of
which is dehydrated-apache2 whatever THAT is
3000[21:20:52] *** Quits: CombatVet (~c4@replaced-ip) (Remote host closed the connection)
3018[21:23:28] <ratrace> no that's just alias dir. dehydated
has no functionality itself to move, install certs or trigger apache
reload. it has hook.sh that's basically emtpy for the admin to
set up
3019[21:23:33] <jelly> perhaps the maintainer decided to do that
last time things broke
3029[21:25:33] <ratrace> it is. also, it's a single shell
script. I pulled that from dehydrated's git when there was that
issue with dehydrated in buster few months ago (ACME something
something compatibility wrt http-01)
3030[21:25:36] <greycat> well... OK... it didn't blow up
yet.
3075[21:37:16] <annadane> i starting adding mailing lists to the
non-english factoids (because those irl channels never get any
people) but one thing i need to do eventually is also prune the
channels that don't exist
3133[22:06:22] <dpkg> <Maulkin>
*mumble*raid0-isn't-raid*mumble* * resiak removes two of
Maulkin's chair's legs. "It's okay, because they
were in a redundant array!"
3243[23:05:58] <greycat> Looks like bullseye has it. Will have
it.
3244[23:06:19] <knives> isn't bullseye a testing distro?
3245[23:06:25] <greycat> !bullseye
3246[23:06:25] <dpkg> The release following Debian 10
"Buster" is codenamed "Bullseye" (Woody's
horse in Toy Story 2) and will be Debian 11. It is the current
"testing" release. Remember that straight after a stable
release, all sorts of mess suddenly lands in "testing" and
it is best avoided if you don't like debugging things.
replaced-url
3282[23:12:32] <jelly> knives: which debian release do you have
installed?
3283[23:13:23] <rander2> mysql -u root -p ---> Enter password:
3284[23:13:24] <rander2> ERROR 1698 (28000): Access denied for
user 'root'@'localhost'
3285[23:13:40] <rander2> how may I solve this proble
3286[23:13:43] <rander2> m
3287[23:13:51] <jelly> rander2: which debian release is this?
3288[23:14:11] * jelly loves to ask that question
3289[23:14:13] <joepublic> pasting that error into your favorite
search engine should have it sorted within 5 minutes or so, if you
are unable to resolve it here
3290[23:14:14] <greycat> rander2: just run "mysql -u
root" as root. Do not use -p. Do not try to pass a password.
3298[23:15:00] <rander2> with id 0 ? root from OS not from mysql
3299[23:15:04] <jelly> !debian suite
3300[23:15:05] <dpkg> cat /etc/debian_version (or lsb_release
-sc). Or check /etc/apt/sources.list. If unsure about the
distribution, $ cat /etc/{*version*,*release*,*issue*} should grab
almost all distributions.
3341[23:32:11] <greycat> The important thing you need to know
about mysql is that it doesn't have "a password for
root". It has "a password for
'root'@'localhost'" and it has another,
SEPARATE, "password for 'root'@*" where * is any
other host.
3342[23:32:17] <greycat> This is all in the upstream docs.
3343[23:32:53] <greycat> Someone thought this was a good idea.