159[02:22:28] *** Quits: hanasaki (~hanasaki@replaced-ip) (Remote host closed the connection)
160[02:23:03] <daegontaven> hey folks, can someone help me with
my wifi driver issue | it's not displaying any connections |
i'm using the 'wl' module | here is the output of
dmsg | grep wl |
replaced-url
161[02:24:00] <adso> run `lspci | grep network`
162[02:24:06] <adso> figure out what kind of network card you
have first
163[02:24:22] <adso> and my advice would be to google around for
it as typically there's a solid amount of documentation for
troubleshooting wifi drivers
164[02:24:39] <adso> i didn't have any problem finding a
plethora of information for my particular BCM wireless card online
165[02:25:23] <daegontaven> adso: i didn;t get any output from
running that
168[02:25:36] <adso> and if that doesn't work, try it
without the grep
169[02:26:12] <adso> you should see some output like this:
03:00.0 Network controller: Broadcom Limited BCM4360 802.11ac
Wireless Network Adapter (rev 03)
170[02:26:17] <daegontaven> adso: running this however gave me
this
replaced-url
171[02:26:20] <adso> in which case you can google around for the
model number
172[02:26:23] *** linuxsex is now known as Sain7
173[02:26:33] <daegontaven> it's only supported by wl as
far as i know
177[02:27:32] <awal1> daegontaven, just for be sure what you get
with lspci | grep -i wireless
178[02:27:59] <daegontaven> awal1: i get no output :(
179[02:28:02] <awal1> or grep -i net for wireless and ethernet
180[02:28:03] <adso> i ended up having to upgrade my linux
kernel to 4.14, upgrade the headers, uninstall previous drivers, and
install the broadcom-sta-dkms package
181[02:28:05] <adso> please be careful though
182[02:28:12] <adso> i imagine you could easily hose a box
messing with the kernel
183[02:28:22] <adso> this wsa the first thing i did with my
install so i was fairly careless, and it ended up working out
184[02:28:35] <adso> if you have anything important that you
wouldn't want to lose, backup obviously
185[02:28:49] <adso> also i'm running debian buster alpha
2.
186[02:28:49] <awal1> daegontaven, lspci | grep -i net ?
187[02:28:52] *** Quits: darkhanb (~textual@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
188[02:29:00] <daegontaven> adso: here is the short and full
output of lspci
replaced-url
189[02:29:06] <awal1> you should get an output with those 2
commands
190[02:29:07] <daegontaven> no need to do grep cause it's
so short
191[02:29:53] <daegontaven> output with "lspci | grep -i
net"
replaced-url
192[02:30:16] *** Quits: frostschutz (~frostschu@replaced-ip) (Remote host closed the connection)
195[02:31:52] <daegontaven> yes i already have wl installed
196[02:32:07] <daegontaven> and i reinstalled it twice
already..still not wotking
197[02:32:12] <daegontaven> *working :(
198[02:32:18] <memo1> Hi, im using iptable-persistace to save
iptables rules. Right now im trying to erase some rules using
iptables -D ... but it keeps failing iptables: Bad rule (does a
matching rule exist in that chain?).
253[02:47:30] <memo1> metastable: thank you so much. Can you
explain me where you find the answer. Google always shows erasing
the rules usig iptables -D ..
254[02:47:43] <metastable> memo1: Not kidding, 'man
iptables'.
255[02:48:46] <memo1> metastable: perfect. Last question, im
really new to iptables. The rules must be in Filter tables?
256[02:49:16] <metastable> memo1: No. Rules go in different
tables depending on their function. Most standard ACL entries go in
the filter table, because you're filtering traffic.
257[02:49:39] <memo1> metastable: in my case, im redirecting. Is
good in nat?
342[04:14:25] <somiaj> ozzloy: this can depend on the pdf form.
I think there is limited ability to edit pdf forms with various pdf
readers. There is an older adobe 32bit reader that can edit some
.pdfs that opensource alternatives cannot.
343[04:15:21] <somiaj> hmm, seems libreoffice can edit .pdf
files
373[04:42:17] <RayReddington> can someone help me? I dd'd a
usb with a debian distro
374[04:42:19] <RayReddington> and i cant boot from it
375[04:42:21] <RayReddington> im on mac mini 1.1
376[04:42:28] <RayReddington> the usb isnt recognized anymore on
the harddrive
377[04:44:47] <awal1> trying to use " volume_down_5 = -
" (default) instead of " volume_down_5 = , " via
'.moc/config' but won't work; I get "Wrong
option name: 'volume_down_5'. FATAL_ERROR: Error in config
file on line 3!". Why?
537[05:40:45] <ngharo> daniel-s: not that i know of but an
option that may work instead is to do an automated install by
preseeding answers to questions
545[05:42:34] <Nawab> Even after changing, its using tje old
servers
546[05:42:57] <somiaj> network-manager may overwrite things. It
should have the ability to set a custom nameserver, or you can use
something like resolvconf to better control the resolv.conf file
547[05:43:25] <Nawab> I dont hav tht installed
548[05:43:56] <somiaj> I really don't know your setup so
unsure what is best, but that should at least give you the abilty to
control /etc/resolv.conf without something changing it to manually
set things
556[05:46:37] <Nawab> But i dont hav tht installed
557[05:46:44] <somiaj> then install it
558[05:47:21] <somiaj> it isn't installed by default, but
it maybe a simple way to just control the resolv.conf file. Though
you should also be able to correctly configure network-manager to
setup nameservers, but I'm unsure on that front.
559[05:47:24] <Nawab> Will resolvcpnf conflict with network
manager?
560[05:47:28] <somiaj> no
561[05:47:47] <somiaj> all it does it create links and will try
to keep things from autmatically changing resolv.conf for you
610[06:29:42] <gubernator> Most IRC clients list them in the
righthand pane
611[06:29:53] <RayReddington> NAMES #debian
612[06:29:54] <RayReddington> fuck
613[06:29:56] <RayReddington> i haveno panel
614[06:29:59] <RayReddington> and i must list
615[06:30:14] <RayReddington> whos the guy
616[06:30:16] <RayReddington> with the name
617[06:30:18] <RayReddington> rpfan or something
618[06:31:52] <RayReddington> can someone with a sidebar
619[06:31:53] <RayReddington> tell me who he is
620[06:31:56] <maxcell_> After installing intel-microcode on my
debian (Buster) i'm getting and permission error with
"uinpit" when i try to connect my ps4 controller using
"ds4drv" : [error][controller 1] Failed to create input
device: "/dev/uinput" cannot be opened for writing > it
is safe to put write permissions to "uinput" for my
regular user? If it is, how do i do that?
621[06:32:49] <RayReddington> omg
622[06:32:51] <RayReddington> someone with a sidebar
623[06:32:53] <RayReddington> anyone
624[06:33:17] *** Quits: pav (~pav@replaced-ip) (Quit: t's now time to walk the dog.)
625[06:35:27] <bites> gubernator: should be there if you have
lightdm installed. what does "apt-cache policy lightdm"
say? is lightdm running? "systemctl status lightdm"
626[06:35:51] *** Joins: solaris (~solaris@replaced-ip)
629[06:38:35] <rant> maxcell_: I think more would've had to
happen than just installing intel-mirocode to change behavior of an
input device.. but yes you could change permissions of that but
you'd have to do it through a udev rules file to make it
persistant
637[06:43:19] <rant> safe is a relative term and fairly
irrelevant if you can't use it without the permissions, most
you can do is setup a group or specific user that has such
permissions
638[06:44:56] * rant bookmarks that reacivated.net guide on udev rules
cause it seems fairly complete
639[06:45:37] <rant> I have yet to memorize how to write udev
rules cause its so rarely I do it
640[06:45:48] *** Quits: daniel-s (~daniel-s@replaced-ip) (Remote host closed the connection)
696[07:28:21] <gubernator> I'm just coming to realize that
the DE determines most of the gui apps
697[07:28:41] <gubernator> rather than debian choosing them
698[07:28:52] <alkisg> A de is a predefined set of gui apps,
yes. Debian offers most des.
699[07:29:25] <alkisg> Many DE apps usually come from the same
team, so they cooperate better, it's not good to have some of
one de and some from another
700[07:30:03] <alkisg> E.g. you're using gnome, it
wouldn't make much sense to use mate's caja instead of
gnome's nautilus as the file manager
701[07:30:35] <klys> so is "desktop environment" an
oxymoron that deserves to be discredited?
702[07:31:11] <alkisg> I don't understand why that would be
an oxymoron, desktop environment and gui apps are closely related
703[07:32:13] <gubernator> I am trying to work with the defaults
apps as much as possible. The other day you were advising me on
symlinks and the user-dirs.dirs config, if you recall. I briefly
thought to use replace Nautilus with Nemo, which I was more familiar
with, but then thought better.
704[07:32:16] <alkisg> A desktop environment is a set of
essential gui apps, along with whatever backends they need to work.
The "additional gui apps" may be unrelated to the DE.
711[07:36:50] <klys> as in, desktop environments are like trusty
fundamentals. you can ask them for a place to rest; just don't
try to wake up
712[07:37:46] <gubernator> alkisg, I'm using a separate
partition for data now, with symlinks connecting them to /home. I
can't use trash now, only permanently delete. Do you know why,
and a workaround?
713[07:38:14] <alkisg> gubernator: in general, I'd advice
against using ntfs for daily use, e.g. Documents etc
715[07:38:43] <alkisg> That said, some file managers create
.Trash directories on each partition to store the trash, and your
user account may not have enough permissions for that
717[07:39:56] <gubernator> I use windows a lot for work, and
need to access the same data. It's my lot in life. Do you know
which sort of permission I might need to add to my user account?
785[08:05:50] <hemebond> If I'm currently doing "echo
1 > /sys/module/bluetooth/parameters/disable_ertm" where
would I make that permanent? Would I create something in
/etc/modprobe.d/?
887[09:27:32] <rant> klys: fyi while that solutions would
certainly work, you can achieve this with a udev rules file using
SUBSYSTEM=="module", DRIVER=="bluetooth",
ATTR{parameters/disable_ertm}="1" or by installing
sysfsutils package which sets up exactly what you were mentioning
and reads /etc/sysfs.conf where you can add
module/bluetooth/parameters/disable_ertm = 1
899[09:35:34] <rant> I added this udev solution to a new factoid
called from the sysfs factoid that already explains the sysfsutils
method for future reference
900[09:35:42] <rant> !sysfs
901[09:35:42] <dpkg> sysfs is a virtual filesystem that exports
kernel data structures and attributes. sysfs attributes can be set
at system boot via /etc/sysfs.conf (sysfsutils package); or via udev
ask me about <sysfs udev>.
replaced-url
922[09:41:13] <hoijui> i have no idea how to track it down
though
923[09:41:17] <rant> I could be wrong.. but it makes more sense
to me
924[09:41:56] <hoijui> i also think you are right.. was my first
guess too.. though i only saw it with ssh so far, and i dont
understand why, or how to track it
925[09:42:40] <rant> well you can go to the server (login via
ssh even) and stop the ssh service and start it manually on the
commandline in debug mode sshd -d and use scp -v to get some more
output
926[09:42:54] *** Quits: cdown (~cdown@replaced-ip) (Remote host closed the connection)
927[09:44:17] <rant> stopping ssh service doesnt affect running
connections but if you have no direct access to the server, you want
to make sure you start that back up and perhaps even fire up a
screen sessiong with a "sleep 300; service ssh start"
kinda command to make sure
928[09:44:50] <hoijui> its not a real server
929[09:45:09] <hoijui> just an other desktop machine, so no
problem there.. will try, thanks!
941[09:51:10] <rant> hoijui: you really should check the bts and
probably also the changelog on the package
942[09:51:40] <rant> hoijui: and fwiw I'd also highly
recommend apt-listbugs and apt-listchanges so you see this stuff
everytime you install/upgrade when using testing or unstable
952[09:54:33] *** Quits: pav (~pav@replaced-ip) (Quit: t's now time to walk the dog.)
953[09:54:54] <rant> hoijui: you should probably get to know
more about debian before you start fooling with our testing/unstable
branches. for one thing we dont normally support testing/unstable
here, thats in #debian-next on irc.debian.org (OFTC) and the BTS is
our online public bug tracking system bugs.debian.org where we list
known issues and report issues so they can be known and fixed
955[09:55:29] <hoijui> well.. before comming here, i joined
#debian-testing
956[09:55:31] <hoijui> and ther ewas nothing
957[09:56:22] <hoijui> maybe.. at least put a channel message
there that redirects people to #devbian-next
958[09:56:37] <rant> yes, well many years ago now, Debian
decided to move their official support irc.debian.org to
irc.oftc.net rather than freenode. We still have a larger and older
support community here but for the most part we only deal with
general debian stable stuff
959[09:57:16] <hoijui> yeah, i expected that, but as said.. how
would i know it is -next, and not -testing
960[09:57:18] <mmohr> Hey. What's the preferred Debian way
to cross-compile a Qt application from x86_64 (host) to
arm/raspberry pi (target)? qemu-debootstrap, Docker, ...?
962[09:58:33] <hoijui> also, sometime, for everyone, it has to
be the first time they hear about BTS ;-)
963[09:58:38] <hoijui> for me, it is now
964[09:58:59] <hoijui> i switched to debian one month ago (and
was not much on the PC Pc before that)
965[09:59:35] <rant> hoijui: I highly recommend you switch to
stable and let use help you learn the system.. ask a million
questions if you need to.. we have a great support community here.
You're really only doing us and yourself any favors using
testing/unstable if you know enough to help actually do the testing
974[10:01:30] <hoijui> and you only learn that, by first being
stupid, adn then (hopefully) not anymore
975[10:01:47] <hoijui> if that is unbearable for the community..
then, in a few years, there will be no tresters anymore
976[10:02:06] <rant> right well its not fair to users who come
here for our support to be hammering away at stuff dealing with
testing.. you limit your support resources by using testing/unstable
because we expect yuou to know this stuff
977[10:02:09] <hoijui> yeah.. which is exactly what happened
here :D
983[10:03:07] <rant> I understand your point though and I
agree.. which is why I've been working on a new project to help
new users and supporters and developers all get on the same page by
unifying our support resources and making easy access to the
information
984[10:03:20] <hoijui> so.. i will suggest in -next to make a
link in -testing
985[10:03:55] <hoijui> oh wow... that sounds like.. a huge plan
:D noas arch
986[10:04:01] <hoijui> good luck!
987[10:04:27] <rant> we can't link channels accross
networks all we could do is create a locked channel with an entrymsg
that tells you to go to OFTC
988[10:04:51] *** Quits: cdown (~cdown@replaced-ip) (Remote host closed the connection)
989[10:04:55] *** Quits: Adbray (~Adbray@replaced-ip) (Disconnected by services)
994[10:06:23] <hoijui> it is what i expect most users to do...
guessing first
995[10:06:33] <rant> if it weren't crickets in here right
now and others were looking in, you'd have gotten this rather
than an answer more than likely
996[10:06:39] <rant> !testing
997[10:06:39] <dpkg> Testing is a continuously updated release
between <stable> and <unstable>, currently codenamed
<buster>. See
replaced-url
998[10:07:11] <hoijui> and if each one gets the answer
immediatly, wihtout anyone else having to do soemthing, becasue he
can read it in the channel message
999[10:07:17] <hoijui> you save yourself time
1000[10:08:49] <themill> /topic is a thing already
1001[10:09:30] <hoijui> whaat?
1002[10:09:40] <hoijui> #debian-next (Channel is invite only)
1003[10:09:55] <rant> mmohr: I apologize I wasnt ignoring your
issue, I just dont have any suggestions.. while I have a pi,
I've not done any cross-compiling yet.
1006[10:10:24] <rant> hoijui: as I've said repeatedly, its
on OFTC not freenode. :P
1007[10:10:52] <hoijui> you said that this is your main server
now, not that this channel is not here
1008[10:11:09] <hoijui> you shoud lget to the fresh air more, my
friend...
1009[10:11:10] <themill> !#debian-next
1010[10:11:10] <dpkg> #debian-next is the channel for
testing/unstable support on the OFTC network (irc.oftc.net), *not*
on Freenode. If you get "Cannot join #debian-next (Channel is
invite only)." it means you did not read it's on
irc.oftc.net.
1041[10:30:02] <hoijui> is it agianst freenode policy to put a
channel message that suggest joining an other IRC server?
1042[10:30:25] <hoijui> ouh.. well.. this channel does it so.. i
guess not
1043[10:30:26] <rant> what I said was that OFTC is now the main
server.. thogu I'm kinda surprised that change lasted
considering by and large nobody other than the Debian project
recognizes the cahnge.. all the devs keep their clients pointed at
freenode by default, and most all the users and supporters still
come here
1044[10:31:16] <hoijui> yeah.... i have like. 40 channels here,
and now 1 on oftc
1045[10:31:47] <hoijui> (* 2 on oftc now)
1046[10:32:16] <rant> things were a bit hectic during the time
that switch occured but all it did in my opinion is fragment our
support community and did nothing to benefit anyone
1086[10:45:29] <petn-randall> petererer: The reason for skipping
it is not clear, it might be a good idea to file a bug report and
point to the upstream patch and the rationale why it's
patch-worthy.
1087[10:46:14] <petn-randall> petererer: But AFAICS it only
allows to create empty files with the permissions of the logged in
user, so the damage is fairly limited.
1089[10:47:47] <jelly> petn-randall: it's fairly limited to
DoS after filling up an ext[234]fs fixed number of inodes
1090[10:48:25] <jelly> ext4 is still the default fs type in
debian
1091[10:48:42] <petererer> It seems the rationale for not
patching it is "Minor issue" which is a pretty poor
description when the CVSS score is a Medium 5.3
1105[11:01:19] <petererer> I was looking for a rationale for why
it was not patched. I think I can piece that together now, just
would be nice if the CVE page stated exactly why it wasn't
considered worth patching, really.
1165[11:46:59] <maxcell_> yeah yeah i know how it is
1166[11:47:01] *** debhelper sets mode: +l 1564
1167[11:47:13] <maxcell_> so, your gonna help or not
1168[11:47:15] <maxcell_> :D
1169[11:47:33] <maxcell_> i mean, with my uinput error:
[error][controller 1] Failed to create input device:
"/dev/uinput" cannot be opened for writing
1170[11:48:57] <rant> idk, I suggested earlier using udev to
change the permissions
1171[11:49:16] <maxcell_> an udev rule you have said
1172[11:49:27] <maxcell_> thats not a flaw in security or
something/
1173[11:49:30] *** Quits: Smithe (~Smithe@replaced-ip) (Remote host closed the connection)
1174[11:49:54] <maxcell_> a regular user being able to run uinput
without sudo
1175[11:50:06] <rant> yeah.. but I suspect your problem may be
unrelated to permissions and have more to do with some buggy 3rd
party code.. cause I never had to do anything fancy to get a
controller to work
1180[11:52:13] <maxcell_> i uninstall it now i will reboot and
install again
1181[11:52:32] *** Quits: maxcell_ (~denis@replaced-ip) (Remote host closed the connection)
1182[11:52:44] <rant> maxcell_: as I said earlier I dont see how
installing mirocode could cause this issue unless something else
happened.. like a kernel upgrade or something
1191[11:59:00] <SanchoPensa> guys, what the heck am i gonna do
witht his? ACPI Warning: SystemIO range
0x0000000000000400-0x000000000000041F conflicts with OpRegion
0x0000000000000400-0x000000000000040F (\SMRG)
(20160831/utaddress-247)
1192[11:59:21] <SanchoPensa> i have a hangign bootsequence, i
DOEs continue after a while, but there is a timeout wait.
1193[12:00:12] *** Quits: v01t (~v01t@replaced-ip) (Remote host closed the connection)
1203[12:05:06] <rant> SanchoPensa: as is often the case with
errors from the kernel, what comes before or after is usually the
relevant part.. whats happening here is that something is trying to
claim an IO range that belongs to your ACPI SMRG and thats notmaking
it very happy.. whatever message comes before that is probably
indicating what it is thats trying to do this
1204[12:05:20] *** Quits: Baudelaire (~Baudelair@replaced-ip) (Remote host closed the connection)
1207[12:06:07] <SanchoPensa> rant: thank you so far! waht is an
SMRGß
1208[12:06:08] <SanchoPensa> ?
1209[12:06:15] <rant> SanchoPensa: while I could avoid this
error, I can't in good conscious tell you how to do that
without knowing more.. because it could result in instability which
can likely be solved some other way
1212[12:06:38] <SanchoPensa> rant: it might be the network
adapter.
1213[12:06:47] <rant> its the System Management Region of the
ACPI
1214[12:06:51] <SanchoPensa> or one of the two, cos the
motherboard actually has two of them
1215[12:07:15] <SanchoPensa> rant: the what? what does that do?
1216[12:08:11] <rant> basically what's happening here is the
ACPI in your system is trying to manage resources like IO ports
intelligently.. you can loosen up its control on this but it could
result in any number of instability issues.. when realy you need to
fix the conflict
1221[12:09:13] <rant> to even begin to know how to address this
issue I need to know what the rest of the message is.. the relevant
part.. this is the effect, not the cause
1222[12:09:24] <SanchoPensa> rant: yeah, i just duckduckwent to
see, and i found that conflict explained
1223[12:09:34] <rant> it'd also be nice to know if this just
suddenly started happening and if you changed something
1224[12:09:52] <rant> heh.. "duckduckwent" that's
a good one :P
1225[12:09:55] <SanchoPensa> rant: and yes, i agree, i am not
quite sure, how to detect that particular conflict, though...
1226[12:10:01] *** Quits: dselect (~dselect@replaced-ip) (Quit: ouch... that hurt)
1227[12:10:04] <SanchoPensa> dmesg isnt all to informative about
it...
1228[12:10:16] <SanchoPensa> heh! :)
1229[12:10:40] <rant> typically when you see kernel errors it
goes like this foo: bars \n bar: foo: barred me
1231[12:11:00] <rant> you only pasted the effect, not the cause
that came before it :P
1232[12:11:09] <SanchoPensa> rant: which is, what i was looking
for, but haven't found so far
1233[12:11:21] <SanchoPensa> but gimme a sec, pls. i will do a
grep on all them logs...
1234[12:11:25] <rant> usually its the very last line before the
message
1235[12:11:55] <SanchoPensa> rant: dude... I do untestand that,
do you on the other hand understand, that this is me asking for
assistance in finding it's origin? :P
1236[12:12:04] <rant> heh
1237[12:12:16] <SanchoPensa> put that tongue of yours out again,
and i shall slap it with a fly squatter! :P
1238[12:12:51] <rant> I dont understand how you got that message
and can't see the rest of it :P usually you're lookin at
dmesg or journalctl or such
1239[12:16:21] <SanchoPensa> oh boy
1240[12:16:58] <SanchoPensa> rant: keep your tongue inside, you
danmed perv, and other than that here:
replaced-url
1241[12:17:07] <SanchoPensa> am I fuckt or what??
1242[12:17:57] <rant> well buggy bios and such arent that
uncommon and are typically workable.. linux really doesnt care what
the f'n bios thinks once it has control of the system :P
1243[12:18:28] <SanchoPensa> not the issue, dude...
1244[12:18:40] <SanchoPensa> why is my root dir remounted read
only?
1245[12:18:44] <SanchoPensa> first of all
1246[12:19:00] <SanchoPensa> second: what the fuck is dbus trying
to connect with on the net??
1247[12:19:22] <SanchoPensa> and what the hell is the error that
results from its incapability to do so???
1248[12:20:03] <BanHammor> sounds fairly easy to me: it tries to
get an SMB share mounted
1249[12:20:13] <rant> seems gvfsd was trying to get on the SMB
network
1250[12:20:26] <SanchoPensa> like Couldn't create directory
monitor on smb://x-gnome-default-workgroup/. Error: The specified
location is not mounted
1251[12:20:32] <SanchoPensa> like, WAHT, WHAT??
1252[12:20:39] <SanchoPensa> dafuq is going on ?!?
1253[12:21:26] <SanchoPensa> hey, how do I fiond out, wha my root
dir is mounted read only?
1254[12:21:40] <BanHammor> is it actually mounted read only?
1255[12:22:06] <BanHammor> e.g. is the output of mount displaying
the ro option?
1257[12:22:39] <SanchoPensa> BanHammor: it is indeed, i just
doble chekced it
1258[12:22:49] <rant> same way you find out why something is
trying to claim your SMRG IO ports.. by reading the kernel msg
buffer rather than grepping logs that cuts out the whole picture
1259[12:22:58] <SanchoPensa> /dev/sda3 on / type ext4
(rw,relatime,errors=remount-ro,data=ordered)
1260[12:23:05] <BanHammor> that is not ro
1261[12:23:08] <SanchoPensa> oh wait
1262[12:23:14] <BanHammor> that is "errors=remount-ro"
1270[12:25:01] <SanchoPensa> yeah, bt why does it RE-mount?
1271[12:25:19] <BanHammor> because of initrd stuff, IIRC
1272[12:25:55] <SanchoPensa> BanHammor: i am confused... I dunno,
if you read aong before, but the task rant and I agreed upon, was to
find out, what is causing the IO conflict on the dbus
1283[12:28:23] <SanchoPensa> BanHammor: yeah, as said before, hen
i did that, there appeared to be a problem with the network adapter
1284[12:28:47] <BanHammor> cool. mind pasting the whole log so we
can actually look at it?
1285[12:28:55] <rant> may also want to lookinto that
systemd-analyze blame if your end goal is to address boot time
1286[12:28:57] <SanchoPensa> BanHammor: i do not know yet,
whether i am dealing with a multiple source problem here, like a fkt
up north bridge or something
1287[12:29:39] <SanchoPensa> systemd-analyze blame? geez, guys,
you are speaking in riddles todas... :D
1288[12:30:33] <rant> yes, systemd-analyze blame shows you how
much time is spent on each unit
1291[12:31:56] <rant> and personally I dont know the systemd-era
quiv of just getting /var/log/dmesg which no longer exists.. but if
your system hasnt been running for ages the dmesg command probably
still has all the boot messages in the buffer
1292[12:32:15] <rant> dmesg > dmesg.log
1293[12:32:53] <SanchoPensa> gusy, mind the line, where it says
Feb 20 11:58:09 sentinel kernel: [ 5247.859328] perf: interrupt took
too long (4924 > 4923), lowering
kernel.perf_event_max_sample_rate to 40500
1298[12:35:08] *** Quits: maxzor (~maxzor@replaced-ip) (Remote host closed the connection)
1299[12:36:11] * rant has no intention of minding any lines from
snippets of logs only containing pieces of the kmsg buffer to begin
with
1300[12:36:20] <SanchoPensa> ok, I got failed samba makedirs
replaced-url
1301[12:36:37] <rant> the linux kernel is verbose.. pretty much
every system has it complaining about something.. doesnt necessarily
indicate a problem
1304[12:37:07] <SanchoPensa> rant: duude. would you quit doing
that nick of yours honor, and just spit it out, what you want me to
do?
1305[12:37:10] <SanchoPensa> huh? ;)
1306[12:38:11] <SanchoPensa> yeah. i wonder.
1307[12:38:12] <SanchoPensa> why does gvfs want to create samba
directories.
1308[12:38:12] <SanchoPensa> i dont have any samba installed. or
at least, i shouldnt.
1309[12:38:26] <SanchoPensa> something is weird.
1310[12:39:55] <rant> I've said it at least 3 times now..
even specifically said dmesg > dmesg.log which is how you create
a logfile that actually has the kmsg buffer intact and unabridged
1318[12:43:30] <SanchoPensa> rant: and i dont like to put out my
interior organs on the net like this, you know? ;)
1319[12:43:40] <maxcell_> how can i make a udev rule to allow the
modules "uinput" and "hidraw" to have a
"660" (rw-rw----) permissions and the change the group
owner to that particular module "uinput" and
"hidraw" to root.imput rather then root.root. There is any
way i can do that without the need to change anymore modules owner
in /dev/?
1320[12:45:05] <rant> SanchoPensa: lines 607-609 were what I was
talking about, 607 shows the cause, 609 suggests a solution.. when
you take the middle part out, you lose the context of what's
actually happening
1321[12:46:10] <rant> maxcell_: that udev rules file you have
already does this.. it specifically looks for certain vendor/prod id
codes.. you just need to change the mode to 660 rather than 666 and
add a GROUP= param and make a group you want to use, and add your
user to it
1328[12:48:19] <rant> SanchoPensa: yeah we're gettin there..
I had assumed it was smbus bt I needed more info :P
1329[12:49:14] <bs0d> Hello. How to add a directory to $PATH in
Debian 9? I've tried editing ~/.profile and created
~/.bash_profile and added there PATH = $PATH:/usr/local/program, but
nothing did work. Am I missing something?
1330[12:49:27] <SanchoPensa> rant: oh do you now? :D
1331[12:49:33] <SanchoPensa> what do you need?
1332[12:49:55] <SPF> bs0d: remove the spaces
1333[12:50:37] <SanchoPensa> rant: can't upload syslog to
debian past, it's too long, and i'd have to truncate it
anways... so, where to look?
1334[12:51:31] <BanHammor> g0zzy, belatedly: /etc/defaults/grub i
think, ArchWiki has a decent article on editing grub
1336[12:51:57] <bs0d> SPF that did not change anything. I tried
to close and open again the terminal window, and the $PATH is still
unchanged. I have a suspicion that the ~/.profile file is not read
at all
1341[12:53:29] <bs0d> SPF no, people in linux forums generally
recommend against doing it, arguing that ~/.bashrc is not a proper
place for environment variables.
1342[12:53:37] <rant> SanchoPensa: only thing I know of offhand
is i2cdetect in the i2c-tools package I was trying to figure out
more.. I never looked up this info.. I know where things like
/proc/interrupts are.. idk where the heck info is on the smbus :P
1343[12:54:14] <rant> SanchoPensa: what about lspci -nn? that may
help
1344[12:54:53] <BanHammor> bs0d, who are those people?
1345[12:55:11] <rant> SanchoPensa: because whichever piece of hw
this is, is on the PCI bus.. that much I do know as of right now
1346[12:55:31] <SPF> bs0d: you could add for example "touch
/tmp/test" and see if it executes. Otherwise you have to view
the bash configuration files in /etc/bash*
1364[13:01:07] <SanchoPensa> rant: i was just gonna ask about
that number
1365[13:01:14] <SanchoPensa> where the heck do i find it?
1366[13:01:40] <tobiasBora> I'd like to know, when I install
debian, is it supposed to be installed in both "modes"
UEFI and CSM/legacy mode, or do I need to do something special to
install it in CSM mode?
1383[13:05:56] <BanHammor> (afaik debian cannot install in both
legacy AND UEFI mode at the same time, but i'll let other
people correct me on that)
1384[13:05:59] *** Quits: Haudegen (~quassel@replaced-ip) (Remote host closed the connection)
1385[13:06:27] <SanchoPensa> rant: and that despite: 00:1b.0
Audio device [0403]: Intel Corporation 82801H (ICH8 Family) HD Audio
Controller [8086:284b] (rev 02)
1386[13:06:27] <SanchoPensa> ?
1387[13:06:52] <rant> SanchoPensa: no, its an smbus controller
typically used for onboard sensors
1388[13:07:12] <tobiasBora> BanHammor: and then what should I do
to make my initial debian partition boot in csm mode? Because
I'd like to avoid to reinstall it in legacy mode
1389[13:07:23] <tobiasBora> (I won't have uefi mode after
that)
1390[13:07:25] <rant> SanchoPensa: you're looking for 1f.3
not 1b.0
1391[13:07:39] <SanchoPensa> rant: onboard sensors? yes, i know!
1402[13:09:55] <BanHammor> i honestly haven't tried this,
but i'm guessing you will need to boot something else via
legacy, chroot into debian, and run something like apt remove
grub-efi; apt install grub-pc
1408[13:10:44] <rant> SanchoPensa: that id is not the product id,
its the bus address.. in your system 1f.3 is mapped to vend/prod
code 8086:283e which is an intel SMBus controller
1413[13:11:45] <SanchoPensa> rant: it is mapped where?
1414[13:12:03] <rant> SanchoPensa: along the PCI bus.. heh..
1415[13:12:11] <SanchoPensa> that's what i was asking
before, if there was somethign like a device list in /proc or so?
1416[13:12:33] <rant> SanchoPensa: you got two ways of solving
this issue, either load a working driver or tell acpi to back off
and deal with the consequences
1417[13:12:34] <SanchoPensa> rant: dude, that is not where the
pci adress is mapped to the vendor ID :D
1428[13:14:26] <SanchoPensa> rant: i don#t know how many times i
gotta tell you, that it doesnt matter, iff its the sound or the
smbus, as the mainboard is still 9 yrs old?
1429[13:14:44] <SanchoPensa> rant: dude watch your language, if
you will, i am not your brat.
1450[13:19:49] <rant> SanchoPensa: thats where I have the issue..
if the driver is loaded, it should be avoiding these issue, which
leads me to think there is an i2c sensor its enabling that is using
that IO address and causing the issue.. which is why I suggested
i2c-tools to try figure out what sorts of i2c devices you got in
there.. this could be say a thermal sensor on your graphics card or
something.. who knows
1452[13:20:45] <rant> SanchoPensa: likely as you say, there is
something in /proc, /sys, or /dev that can get this info but idk
where.. all I know is there are utilities in i2c-tools that query
the right locations and tell you
1453[13:21:51] <SanchoPensa> rant: good place to start then,
thanks once more!
1457[13:23:07] <rant> SanchoPensa: you see how those lines I
picked out of your kmsg buffer all correlate.. pci 1f.3 points to
device 8086:283e which is the smbus controller, which points to io
0x0400-0x041f which is the address range complained of in the
error.. and we see i801_smbus enabling device 0001 -> 0003 and
thats where we dead end.. knowing what device that is on the smbus
1458[13:23:37] <rant> its right after the smbus controller
enables that device that the error occurs
1460[13:24:18] <SanchoPensa> rant: trail of breadcrumbs,
business, as usual... :D
1461[13:24:23] <rant> SanchoPensa: if this is the controller
itself, you may be able to pass an argument to the module, see
modinfo i2c_801 to see if it accepts and arguments related to IO
ports
1462[13:24:40] <SanchoPensa> oh boy...
1463[13:24:50] <SanchoPensa> who are you really, Katy?
1465[13:25:45] <rant> SanchoPensa: the end all is to pass
acpi_enforce_resources=lax to the kernel command line at boot and
tell ACPI to back off :P if you can't work around another way
1466[13:27:09] <SanchoPensa> rant: i gotwo other options: to
ignore that 1 Minute wait at boot time, or to just turn the damned
ACPI off in BIOS
1535[14:03:19] <tobiasBora> BanHammor: The thing is that for some
strange reason, the computer of a (remote) friend does not boot
since the installation of debian 9, neither on Windows or debian.
1536[14:03:38] <tobiasBora> it just bootloop on the Bios setup
(EZ mode)
1537[14:03:47] <BanHammor> that sounds...very familiar.
1538[14:03:55] <tobiasBora> but he can still boot on usb
1539[14:03:57] <tobiasBora> BanHammor: really ?
1540[14:04:03] <BanHammor> you talked about this here before,
right?
1541[14:04:07] <tobiasBora> yes
1542[14:04:29] <tobiasBora> my friend was not here for a week, on
I will talk with he in a few mn
1543[14:04:36] <tobiasBora> to solve the problem
1544[14:05:19] <tobiasBora> that's why meantime I try to see
if it has some chance to boot on CSM mode even if it was not install
in CSM mode
1545[14:05:35] <tobiasBora> so I guess there is no chance
^^'
1553[14:09:46] *** Quits: RebelCoder (~RebelCode@replaced-ip) (Remote host closed the connection)
1554[14:10:33] <regenkind> So what I'd like to do is to
partition the linkstation via my ssh connection and prepare the link
stations filesystem and then reboot into the new kernel
1562[14:12:40] <petn-randall> regenkind: Personally I'd also
go with a current release, wheezy is practically EOL.
1563[14:12:43] <petn-randall> !wheezy-lts
1564[14:12:43] <dpkg> Security support for Debian 7
"Wheezy" from the Debian Security Team ended on
2016-04-25. The amd64, i386, armel and armhf architectures receive
additional long term support (LTS) via <wheezy/updates> until
2018-05-31. See
replaced-url
1565[14:12:45] <regenkind> Ok, so I absolutely need a secondary
boot device.
1567[14:13:33] <regenkind> Will do update late for sure, but
first I want to get it up and running
1568[14:13:42] <petn-randall> regenkind: Yes, which is why
I'm assuming it's easier to take the device out. Also,
many steps on the tutorial are IMHO bad practice, such as
downloading the debootstrap.deb and installing it without dependency
resolution.
1569[14:14:34] <regenkind> Definitely is, as the mentioned
version did not work, had to go back one version earlier even.
1570[14:15:03] <petn-randall> regenkind: I don't know the HW
details of that device, but there might easier ways to install
Debian on it. That tutorial is already several years out of date.
1571[14:15:04] <regenkind> But it build without complain, so ...
will try it.
1572[14:15:28] <petn-randall> regenkind: Do you already have
Debian on your workstation?
1573[14:15:33] <SanchoPensa> rant: you know what? i2cdetect -l
doesnt detect anything. just returns NILL
1574[14:15:59] <regenkind> No, I am using manjaro on my main
server but the scripts worked alright
1575[14:16:38] <regenkind> On the link station I created the
chroot env as described and that worked all out
1576[14:17:05] *** Quits: RebelCoder (~RebelCode@replaced-ip) (Remote host closed the connection)
1577[14:17:13] <regenkind> The problem with the buffalo firmware
is that there are no updates for the latest security issues
1579[14:18:18] <regenkind> Therefore I want to transform the link
station to a Debian home server with OpenMediaVault
1580[14:18:33] <petn-randall> regenkind: Well, wheezy will only
get security updates until the end of May, so that won't buy
you much.
1581[14:18:52] <petn-randall> !omv
1582[14:18:52] <dpkg> OpenMediaVault is a NAS operating system
with a web-based control panel. It is <based on Debian> but
not supported in #debian. Support can be found at
replaced-url
1583[14:19:23] <regenkind> Well I thought once I have Debian
installed I thought I could do a distribution update ?
1584[14:19:24] <petn-randall> regenkind: Note the omv and Debian
are separate projects, so you might want to ask for support in their
channel. ^^^
1586[14:19:48] *** Quits: xcm (~xcm@replaced-ip) (Remote host closed the connection)
1587[14:19:55] <regenkind> The openMediaVault is not that
important.... More like an example
1588[14:19:56] <petn-randall> regenkind: I'm not sure that
would even work. If you want omv, I'd try to install that.
1589[14:20:34] <regenkind> I just want a regular linux
distribution on this box, either arch or Debian is fine
1590[14:20:55] <BanHammor> tobiasBora, sorry, i got distracted:
you should ask your friend to reinstall in CSM mode, it will be MUCH
easier than mucking about with chroot
1593[14:22:08] <tobiasBora> BanHammor: the thing is that if he
reinstall with CSM, he won't be able to boot on Windows anymore
1594[14:22:11] <petn-randall> regenkind: I'd try installing
it as described in the release notes, you'll have the best
chances for it to boot that way.
1595[14:22:14] <tobiasBora> and I think he still need windows
1599[14:23:18] <petn-randall> tobiasBora: At this point it's
probably best to get your friend into the channel to respond, that
would save you the time relaying ideas back and forth.
1600[14:23:31] <petn-randall> (just an idea)
1601[14:24:01] <BanHammor> tobiasBora, i think debian should
still pick up the windows install via os-prober
1640[14:53:53] <opv> hey guys, i have an inquiry about a debian 8
server running on a 10G drive. there is a 1.8G discrepancy between
"df -h" and "du -sch / 2>/dev/null". can you
help me make sense of this?
1656[14:58:47] <alexises> Hello, I need to be able to use md5
signed certificate under debian 9. How to renable md5 support ?
1657[14:59:06] <jelly> (tell it not to du other filesystems)
1658[14:59:13] <opv> thank you very much, dmn_ , jelly
1659[14:59:42] <opv> i've identified a couple of deleted
files, but somehow can't have the processes stop using them. we
have swapspace and apache. i've done an apache restart, and a
swapon;swapoff, but no change in the opened, deleted files
1660[14:59:44] <jelly> alexises: what kind of certificate, with
which tool? What's the exact command that fails and its output?
1694[15:19:21] <b-04> I have a problem with squid 3.5,
specifically with the site of linkedin.com, sometimes when they
enter the site it looks broken, like not loading the whole page, the
css sheets and some js. I do not cache with the proxy, and the
domain linkedin.com as licdn.com are in an acl of allowed without
authentication. Any idea what could it be?
1699[15:22:16] <metastable> alexises: Under any other
circumstance, I wouldn't advise this, but given that you have
little choice, there is ONE thing you can try.
1701[15:22:35] <metastable> alexises: I don't guarantee
success. Can you pastebin your nginx configuration, so I can
understand better what's going on?
1721[15:34:32] <oo_miguel> tw: works perfectly thank you
1722[15:35:44] <tw> the bpf syntax is a bit annoying to get used
to, and definitely use -n because dns lookups will bottleneck your
dumps, but it's a really useful network analysis tool to learn.
1723[15:36:23] <oo_miguel> yeah. the syntax is very confusing. I
just copy and paste examples I find, and adapt them with litttle
understanding :P
1724[15:36:26] <tw> wireshark is a similar gui tool, but you
still should use the bpf filter and not the gui filter.
1725[15:36:37] <oo_miguel> I prefer a cli solution
1726[15:36:45] <oo_miguel> do not have a gui available everywhere
1727[15:37:10] <BCMM> is there an easy way to get a list of
packages that weren't installed from repos? (e.g. installed
with wget ; dpkg -i)
1728[15:37:23] <oo_miguel> tw: thank you once again!
1729[15:38:13] <tw> oo_miguel: iptraf-ng may also be interesting
to you.
1746[15:45:14] <metastable> alexises: Okay, that's fair.
Sorry, I get hyper-annoyed when I see people acting like their
192.168.x.x addresses are SOOPER SEKRIT.
1747[15:46:17] <alexises> metastable: it could be if it's
can be linked to the relevent company, it's gain (a little bit)
of information about internal addressing
1755[15:50:09] <metastable> alexises: I dismiss the notion that
your internal IP scheme is somehow sensitive as a violation of
Shannon's Maxim. If I can initiate a connection to the system,
I'm inside your network, and nmap will tell me anything I could
want to know about your IP scheme anyway.
1756[15:50:22] <metastable> alexises: Crap. That was the only
thing I could think of that might force OpenSSL into accepting the
CA.
1757[15:50:35] <tw> OPENSSL_ENABLE_MD5_VERIFY=1
1758[15:50:51] <tw> in nginx startup environment.
1759[15:51:15] <alexises> ok, I check this :)
1760[15:51:45] <metastable> tw: You're a font of knowledge,
today.
1860[17:02:06] <cluelessperson> I did this command, dpkg
--add-architecture i836
1861[17:02:09] <cluelessperson> is that bad?
1862[17:02:20] <cluelessperson> and if so, how do I undo before I
do anything else?
1863[17:02:21] <somiaj> cluelessperson: with that typo?
1864[17:02:45] <somiaj> cluelessperson: it shouldn't be, you
can use --list-archietecture or something like that to see if it
actually accepted it, then there is a remove command
1882[17:06:09] <somiaj> cluelessperson: I don't know if you
can see when it was added. But on an amd64 system that is safe and
many do it so they can run 32bit software.
1883[17:06:27] <somiaj> cluelessperson: dpkg -l | grep i386 (look
for packagename:i386) well let you know if you have anything
installed from the i386 repos.
1885[17:07:38] *** Joins: solaris (~solaris@replaced-ip)
1886[17:07:39] <cluelessperson> somiaj: ah, looks like I was
previously using it
1887[17:07:47] <cluelessperson> somiaj: Also looks like Steam is
32bit based
1888[17:08:14] <somiaj> correct, many games are still made only
32bit since they run just fine on 32bit and 64bit oses
1889[17:09:50] <ageis> I'm confused about the output of dpkg
--force-help
1890[17:10:22] <ageis> [!] confask Offer to replace config files
with no new versions -- shouldn't that say "new
versions" not "no new version"
1891[17:10:28] <somiaj> ageis: in general forcing things is
usually not the way to fix problems. Also what part is confusing
you?
1892[17:10:43] <ageis> is the [!] an inversion?
1893[17:11:01] <somiaj> read the full output. WARNING - use of
options marked [!] can seriously damage your installation.
1894[17:11:14] <ageis> let's say I want to --force-confmiss,
(install missing,) and also force-confask ( ask about ones already
present)
1895[17:11:38] <jelly> !confmiss
1896[17:11:39] <dpkg> You have to especially tell the packaging
system to reinstall config files because when they are gone, it is
assumed that you want them to stay deleted. "aptitude -o
DPkg::Options::='--force-confmiss' reinstall
$packagename" will restore them (man dpkg for details). If the
package uses <ucf> for config file management, ask me about
<ucf confmiss>.
1897[17:12:31] <jelly> most real life situation, you won't
do it directly with dpkg, ageis
1898[17:12:46] <petn-randall> ageis: The man page is more verbose
and uses correct english. You might want to file a bug report to
clarify the terse output.
1899[17:12:47] <ageis> I have a specific use case for this
1911[17:16:04] <somiaj> ageis: I think the no is correct, it
basically forces fvwm to prompt about replacing config files if
there is no new version aviable. By defualt it does ask to replace
when there are new versions avaiable.
1912[17:16:16] <somiaj> ageis: arg my mind is elsewhere, forces
dpkg
1913[17:17:02] *** debhelper sets mode: +l 1560
1914[17:17:04] <jelly> ageis: what do you mean, specific dir
1915[17:17:32] <ageis> confmiss prompts when there are new
version available?
1932[17:21:29] <jelly> the manual pretty much says you CAN use
--force-confmiss and --force-confask at the same time in the
description for confask
1933[17:21:34] <ageis> lke I could rm -f /etc/pam.d/* beforehand,
use conf-miss, and anything outiside of there willl be prompted
about, but ill get the new files
1939[17:22:27] <jan_nabot> Command ' setterm --dump '
supposedly 'writes current virtual console snapshot' to
file. I run it on Debian Xterm Tmux and all it returns is
replaced-url
1943[17:23:14] <ageis> my thing is that these various autologin
PAM things have stuff insecure... and i'm gonna do this for
/etc/dbus-1 too cause a ton of dbus contexts are broken
1982[17:33:38] <wrksx> lots of nice feedback, thank you guys
1983[17:33:44] <jelly> wrksx: either use root or su -
replaced-url
1984[17:35:20] <wrksx> tw: looks like some aliases were rightly
setup by the system (mailer-daemon -> postmaster, postmaster
-> root, root -> <myuser>)
1985[17:35:32] *** Quits: solaris (~solaris@replaced-ip) (Quit: solaris)
1986[17:35:38] <jelly> wrksx: alternatively, if you're the
only admin, let all the relevant stuff deliver to your user via
/etc/aliases
1987[17:36:00] <wrksx> yeah I'll probably end up doing that
1988[17:36:11] <ageis> but yeah aliases is the best way
1989[17:36:25] <jelly> then filter those however you like if
needed
1990[17:36:30] <tw> remember to run `newaliases` after updating
the /etc/aliases file
2074[18:17:18] <laurus> Does anyone know a simple way to convert
a PDF that has filled-out form fields to a PDF that is exactly the
same but where everything is just an image?
2102[18:25:54] <gr8> are the default repositories ok for having a
stable system without surprises, or should I disable the repos
except the security updates?
2147[18:55:28] <Aebian> if I want all files within a folder to be
immutable even when new files are being uploaded, should I go with
'chattr +i' or is there a better way?
2163[19:04:24] *** Quits: RebelCoder (~RebelCode@replaced-ip) (Remote host closed the connection)
2164[19:04:40] <alkisg> Aebian: same as in /tmp, any user can
create, but they can delete only their own files. And you can
periodically chown them to nobody if you don't want to consider
them their own anymore
2178[19:09:47] <donald> Hi. I made a lab to experiment virus with
virt manager. I want to be sure they can not infect any computer via
internet while test there power via intern network. When I check the
NIC of any computer of the network, the source of the network is:
"virtual network x: network isolated, intern or routing via
hypervisor only.". can a virus or any component of my system
access internet via the hypervisor? or else?
2179[19:09:48] <tw> Maybe incron w/ IN_CRON_WRITE and chown to
another user?
2180[19:09:58] <tw> *IN_CLOSE_WRITE
2181[19:10:22] <tw> openssh doesn't support post-write hooks
that I'm aware of.
2183[19:11:35] <tw> Though you might be able to do post-logout
actions in PAM but that sounds difficult to get right.
2184[19:11:56] *** Joins: wilson (~wilson@replaced-ip)
2185[19:12:08] <jelly> Aebian: if the user has +w and +x
permissions on the directory they will be able to delete any file,
owned by any user. What alkisg says is the simplest thing to do but
inherently racy.
2204[19:18:29] <tw> donald: short version is no virtualization
platform is perfectly secure, so it's a matter of acceptable
risks. You can delete the nic or set it to none, eg the `-net none`
qemu option.
2205[19:18:31] <jelly> donald: there's really no need to
repeat after just 7 minutes
2206[19:18:58] <tw> by default qemu supports usermode networking,
so test it first.
2207[19:19:08] <tw> (even with no NIC assigned)
2208[19:20:16] <Freenoodle> just trying to install sid on a Dell
laptop. Installer says it needs firmware. But it does not detect the
usbstick, which contains the needed file in /lib/firmware. So I
follow the advice at
replaced-url
2210[19:21:04] <donald> tw I can not remove the nic. else I will
not be able to connect two virtual machine together
2211[19:21:26] <BanHammor> how does one even install sid,
Freenoodle?
2212[19:21:48] <metastable> !unstable
2213[19:21:48] <dpkg> Unstable is the status of a Debian release
when packages can be added at any time, that can disrupt the
integrity of the whole system! Ask me about <break>. If you
have to ask whether you should use it or how, you shouldn't.
replaced-url
2214[19:22:05] <metastable> And we also don't support
unstable in this channel.
2216[19:22:26] <tw> donald: You will need to test your
configuration manually to see if it can escape. That's not
something I can help with though.
2217[19:23:26] <jelly> detecting a hypervisor breach is left as
an exercise for the reader?
2218[19:23:47] <tw> More like "unknown default device
options may be created that you didn't specify."
2219[19:23:59] <jelly> might as well run the VM without kvm,
fully emulated
2220[19:24:05] <donald> tw did you say " short version is
no. Virtualization platform is perfectly secure, so it's a
matter of acceptable risks. You can delete the nic or set it to
none, eg the `-net none` qemu option." or "short version
is: no virtualization platform is perfectly secure, so it's a
matter of acceptable risks. You can delete the nic or set it to
none, eg the `-net none` qemu option." ?
2221[19:24:26] <jelly> on an air-gapped raspberry pi without
wifi!
2222[19:24:50] *** Joins: solaris (~solaris@replaced-ip)
2223[19:25:00] <tw> The latter. People do what you're trying
to do with qemu, but you'll have to come up with a
configuration that meets your needs specifically.
2224[19:25:19] <tw> And even in non-kvm mode, it creates some
default devices.
2225[19:25:28] <jelly> donald: "no virtualization platform
is perfectly secure" is QUITE different from "no.
Virtualization platform is perfectly secure"
2235[19:30:54] *** Quits: alexgr (~alexgr@replaced-ip) (Remote host closed the connection)
2236[19:30:55] <donald> tw when I defined my policy of security I
just supposed the virtualization software is enough secure. I just
ask of if the software is nicley configured to block a virus who has
a complete control of the system, the configuration of wifi and is
made to connect to internet.
2238[19:36:04] *** Quits: silverwhitefish (~hidden@replaced-ip) (Quit: One for all, all for One (2 Corinthians 5))
2239[19:36:06] <tw> The default configuration through libvirt is
not something I'm qualified to answer, I'm afraid. If it
were me, I'd go into the qemu monitor after setting up the VM
to see what devices were created and check them. I'd also
disable acpi and a number of other APIs that are passed through by
default.
2278[20:10:04] <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>.
2279[20:10:10] <petn-randall> hid3: This maybe? ^^^
2315[20:19:25] * jelly almost always keeps a separate /boot, that works
well with older versions of stuff and interesting configurations
(LUKS) if need be
2316[20:19:43] <hid3> I ran grub-probe --device /dev/sda1
--target fs and I get unknown FS
2317[20:19:52] <hid3> okay, I will reformat it to EXT3
2318[20:20:03] <jelly> reformat what?
2319[20:20:14] <hid3> wipe the disk nad start over
2334[20:24:09] <jelly> it's not safe to assume the issue is
caused by xfs, maybe you missed an important step and if you do
things exactly the same but with ext3 this error happens again
2345[20:32:56] <jelly> also, xfs probably has different features
based on which version of its mkfs you use. ext4 created with, say,
a debian testing live cd, will be unusable on debian 7 or 8 because
it defaults to enabling too new features
2346[20:33:31] <hid3> might be
2347[20:33:38] <hid3> so I created ext3 for now, like it was
2348[20:33:46] <jelly> nod
2349[20:33:47] <hid3> will uprade to debian 9 (curently it's
7)
2350[20:33:58] <hid3> then fix the hardware I'm temporary
migrating off
2351[20:34:06] <hid3> and then try xfs there
2352[20:34:11] <hid3> (upon migration back)
2353[20:34:20] *** Quits: solaris (~solaris@replaced-ip) (Quit: solaris)
2354[20:34:21] <jelly> I have like 60 debian 7 installs to
upgrade
2395[20:52:55] <Sansar> Greetings, I installed fprintd, and other
dependencies, but unlike my previous debian installations I
can't use fingerprinting device for user auth. or login
2396[20:53:03] <dumbl3d0re> What's considered best practice
to name the directory that have the same function as the /opt
directory but is located locally in a users home directory?
2397[20:53:07] <Sansar> Any idea what may cause this?
2402[20:55:49] <apteryx_> hello! Is there a channel dedicated for
packaging in Debian?
2403[20:55:54] <apteryx_> or should I ask here?
2404[20:57:46] <somiaj> apteryx_: do you want to learn how to
make .deb packages, have you read the debian matainers guide? Or do
you want to get a .deb package into debian?
2406[20:58:51] <petn-randall> apteryx_: There are two channels:
#debian-mentors for if you want to get software packaged *for*
Debian, and #packaging if you want learn how to use the packaging
toolchain but *not* package for Debian. Both are on irc.oftc.net
(not freenode).
2409[21:00:09] <apteryx_> somiaj: I want to learn how to make
.deb packages. Well, I have a good start at it (already functional
package), but I'm currently struggling with 'gbp
buildpackage --git-pbuilder'.
2410[21:01:11] <apteryx_> petn-randall: OK, thanks. I'll
check into #debian-mentors.
2424[21:10:12] <jelly> petn-randall: if I had a fresh new ansible
setup to, er, set up, would I want 2.2 from stretch or 2.4 from
stretch-backports?
2425[21:10:16] <apteryx_> somiaj: thanks
2426[21:11:30] <petn-randall> jelly: I'd go with backports.
2427[21:11:36] <jelly> coworker has already made decent progress
with 2.2 but apparently 2.4 has things for vmware
2428[21:11:45] <jelly> THINGS
2429[21:12:54] <greycat> as in "a few of my favorite",
or as in "that should not be"?
2430[21:13:12] <petn-randall> jelly: ansible has deprecation
cycles over several releases, so a 2.2 playbook (not using
deprecated stuff) should work fine with 2.4, albeit getting
warnings, but you'll have time to fix those before upgrading to
2.5.
2431[21:14:09] *** Joins: solaris (~solaris@replaced-ip)
2436[21:16:15] <petn-randall> jelly: And also the things that get
deprecated are few and far inbetween. In 2.4 it was a larger leap,
they deprecated the include function, but I simply replaced it with
include_tasks everywhere.
2441[21:17:34] <petn-randall> jelly: Currently the amount of
modules is growing fairly fast, for example the windows modules were
lacking a bit but they catched up quite a bit. So IMHO it's
best to go with the latest release and upgrade when needed.
2442[21:17:38] <petn-randall> jelly: There is?
2443[21:17:57] <jelly> I forget, will have to drag coworker to
irc
2444[21:18:06] <jelly> but he found at least 2-3 things
2445[21:18:16] <petn-randall> jelly: I don't really use the
CLI settings much, unless you want to use one-liners to reboot
machines or something.
2451[21:18:56] <petn-randall> But even then it's smarter to
write a small playbook, so you don't have to remember the exact
command line that does that.
2468[21:24:37] <jelly> donald: but avoid repeating the same
question verbatim; include the results of suggestions from the last
time you asked if there were any that were worth trying
2489[21:34:02] *** Quits: darkhanb (~textual@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
2490[21:34:55] <jelly> DammitJim: it's really hard to say
without more info, ie. how precisely is this checking done; whether
the app is keeping CWD there or not
2491[21:36:05] *** Quits: nuno_nunes (~PC@replaced-ip) (Remote host closed the connection)
2492[21:36:21] <toli> guys, i have tried to install wicd and
remove the networkmanager, and since, I cannot connect to lan nor
wifi. it connects it gets ip for couple of seconds, then again no
ip. I have message an error message saying (network is unreqchqble |
dhclinet.c.2733: Failed to send 300 byte long packet over fallback
interface
2493[21:36:43] <DammitJim> good points jelly !
2494[21:36:51] <toli> I know I can reinstall debian, but i
don't really want this
2495[21:36:56] <DammitJim> what is cwd?
2496[21:37:16] <greycat> current working directory
2497[21:37:19] <DammitJim> oh
2498[21:37:20] <jelly> DammitJim: current (wirking) directory,
each process has one
2499[21:37:30] <JustASlacker> cwd=$(pwd)
2500[21:37:46] <greycat> or as in "man getcwd"
2501[21:38:13] <jelly> DammitJim: did you ssh to the same user as
the one running the app, eth
2502[21:38:16] <jelly> etc
2503[21:38:21] <DammitJim> the application runs as a particular
user called quartz and the current working directory is something
like /usr/local/myapp/
2505[21:38:36] <DammitJim> jelly, I ssh'd as myself, then
sudo su - quartz
2506[21:38:39] <jelly> are you SURE that's the current
process cwd?
2507[21:38:47] <memo1> hi, im working with video files (mp4)
store on folders. What is the best way to sort them and selected by
date?, or filter and copy by date?
2508[21:38:53] <jelly> DammitJim: confirm with lsof
2524[21:41:12] <toli> no one have any idea of what is going on?
2525[21:41:20] <DammitJim> but the java application checks on
/mnt/myserver_data/opportunity.csv
2526[21:41:26] <DammitJim> how does that relate to cwd?
2527[21:42:03] <jelly> toli: you can akways stop wicd and
configure /etc/network/interfaces manually ?
2528[21:42:23] <greycat> DammitJim: The reason it was brought up
was because, if your program *sits* in /mnt/mysever_data/ full time,
then it might run into stale file handles or whatever. Or maybe
that's only NFS and not Samba.
2529[21:42:47] <greycat> DammitJim: and it costs you very little
time to find the PID and lsof it to see what its cwd is.
2538[21:44:44] <DammitJim> but now the java application seems to
only "see" the file if I stop the application and start it
again
2539[21:44:54] <DammitJim> but the next time the file is placed
there, the application can't find it
2540[21:45:11] <DammitJim> jelly, that's a very good
question... I'm not sure of that
2541[21:45:30] <DammitJim> hhhmmmm... I wonder what logs I would
have to look for to determine that
2542[21:46:16] <DammitJim> man, but you guys are great! I would
have thought you would have sent me to #java already, but you have
been good at making sure it's not Debian (or me)
2543[21:46:20] <jelly> DammitJim: you should probably debug
exactly what the app does. Maybe it keeps directory fd open and uses
fstatat()
2544[21:47:01] <DammitJim> oh gosh, I am going to be learning
more about fd's, huh?
2550[21:48:03] <jelly> I have no idea how java checks for path
existence.
2551[21:48:23] <DammitJim> oh then what did you mean about
fstatat()?
2552[21:48:46] <DammitJim> as I said... at this time, if I
restart the application, it finds the file, processes it, but the
next time the file comes in, it doesn't see it
2553[21:48:47] <jelly> just guessing based on the few symptoms
you described
2561[21:50:25] <DammitJim> but it's true... I haven't
changed anything on the server sharing the directory over samba nor
the server that is running the java app
2562[21:50:28] <greycat> DammitJim: Basically, C is the only real
programming language in Unix, and everything else is layered on top
of it.
2563[21:50:53] * jelly waiting for libc written in rust.
2564[21:50:54] <DammitJim> oh, so I can't use fstat directly
on the command line, right?
2565[21:51:06] <greycat> the command line equivalent is stat(1)
2566[21:52:25] *** Parts: donald (~donald@replaced-ip) ()
2567[21:52:32] <jelly> fstat and its younger *at sibling are
syscalls, those things in section 2 of the manual which you can get
by installing manpages-dev
2587[21:58:17] <toli> when i try to manually add an ip address
using (ip addr add XX.XX enp0s31f6) re response is a bit scary
(Error: either "local" is duplicate or enps31f6 is a
garbage
2600[22:05:05] <laertus> after doing an "aptitude
update", "aptitude upgrade", and rebooting a debian
system i'm running under virtualbox, after it boots it goes in
to emergency mode and i don't know why
2601[22:05:07] *** Joins: solaris (~solaris@replaced-ip)
2602[22:05:13] *** Quits: solaris (~solaris@replaced-ip) (Client Quit)
2603[22:05:13] <laertus> here's the boot log:
replaced-url
2616[22:11:08] <^||> i cant get sftp to work. example: put
stallman.jpg / error: "Couldn't open local file
"stallman.jpg" for reading: Permission denied
2617[22:11:34] <^||> i tried chown outside sftp for the remote
user but comes up with invalid user
2622[22:14:15] <jelly> ^||: what does "id remoteuser"
say on the remote system, and is the remote running debian and which
release
2623[22:15:23] <jelly> ^||: couldn't open local file for
reading sounds as if... the user running sftp _client_ does not have
permissions to, well, read the file
2684[22:47:17] <tobiasBora> BanHammor: Problem solved, for some
reason my friend aborted the debian install and I don't know
why but debian removed all entries in /boot/efi, including windows
ones. After a clean reinstall debian came back, and Windows should
come back after a few copy
2780[23:32:15] <levkusanagi> Mathisen: I get the following error
when trying to insert guest additions cd image: Could not mount the
media/drive
'/Applications/VirtualBox.app/Contents/MacOS/VBoxGuestAdditions.iso'
(VERR_PDM_MEDIA_LOCKED).
2802[23:39:21] <levkusanagi> Mathisen: I force unmounted the
current image (incidentally, the guest additions image), remounted,
and behold! Autorun dialogue pops up. It asks if I want to run it..
then: "Oops, there was a problem running this software. Unable
to locate the program"
2803[23:40:13] <Mathisen> when you pressed run ?
2804[23:40:48] <levkusanagi> Yes
2805[23:41:26] <Mathisen> im not sure then. it should not do
that. if it had worked as it should you would have gotten the
guest.iso mounted and could install it
2806[23:41:36] <Mathisen> someone else may know.
2807[23:42:10] <levkusanagi> Thanks for the help Mathisen! I
appreciate the time taken.
2809[23:42:20] <rant> are you installing Debian as a guest in
virtualbox on MacOS?
2810[23:42:57] <levkusanagi> rant: That's what I've
done, I'm trying to fix the limited screen resolution issue
(only 4:3 resolutions) by installing virtualbox guest additions, but
I cant
2811[23:43:42] <rant> right well the virtualbox-guest-dkms is the
best way for a debian guest however the issue there is that the
version may not match the macos version of vb you're using and
thus likely will not work
2813[23:44:29] <rant> you would need to mount the vb guest
additions iso as suggested, but screw all that autorun crap.. just
get a root prompt and run the linux additions file manually
2814[23:44:55] <levkusanagi> rant: Brilliant!
2815[23:45:05] <rant> I have my moments :P
2816[23:45:37] <Mathisen> rant, how would he mount it when the
gues.iso is on the host os
2817[23:45:54] <levkusanagi> I succeeded in mounting, it's
the autorun that's crashing apparently
2825[23:47:14] <rant> you may still need to have kernel headers
installed.. I dont recall exactly as I typically dont run the best
OS on the planet under other crappy OSes :P
2834[23:48:46] <rant> they're both two of the best made
laptop hws and have well documented proceedures for linux
installation on em
2835[23:48:54] <levkusanagi> I'm seriously going to look
into that. I'm new to Debian but I feel at home with its stable
philosophy
2836[23:49:12] <rant> levkusanagi: try ls -l
vboxlinuxassitions.run or w/e its called
2837[23:49:36] <memo1> im running a python script as a service.
The scripts needs to create a file.log, but its failing. Usning
systemd service, i cant create files?,
2838[23:49:40] <rant> levkusanagi: I read Why Debian? and Social
Contract on Debian.org some 16 years or so ago, and I was sold
2839[23:49:47] <levkusanagi> -r-xr-xr-x
2840[23:50:02] *** Quits: jazz (~jazz@replaced-ip) (Remote host closed the connection)
2841[23:50:16] <rant> levkusanagi: and you have a # prompt, id
returns 0, and you did ./filename?
2843[23:51:11] <levkusanagi> rant: It returns "Permission
denied", not sure what you mean about id returns 0
2844[23:51:30] <rant> levkusanagi: if thats the case, then I need
the message verbatum.. not truncated to "permission
denied" as for all I know that could be a message from a
subshell or child process
2845[23:51:41] <levkusanagi> Sorry, incoming
2846[23:51:48] <rant> levkusanagi: the id command tells you what
the current user is
2847[23:52:32] <levkusanagi> id returns all 0
2848[23:52:46] <zer0G> levkusanagi, chmod +x the file so
it's executable.
2849[23:52:49] <zer0G> as root.
2850[23:52:54] <levkusanagi> got it
2851[23:53:08] <rant> we covered that already.. he said the file
is -r-xr-xr-x
2859[23:55:07] <rant> ok well thats bash suggesting that the file
is not executable.. I'm beyond stressed right now so other than
the obvious of file permissions I'm not thinkin real clearly
right now