12[00:09:24] <MrHatter123> If the 'mv' command is
interrupted during the move, is the data still in its original
location until all the data is moved ?
13[00:10:51] <oxek> MrHatter123: no
14[00:11:00] <oxek> well, depends
15[00:11:12] <oxek> if it is single file, then yes, original
data is in original location
16[00:11:24] <oxek> if it is multiple files, then some files
are transferred, some are in original location
45[00:39:59] <ratrace> clearly the premise here is that tools
work as designed except, of course, when there's bugs. if we
don't make that assumption, then technology would not work at
all.
46[00:41:54] <ratrace> how often mv fails to atomically rename?
by itself, on its own, with no other symptom or failure? Near never.
Can it fail like that in case of a new bug? Sure. You'd
probably sooner win the EU jackpot tho.
47[00:42:25] <ratrace> in other words, this is kinda like
techno-reverse-solipsism. :)
48[00:42:33] <oxek> true
49[00:42:56] <oxek> maybe I'm just a cynical old debian
user, having seen mv fail in all sorts of ways
50[00:43:15] <ratrace> on its own or becasue the filystem had a
catastrophic failure elsewhere?
51[00:43:21] <oxek> both
52[00:43:41] <oxek> e.g. tried to mv a file on same filesystem,
and the file *disappeared*
53[00:43:47] <ratrace> I only saw the latter, in over 20 years
of *nix'ing. power failures, hardware glitches, memory
corruption, ...
55[00:46:51] <ratrace> I mean.... the ACID'ity of
databases is, partially, based on that atomicity (on atomicity of
filesystem operations), backed by a replay-able WAL. And if those
were failing willy nilly, teh whole civilization would collapse :)
60[00:47:38] <ratrace> can they fail? oh sure, absolutely. but
they don't in vast, vast, vast majority of cases, unless the fs
itself is experiencing a headshot. OR, it's mysql we're
talking about :)))
91[01:21:14] <dpkg> Welcome to #debian, the larger but
unofficial support channel for Debian GNU/Linux stable releases.
Official channels have moved to OFTC (irc.oftc.net) see
replaced-url
92[01:22:00] <Peppi> I'm using octopi (based on Debian).
I'd like to make it so when I do a ll (list long) it call ls -l
how do I do that? I don't even know what to google sorry lol
93[01:23:10] <ectospasm> alias ll='ls -al'
94[01:23:33] <ectospasm> Or, don't pass `-a` to ls if you
don't want to see . and ..
95[01:23:58] <ectospasm> Whenever I'm in a new
environment, I set up that alias.
96[01:24:19] <ectospasm> ...if I can't copy my full
environment to the system.
97[01:24:31] <riff-IRC> Or ls -Al if you want to see hidden
files, but not . and ..
162[01:38:22] <afidegnum> mutante: i'm choosing option 3,
Mail Sent by smarthost no local mail
163[01:38:25] <afidegnum> correct ?
164[01:39:19] <afidegnum> 1. internet site; mail is sent and
received directly using SMTP 3. mail sent by smarthost; no local
mail 5. no configuration at this time
165[01:39:19] <afidegnum> 2. mail sent by smarthost; received
via SMTP or fetchmail 4. local delivery only; not on a network
166[01:39:19] <afidegnum>
167[01:40:24] <afidegnum> what do you suggest?
168[01:40:25] <mutante> afidegnum: yea, well, do you have
external mail servers to use? yes, 3 is the common one
169[01:41:01] <afidegnum> mutante: yes, I have 2 exim servers
i'm confuguring, 1 as a clients with multiple mail accounts
from multiple domains
170[01:41:38] <afidegnum> and the smarthost which is in docker.
which will forward mails from server 1 to their respective
destinations
171[01:41:40] <mutante> It used to say ""(3) Satellite
system: All mail is sent to another machine, \called a
"smarthost" for delivery.""
172[01:41:47] <mutante> that's why I called it satellite
173[01:42:08] <afidegnum> oh, ok, then the name changed
174[01:43:19] <mutante> afidegnum: 3 would be if this host is
NOT doing mail itself and instead sends it to a different host, like
the mail servers of your org or Google or whatever feels responsible
for your domain
175[01:43:46] <afidegnum> yes,
176[01:43:54] <mutante> if you are setting up that smart host
itself then it's more like 1 i guess
183[01:46:40] <afidegnum> ok, let me clarify, I have server A
and B, server A has mails*@domains* it's a company mail system.
and sever B will forward mails sent from server A to their final
destinations
184[01:46:58] <sney> relay server relays all incoming mail to
another server. a smarthost (usually) actually handles mail delivery
for a local domain
185[01:47:12] <mutante> maybe the difference is only the list of
domains it accepts
186[01:47:31] <sney> and yeah, definitions in smtp can be a
little vague
187[01:47:31] <afidegnum> i have been battling this confusion
since morning
188[01:47:47] <sney> it is one of the oldest things on the
internet, after all.
189[01:48:02] <mutante> option 1 is if you are the mail server
yourself and option 3 is if the mail server is elsewhere
190[01:49:10] <afidegnum> sney: does smarthost send mails out
for a local domain?
191[01:49:16] <mutante> server A would pick 1 and server B would
pick 3
192[01:49:21] <afidegnum> mutante: ok, i'm working on
server B now
193[01:49:26] <afidegnum> where mails should be sent out
194[01:49:33] <afidegnum> so i choose 3 ?
195[01:49:43] <afidegnum> or 1 ?
196[01:49:53] <mutante> final destination .. means ?
197[01:49:56] <mutante> another mail server?
198[01:50:06] <mutante> or the user inboxes on that system
199[01:51:13] <mutante> well, now I think it might be 2 variants
of option 1 but with different list of domains it accepts mail for
200[01:51:29] <mutante> sorry, not sure, maybe sney can chime in
201[01:51:51] <afidegnum> yes, i.e Server A sends mails from
user1@domain1.com to some@google.com, user1@domain2.com to
@yahoo.com and server B pick those mails to send to the respective
destinations
202[01:51:59] <sney> usually the smarthost also does outgoing
mail, yes
203[01:53:24] <afidegnum> so i chose option 1 or 3 ?
204[01:53:31] <sney> for your docker container that is not
handling mailboxes or mail routing, you would definitely choose 3,
where server A is still the core of your company's email
infrastructure
218[01:58:46] <afidegnum> sney: ok, i chose option 3. do i use a
domain name or part of the fqdn here?
replaced-url
219[02:00:38] <Vanfanel> Hi! I have removed the systemd loging
service, so now I get "Failed to connect stdout to the journal
socket" errors on services launch. How I can stop systemd
trying to connect stdout to the journal socket? I don't need it
on this particular system
231[02:05:30] <ectospasm> It seems too much effort to set up
(r)syslog for systemd instead.
232[02:05:37] <ectospasm> But that's just my opinion.
233[02:06:12] <Vanfanel> ectospasm: I just don't want any
logging, and this is going to be an embedded system that works
perfectly fine for what it does
234[02:06:25] <Vanfanel> ectospasm: I just need to know how to
get rid of that message
235[02:07:16] <ectospasm> You could configure systemd-journald
to write to /dev/null
236[02:07:31] <ectospasm> That way, it doesn't actually
fill up your small embedded disk, but the socket is still there.
237[02:08:35] <Vanfanel> ectospasm: yes, but I would like to
remove the socket too. It was easy to do by masking it in the past,
now this error appears. Is that error unavoidable?
238[02:08:47] <mutante> Vanfanel: "LogTarget" in
/etc/systemd/system.conf sounds like it.
239[02:08:51] <mutante> Vanfanel: "One of console (log to
the attached tty), console-prefixed (log to the attached tty but
with prefixes encoding the log level and "facility", see
syslog(3), kmsg (log to the kernel circular log buffer), journal
(log to the journal), journal-or-kmsg (log to the journal if
available, and to kmsg otherwise), auto (determine the appropriate
log target automatically, the default), null (disable log
280[02:20:41] <Peppi> so where do I place the .bash_profile
file?
281[02:20:46] <ectospasm> It doesn't make that much of a
difference, but if you have things that print anything in .bashrc
they can mess with programs that don't expect that output (for
non-login shells)
282[02:20:50] <trench> depends on what you want to do?
378[03:42:49] <oxek> (unless we're close to release)
379[03:43:18] <oxek> we're still lacking a distro that
would have DFSG, but also up-to-date software
380[03:43:58] <mutante> but why would you want bleeding edge?
you like to be beta tester?
381[03:44:04] <oxek> or at least a service that would make keep
sid always up-to-date, and allow automatic (or at least trivially
easy) backports of all software
382[03:44:16] <oxek> mutante: sometimes there's a bugfix
that you need for work
383[03:44:23] <oxek> and compiling stuff yourself can be a
problem
384[03:44:31] <ectospasm> If you want bleeding edge or near
bleeding edge, go with a rolling release distribution.
385[03:44:49] <sponix2ipfw> Exactly
386[03:45:11] <abrotman> or learn how to make packages yourself
387[03:45:17] <mutante> ok, but generally it seems chances are
higher that you get bugs in work that still need to be fixed later
388[03:45:25] <velix> mutante: At least you can mount overlayfs
again, BUT I've read that you also can mount as non-root user.
389[03:45:54] <velix> sponix2ipfw: I've got virtual
machines to test new stuff and to make decisions about the way to go
next.
390[03:46:05] <mutante> velix: I can already mount as non-root
though, what am I missing?
391[03:46:24] <velix> mutante: I never had priviliges to this
without sudo?
392[03:46:39] <sponix2ipfw> mutante: apparently a higher version
number
393[03:46:45] <velix> sponix2ipfw: When next kernel supports
something I need, I either need another solution or need to change
distribution.
567[08:23:56] <asarch> What package do I need in order to run
i386 binaries on Buster for amd64 (I already did dpkg
--add-architecture i386 && apt update)?
575[08:33:51] <alkisg> asarch: I think it was like `apt-get
install libc6:i386 libncurses5:i386 libstdc++6:i386`, if it's
not I can look at the dependencies of some of my packages...
576[08:33:57] <oxek> asarch: the package should tell you
577[08:35:08] <afidegnum> hi, any insight about my issue?
579[08:35:48] <afidegnum> I have installed docker in debian and
running a debian based image, yet i'm faceing this error:
invoke-rc.d: could not determine current runlevel invoke-rc.d:
policy-rc.d denied execution of restart.
584[08:40:29] <asarch> ERROR: ld.so: object
'/usr/lib/x86_64-linux-gnu/pulseaudio/libpulsedsp.so' from
LD_PRELOAD cannot be preloaded (wrong ELF class: ELFCLASS64):
ignored.
585[08:40:58] <asarch> And this:
/home/asarch/Projects/Squeak/Seaside/Seaside
3.0.app/Contents/Linux-i686/lib/squeak/4.0.3-2202/squeakvm: error
while loading shared libraries: libuuid.so.1: cannot open shared
object file: No such file or directory
588[08:43:40] <alkisg> asarch: you'll need all the missing
libraries. Try `apt install squeak-vm:i386`, it should install some
of the missing dependencies for your project. Isn't it
available as a .deb package?
589[08:46:14] *** Quits: Z4CHe (uid496478@replaced-ip) (Quit: Connection closed for inactivity)
686[10:16:11] <alkisg> In extreme scenarios, I can imagine that
it could make a difference, but it's not common. E.g. package a
depends on b, newer package a no longer depends on b. If you purge b
before updating, you'd also purge a.
687[10:16:17] *** Quits: citypw (~citypw@replaced-ip) (Remote host closed the connection)
692[10:20:08] <put_in> alkisg: thank you for your answer. i mean
however just "update" not "upgrade" or
"full-upgrade"
693[10:20:16] <put_in> does it still apply then?
694[10:20:43] <alkisg> I gave an example where
"upgrade" would make a difference. "Upgrade"
requires "update", so yeah "update" would be
needed then. But that example isn't common.
695[10:20:48] <TheBigK> put_in: i would think purging doesnt do
much in that sense... since dependency is calculated on the
currently installed package
696[10:21:20] *** m1dnight__ is now known as m1dnight_
697[10:22:17] <put_in> the reason i am asking is because i write
a bash script for my server and i want to remove logrotate right
away. i will install some packages right afterwards but suddenly i
wondered if i should put the purge before the update or after.
698[10:22:30] <put_in> fair enough, i will purge after the
update. :-) thank you both
700[10:23:02] <TheBigK> put_in: its not wrong doing the update
first... so ur welcome :)
701[10:23:04] <jelly> you can always do an update
702[10:23:23] <jelly> it doesn't change anything in package
state
703[10:23:24] <alkisg> It's also possible for plain
"update" without "upgrade" to make a difference
in what's getting purged, but it's even more uncommon...
704[10:23:39] <ratrace> I think more joules and time were spent
thinking and asking about it, than running an update just in case :)
705[10:23:57] <jelly> alkisg, only on significant changes in
repos, eg. release upgrade
706[10:24:14] <alkisg> jelly: a depends on b|c; purging b might
trigger c getting installed
707[10:24:17] <ratrace> there's also in teh back of my mind
a situation where removing a package wanted to _install_ another
because there was an either-or dependency.... could be remembering
wrong and that wasn't debian
708[10:24:36] <jelly> alkisg, and existence of c won't
change on update.
709[10:24:46] <alkisg> Dependencies of c might though
710[10:24:53] <jelly> no, they won't
711[10:24:53] <alkisg> So different packages might get installed
712[10:24:57] <TheBigK> isnt that the case with
apache2-mpm-worker/prefork? or am i wrong there
713[10:25:04] <jelly> not within a release
714[10:25:27] <jelly> (well, okay, there are exceptions)
715[10:25:33] *** m1dnight_ is now known as m1dnight__
716[10:26:07] <jelly> alkisg, changes in binary deps and ABI are
a big no-no within a release
717[10:26:49] <alkisg> jelly: I think I've seen some, for
missing dependencies that were added as a bug fix... but I think
it's in the "too uncommon to search for" at that
point :)
718[10:26:50] <jelly> so if c follows Debian Policy, it's
not bloody likely to happen
720[10:27:23] <jelly> if it's a bug fix, then the bug will
be fixed, that sounds fine to me
721[10:27:52] <jelly> and if it's in stable, that'd
mean it was a pretty serious bug in the first place
722[10:28:47] <jelly> testing and sid are completely different
in this regard, and so are third-party repos which might not follow
debian policy
723[10:32:51] <jelly> put_in, you can do installation of new
stuff and purge in a single commmand with aptitude: aptitude -s
install vim-gtk logrotate_
724[10:33:31] <jelly> useful if purge would temporarily break
some deps
725[10:36:23] *** Quits: sputnik (kli0rf@replaced-ip) (Remote host closed the connection)
744[10:59:24] <ratrace> but, the $1M question is ... is tehre a
policy in packaging that forbids package names ending in _ ? I think
there is something about - as that has function... but _ ?
760[11:29:15] <jelly> _ is the delimiter for
package_version_arch.deb
761[11:31:06] <ratrace> jelly: ah yes, the arch. thanks.
762[11:32:26] <jelly> > Package names (both source and
binary, see Package) must consist only of lower case letters (a-z),
digits (0-9), plus (+) and minus (-) signs, and periods (.).
764[11:33:26] <jelly> ra+tra.ce is a valid package name,
rat_race would not be
765[11:36:28] <jelly> nothing about package names ending in - so
that's a problem, but there are already ambiguity issues with
those ending in + so it's not too huge a deal
766[11:37:15] <jelly> interestingly aptitude CLI conflict
resolution uses _prefixing_ with - and _ instead of suffixing
775[11:45:12] *** Quits: emOne (~emOne@replaced-ip) (Remote host closed the connection)
776[11:49:37] <jelly> does sudoers have a syntax to shortcut
several alternative parameters in a command line? %devops
ALL=(ALL:ALL) /bin/systemctl {start,stop,restart,status}
grafana-server.service
777[11:51:17] <ratrace> jelly: comma separated maybe, via
Cmnd_Alias spec
786[11:57:13] <ratrace> question is only if something liek this
is doable: Cmnd_Alias SYSTEMCTL_CMNDS = /bin/systemctl start,
/bin/systemctl stop, ... \n %devops ALL=(ALL:ALL) SYSTEMCTL_CMNDS
grafana-server.service
787[11:57:39] <ratrace> OR grafana-server.service _has_ to be
part of Cmnd_Alias, in which case this wouldn't work
792[12:00:09] <ratrace> jelly: MEANWHILE .... the penguin
demigods invented polkit. this would be super easy with polkit and
some hard dbus on dbus action, methinks.
816[12:22:18] <Heradon> Hello guys, i had a problem with my
debian VServer. i cant ping a host on the internet (mail-tester.com)
on my second root server i can ping (home too) any ideas?
831[12:40:04] <ratrace> Heradon: check if you can ping via ipv4
and ipv6 separately (-4 and -6 flags), check if you can ping by
hostname and IP address separately
832[12:40:42] <ratrace> and, describe "can't
ping", there's specific error message for specific trouble
(dns resolution, routing, ...)
833[12:40:46] <Heradon> i already checked, i think my ip is
blocked
983[15:24:12] <H-var> is there a good guide anywhere that covers
all those things, like "oldconfig" and stuff?
984[15:24:18] <ratrace> H-var: it's going through _new_
options and asking to set a value. missing this step would've
used defaults which may or may not've been sufficient
985[15:24:37] <ratrace> 5.12 is quite a jump from 4.19 btw, so
there'a ton of changes
986[15:25:04] <jelly> H-var, was it the kernel that failed, or
initrd?
987[15:25:10] <jelly> !debug initramfs
988[15:25:10] <dpkg> methinks initramfsdebug is
replaced-url
989[15:25:34] <H-var> jelly is there a logfile anywhere that I
can open?
990[15:25:44] <jelly> H-var, remove the "quiet" option
from kernel boot parameters, and try to detect where things go wrong
991[15:25:45] <ratrace> if the kernel didn't load,
there's no initramfs
996[15:26:30] <H-var> actually I am noob, but I think initramfs
is the first thing that loads??
997[15:26:40] <jelly> kernel panic happens if there's no
device with root fs found, too
998[15:26:48] <H-var> according to what I see during boot?
999[15:27:05] <ratrace> H-var: no. grub loads the kernel and
gives the initramfs _path_ as param, so the kernel unpacks the
initramfs in memory and seeks /bin/init and runs it
1000[15:27:38] <H-var> was anyone able to compile and load into
debian with 5.12?
1001[15:28:08] <H-var> on an AMD architecture system?
1002[15:28:09] <jelly> is this your first custom kernel?
1003[15:28:22] <H-var> first time always hurts
1004[15:28:24] <ratrace> jelly: yes yes but my point was, if the
kernel panicked, it didn't fully load or complete the loading
process so there's no initramfs to debug from
1005[15:29:12] <jelly> ratrace, the factoid shows how to stop
initramfs in several places to inspect things. You stop it before
the panic.
1006[15:29:14] <H-var> ratrace, I can record a video I guess, if
you want?
1007[15:29:16] <ratrace> H-var: "into debian" here is
irrelevant. there's no "debian" until the kernel
loads and starts /sbin/init from the rootfs
1008[15:29:49] <ratrace> H-var: no need, just please imgur a
screenshot of the panic message and trace, AND run without
"quiet" as jelly suggested
1009[15:29:54] <H-var> ratrace, you're right - maybe there
are issues with my hardware on 5.12
1010[15:30:11] <jelly> it's too early to say that
1011[15:30:59] <H-var> I will finish compiling 5.11, and try
installing that first, and then I'll boot into 5.12 and
screenshot the error
1012[15:31:31] <jelly> and based on my experience, if you started
with debian's everything-and-the-kitchen-sink .config,
it's more likely you have a working kernel and initrd failing,
than the kernel failing before initrd
1013[15:31:34] <H-var> have you tried 5.12 already? Is it working
for you?
1014[15:31:58] <jelly> debian thankfully allows us to NOT have to
build custom kernels
1016[15:32:41] <jelly> in most situations, that is
1017[15:32:43] <ratrace> unless you need to :)
1018[15:32:56] <jelly> you mean, unless you buy very latest
hardware
1019[15:33:21] <ratrace> no I don't mean that. I run
kernel.org current LTS, ahead of Stalebian's updates
1020[15:33:39] <jelly> I like what debian kernel team ships.
1021[15:34:02] <jelly> we roll with that
1022[15:34:14] <ratrace> because the kernel devs _want_ everyone
to run latest, and they designed the (stable) patching policies
around that requir^W desire.
1023[15:34:34] <H-var> I bought this fucking mini pci-e card and
it's not supported in 5.10. Debian plans 5.11 in 2022, so now I
have to compile this shit on my own like some hacker. I don't
even know what I'm doing.
1024[15:34:39] <jelly> I think the only place we have custom
kernel on debian is on some public exposed infra, where grsecurity
is applied
1025[15:35:28] <jelly> H-var, please file a bug report for the
kernel version in bullseye and ask for a backport of all the
bluetooth and wifi bits to bullseye kernel.
1026[15:35:29] <ratrace> H-var: since you've been at this
for the past several days..... could it be fastre/cheaper to just
buy another, more compatible card? :)
1027[15:35:37] <H-var> I have no idea what is going on
1028[15:36:12] <jelly> _you_ have to compile it now, but maybe
the next user with AX210 won't have to
1029[15:37:09] <H-var> so I just paste a screenshot, and explain
what is not working, and they might add it to 5.10?
1030[15:37:29] <H-var> but they already fixed it in kernel
1031[15:37:39] <H-var> only, the fix is available in 5.11
1032[15:37:59] <ratrace> jelly: how does that process work btw...
are such driver backports trickling down into stable point releases
or one has to wait for debian-next?
1033[15:38:20] <H-var> you mean report on the debian's
bugreport website, not the kernel.org website?
1034[15:38:33] <ratrace> yes
1035[15:38:50] <H-var> !bugreport
1036[15:38:53] <H-var> !bug
1037[15:38:54] <dpkg> Bug Tracking System for Debian packages,
replaced-url
1063[15:56:32] <jelly> so if you do a dpkg -S /boot/vmlinuz-* ...
it will report all the package names for kernels that are installed
and that ship files starting with vmlinuz- in /boot directory
1065[15:57:15] <jelly> is an example result of that, and
"linux-image-5.10.0-6-cloud-amd64" is the package name
1066[15:57:41] <H-var> oh, I see now. I will make a screenshot of
the error during the boot of 5.12, and then report it to the debian
bugtracking system mentioning the output of dpkg -S
/boot/vmlinuz*5.12 as the package
1072[15:59:44] <H-var> so I will need to report it to kernel.org
instead then
1073[15:59:56] <jelly> but fail to work. So you'd complain
to Debian to say "your bloody linux-image-5.10.0-6-cloud-amd64
doesn't work right with my AX210 PCI ID [8086:2725] Subsystem
[8086:4020] card,"
1074[16:01:03] <jelly> H-var, there are two different tracks
here. First is making a custom kernel and verifying it supports your
hw. Second is reporting that the distro kernel does not make your hw
work correctly.
1075[16:01:40] <jelly> H-var, you don't, in general complain
to upstream kernel that you didn't know how to build and run
your kernel
1134[17:10:54] <H-var> I guess the issue is that the driver is
written for 5.11 and is not adaptable to 5.12 or something. Idk, but
if anyone wants to buy AX210 - you can buy it. It works perfectly on
5.11 kernel.
1146[17:13:38] <vsk> Hi my debian 10 laptop freezes all of a
sudden when I watching Youtube videos. I have checked all the logs
in the log viewer but could not identify anything.....what other
facts I have to search please guide me
1147[17:14:20] <H-var> overheating CPU?
1148[17:14:35] <H-var> or damaged GPU\CPU
1149[17:14:36] <vsk> all logs mean : kern.log, sys.log,
message.log
1150[17:14:44] <H-var> maybe it was overheated once before
1165[17:18:17] <vsk> H-var as per my knowledge and understanding
which is limited ofcourse.......regarding the overheating thing I
could see message like this i am not sure if that is
cause............my point is I could not find any major descriptive
thing....which seem like error there could be minor things
1166[17:18:25] *** Quits: xet7 (~xet7@replaced-ip) (Remote host closed the connection)
1167[17:18:52] <H-var> actually what you can do is you can
download a stress test application
1168[17:19:03] <H-var> stress test your CPU GPU and RAM under
full load
1169[17:19:12] <H-var> without opening any other stuff
1170[17:19:28] <H-var> better loading into terminal instead of
desktop, and test from there
1171[17:20:45] <H-var> but even, if they pass the tests, there
could still be some other hardware issues - something wrong with the
motherboard
1172[17:20:58] <vsk> H-var this is what I saw in syslog few
seconds before it froze.........also i does not happen occasionally
once in 2 days esp when i watch youtube....first video stops playing
then I close browser tab....slowly browzer also stops responding
then whole pc
1173[17:21:09] <vsk> H-var Apr 29 19:33:03 Orion dhclient[650]:
DHCPREQUEST for 192.168.100.8 on wlp12s0 to 192.168.100.1 port 67
1174[17:21:09] <vsk> Apr 29 19:33:03 Orion dhclient[650]: DHCPACK
of 192.168.100.8 from 192.168.100.1
1175[17:21:58] <vsk> H-var my pc is connected to wifi router my
IP address is given dhcp so occassionaly the above thing is logged
1184[17:24:30] <vsk> H-var around the same time I saw cpu
throttle message "Apr 29 18:56:52 Orion kernel: [16536.254536]
CPU1: Core temperature/speed normal
1185[17:24:30] <vsk> Apr 29 19:02:01 Orion kernel: [16845.151198]
CPU1: Core temperature above threshold, cpu clock throttled (total
events = 14075)" like this messages are repeating for about a
min
1186[17:24:56] <oxek> /var/log/dpkg has nothing of interest
1188[17:25:38] <vsk> H-var this CPU temperature above threshold
messages came about 20m before the freeze event
1189[17:25:38] <H-var> vsk that's the problem - cpu should
be able to continue working even under 100% continuous load
1190[17:26:02] <H-var> even under 24/7 continuous 100% load CPU
should never fail
1191[17:26:49] <afidegnum> hi, please what can you suggest for
this exim4 config? I want to configure 2 servers, one relay mails to
the second server which send outside. what additional confugraitons
do i apply?
1192[17:27:00] <vsk> H-var right i am using 4GB RAM, 4GB swap and
core 2 duo processor 2.33GHz.........is it supported for modern
firefox-esr browser and debian 10
1208[17:30:57] <H-var> his CPU is overheating probably because
fan is dirty, and cpu throttling is not implemented well
1209[17:31:20] <vsk> jelly : last time I opened the CPU fan 2
years back
1210[17:32:26] <vsk> H-var ok i will try that ....regarding
"CPU throttling is not implemented well" anything I can do
in this regard
1211[17:32:37] <H-var> you can do a really simple life hack - buy
those plastic sexy stockings, cut them to size, fold into 2 layers,
and cover all the PC holes with them
1236[17:44:52] <andrew_smart> vsk consider seeing if in this
"frozen" condition you're able to switch to a virtual
console, e.g. ctrl+alt+f2, or using magic sysrq keys to see if
you're able to get the system responsive again
1237[17:45:54] <jelly> "hey your laptop looks sexy"
1255[17:59:59] <H-var> two pumps was totally unnecessary, but it
looks cool
1256[18:01:31] <H-var> the PSU is hidden behind the video card.
Air from the front fans cools down the radiator, and then goes right
into the PSU's fan, and then it comes out from the top of the
PSU and right into the top radiator, and fans, and then finally
escapes through the top
1257[18:03:32] *** Parts: Kel (~KelMonsta@replaced-ip) ("Closing Window")
1303[19:00:02] <another> H-var: got any references for that
claim?
1304[19:00:16] <sney> kernel.org is supported by sponsors,
because the linux foundation is a non-profit. big non-profits get
support from corporations. this is normal. companies like atmosera
and google and microsoft and so on have a lot of money, so they
provide things, that might include servers.
1306[19:00:33] <petn-randall> H-var: "This site is operated
by the Linux Kernel Organization, Inc., a 501(c)3 nonprofit
corporation, with support from the following sponsors."
1307[19:00:41] <petn-randall> None of those are Atmosera.
1308[19:00:51] <sney> "atmosera owns kernel.org" is
wrong; "atmosera provides some infrastructure to the linux
foundation" is accurate but your interpretation is stupid.
1309[19:01:20] <H-var> just found it interesting that kernel.org
is owned by Atmosera Inc, and this company is owned by
Microsoft's CEO, that's all...
1310[19:02:02] <sney> some parts of kernel.org live on a computer
provided by atmosera, you mean.
1311[19:02:05] <sney> oh noes.
1312[19:02:07] <petn-randall> H-var: You haven't provided
any sources. Until you do, I'll assume you made that up.
1316[19:05:46] <H-var> oh, I see. It's just when I tried to
open an email that arrived from kernel.org, my email client warned
me about "Dangerous content", and I was prompted to verify
the owner of that email. After a verification, it said that the
owner of the electronic letter was Atmosera
1325[19:10:15] <another> there's so much wrong with that, i
don't even know where to begin
1326[19:10:33] <H-var> I know, right...
1327[19:10:51] <H-var> Microsoft holds the servers of
kernel.org...
1328[19:11:00] <another> to clarify: i meant your statement
1329[19:11:35] <sney> "hey this is weird, why is this other
company verifying emails from kernel.org?" "because
corporate sponsors provide stuff to big non-profits whose tech they
use." "ohh that makes sense, thanks for the
explanation"
1332[19:14:18] <H-var> okay-okay - I know you guys know about it
since forever, but I just found that out now. Just wanted to share.
I didn't know everyone knows about it
1343[19:22:32] <ratrace> even if microsoft literally directly
owned the domain and servers .... so what. they're already a
significant contributor to the kernel code. they already own the
infrastructure for vast majority of FOSS ecosystem, namely teh
github.
1344[19:24:00] <alex11> (off topic but) i do think we ultimately
need some sort of decentralized code hosting website
1345[19:24:48] <jelly> alex11, every time you do a "git
clone" there's a decentralized copy.
1346[19:25:06] <another> if anyone cares: i dug a bit into it.
198.145.29.0/24 is registed to the the Linux Foundation and routed
and announced by 198.145.0.0/19, AS2044 Atmosera
1347[19:25:25] <alex11> yeah, but i mean stuff like github but
unable to be owned by a company
1348[19:25:36] <sussudio> without googling, is it just hosted on
azure?
1349[19:25:50] <sney> do it like kernel.org, where the companies
are welcome to sponsor but they don't control what the service
does
1350[19:26:17] <sney> atm with github, technically microsoft
could decide that some part of it isn't making enough money to
justify keeping online, and bye-bye that part
1351[19:26:31] <sney> but yes we're getting into the ot
weeds
1352[19:28:05] <aminvakil> alex11: i can't remember very
well, but there was an incident in 2012 or 2013 where kernel.org got
hacked somehow and nothing happened as everyone had a copy of source
code on their pc
1356[19:29:42] <ratrace> alex11: git != github. if github
literally disappeared today, all that code is in git trees of each
contributor for each project. issues and other things may've
perished, but that's all
1357[19:30:00] <alex11> ok, but it's still the basis for
project discovery
1358[19:30:46] <aminvakil> alex11: do you have any other
alternative?
1359[19:30:50] <ratrace> then everyone would re-origin to gitlab
and discovery would be restored
1360[19:31:14] <alex11> not currently, but i'm talking in
idealist terms
1361[19:31:37] *** Quits: szorfein (~daggoth@replaced-ip) (Remote host closed the connection)
1362[19:31:38] <another> there are people who use github for
discovery?
1367[19:34:55] <alex11> okay, but it's still the de facto
place for a lot of things
1368[19:35:19] <alex11> and if microsoft decided (or any company,
this isn't specifically anti microsoft) to Be Evil migrating
all these things to some other website would be a chore
1369[19:35:41] <another> sure. but i discover project usually by
a link somewhere. not by using gh
1397[20:01:35] <H-var> I honestly don't understand why
people called me "conspiracy" and lost their minds just
because I casually mentioned it. I don't accuse anyone. I was
just like "huh, that's funny" that's all. If I
knew someone would be upset, I would rather just not say anything at
all
1398[20:02:09] <imMute> I think most of the upsetness comes from
it being wildly off topic
1399[20:02:51] <imMute> and it was stated more as a fact than an
a curiosity. and it being very wrong played into that.
1401[20:05:06] <H-var> it's what my email client said
"would you like to verify the owner"? I clicked YES, and
it said "Owner: Atmosera Inc". So I thought
@kernel.org's owner is Atmosera Inc. I mean, I trust my email
client... In any case, it was programmed by a way more intelligent
person than me, so I didn't even have a thought of doubt
1406[20:09:52] <imMute> I think it's more you're not
understanding what "Owner" means in your email client.
whois says kernel.org "Registrant Organization: The Linux
Foundation"
1407[20:10:26] <greycat> owner *of what* was my initial thought
1443[20:26:50] <H-var> I feel like something's missing in
there?
1444[20:26:53] <greycat> that's not how log files work
1445[20:27:27] <oxek> H-var: what do you mean? That's the
entire output I saw on the screen
1446[20:27:41] <greycat> I don't *see* any other failures in
there, so maybe google the error message that you got and see if
there's a known bug in needrestart
1447[20:28:40] <oxek> I googled before coming here
1448[20:30:20] *** Quits: szorfein (~daggoth@replaced-ip) (Remote host closed the connection)
1466[20:41:14] <H-var> On August 8 it will be a full year since I
logged into windows last time. I wanna buy a cake lol, but I do not
know what to write on it. How about "sudo apt upgrade" xD
1508[21:02:26] <H-var> chriss_toni if you see firmware: failed to
load - it means that probably you do not have that file. You can get
that file from here:
replaced-url
1509[21:02:31] <jelly> BCM4314 sounds similar to those wifi card
from broadcom where you have to use a fwcutter tool to slice the
firmware out of their blob
1510[21:02:45] <sussudio> firmware-brcm80211 is in non-free, so
what i said before. don't know if this also contains BCM4313.
1511[21:03:01] <jelly> chriss_toni, you would put the file
exactly at /lib/firmware/brcm/BCM43142A0-0489-e062.hcd
1518[21:06:05] <dpkg> The Broadcom BCM43142 is a hybrid 802.11n
and Bluetooth 4.0 device. Its wireless LAN component (PCI ID
14e4:4365, incorrectly identified as BCM4365 by Broadcom's
proprietary driver) is currently only supported by Broadcom's
proprietary driver since version 6.30.223.126 (packaged for Debian 8
"Jessie" as broadcom-sta-dkms).
1527[21:11:13] <chriss_toni> I've found the driver on
Github. thx for the Link. Who can i load this driver with no reboot.
I have packed it on the right Path.
1528[21:11:19] <jelly> first result on goog's gives a github
repo with that binary file. Maybe it's kosher, maybe it's
malware for your radio chip, who knows!
1529[21:11:33] <chriss_toni> what?
1530[21:11:44] <jelly> chriss_toni, the easiest thing to do is
indeed a reboot
1531[21:11:54] <chriss_toni> oki.
1532[21:12:08] <chriss_toni> what is malware? jelly
1535[21:13:00] <chriss_toni> thx but i need this driver for my
Mocrosoft Mous.
1536[21:13:28] <chriss_toni> When i install this driver what is
waiting for me?
1537[21:13:30] <greycat> firmware, not driver
1538[21:14:13] <chriss_toni> yes. oki i was a windows user.. ^.^
1539[21:14:19] <chriss_toni> befor
1540[21:14:35] <sney> it's probably as safe as anything
you'd download for windows
1541[21:14:41] <sney> we just don't know, because the source
isn't visible
1542[21:14:49] <greycat> Drivers are loaded into the Linux kernel
and run on your computer. Firmware is loaded into your device, and
runs on that device.
1543[21:15:08] <jelly> it's as safe as any random github
repo -- we don't know what's in there
1544[21:15:43] <chriss_toni> okay.
1545[21:15:52] *** Quits: Night-Shade (~TimF@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
1546[21:16:38] <chriss_toni> the security on my laptop is okay
when i use this firmware. it's safe for me. best thx.
1547[21:17:06] <chriss_toni> But a nother firmware was a kosher
way
1548[21:17:09] <chriss_toni> ...
1549[21:17:17] <chriss_toni> öhm...
1550[21:17:24] <sney> with broadcom + bluetooth, this is probably
your only option.
1551[21:17:58] <chriss_toni> sorry you mean a stick per usb??
1673[22:49:01] <metbsd> hmm should i keep debian testing or
ubuntu? hard to pick
1674[22:49:07] <metbsd> they are pretty much same
1675[22:49:39] <mutante> this channel has bias for debian of
course
1676[22:50:18] <metbsd> yes ok
1677[22:50:40] <monkwitdafunk> as long as you dont waste precious
data you are okay. i am using 4G LTE
1678[22:51:00] <jhutchins> monkwitdafunk: There's not much
luck involved with burning an iso to a usb drive. Checksum your
download, use the correct command and you should have no problems.
If you have problems, post the command you used here and will set
you straight.
1680[22:51:53] <mutante> nowadays an actual "cp" of an
iso to a (flash) drive works, back in the days you had to dd
1681[22:52:04] <monkwitdafunk> thank you jhutchins. yeah. i am
rusty with my command line as i stopped using a computer longer than
30 minutes for maybe 3 years
1682[22:52:26] <monkwitdafunk> cp?
1683[22:52:28] <sney> a common problem with the DVD isos is
people trying to boot from the DVD-2 (or 3, 4, etc) images, which
don't contain a bootloader or installer progam, just packages
1684[22:52:32] <mutante> yea, like copy
1685[22:52:34] <sney> only disc 1 can boot and install
1686[22:52:51] <jhutchins> monkwitdafunk: Actually, I believe cp
always worked, it just wasn't much advertised.
1687[22:53:05] <metbsd> does testing have security fix?
1688[22:53:26] <monkwitdafunk> jhutchins, have you herd of
die.net?
1695[22:56:29] <monkwitdafunk> yes. i almost forgot. i am so
rusty but i am back
1696[22:56:45] <mutante> monkwitdafunk: I couldn't resist
the chance to say it: 'man man' (but to give you a better
answer, yea. die.net is actually somewhat well known for man pages)
1697[22:57:47] *** Quits: Deyaa (uid190709@replaced-ip) (Quit: Connection closed for inactivity)
1698[22:57:57] <mutante> more literal answer maybe something like
"man --all *"