7[00:03:47] <rydare> might get my head bit off for this, but
do you have a spare USB stick? I would have you boot a ubuntu livecd
to see if the buttons work, because i found a post from a month ago
with someone on debian having problems while a ocmmenter using
ubuntu said it was fine on their machine
8[00:03:59] <rydare> then we can figure out what ubuntu is
doing differently
67[00:14:50] <d0uglas> hey rydare, thanks for your help
yesterday.. I fired up memtest which has been running a solid 21
hours, 638 errors. Went into bios (new mb), adjusted the voltage and
speed of the ram in line with the product specs, ran memtest again,
no dice. Based on this -
replaced-url
97[00:22:26] <d0uglas> so i enabled it in bios, now in grub,
editing commands, one that's up top is setparams '...
' .. do I want to tuck iommu=soft inside its quotation marks,
or a separate line?
126[00:37:29] <d0uglas> This time I hit f7 or the right arrow by
mistake at the boot menu, didn't pass iommu, new weirdness like
rcu_sched kthread starved for 20541 jiffies! (gibberish) ... NMI
watchdog: BUG: soft lockup - CPU#[1,2,..8] stuck for 22s!
[systemd-udevd..]
127[00:37:38] <lenswipe> ah
128[00:37:44] <rydare> i'm on testing
129[00:37:48] <rydare> but it does come with issues
130[00:37:57] <rydare> i find it better for a desktop use
131[00:38:03] <rydare> for a server i'd definitely use
stable
205[01:47:27] *** Quits: will_haven (~will_have@replaced-ip) (Quit: I was chosen because I was the bravest, the most worthy.
The honor was mine until another came
to{#`%${%&`+'${`%&NO CARRIER)
216[01:57:52] *** jdd is now known as baconicsynergy
217[01:58:07] <syrius> mmm. ChmEarl Build-Depends has all the
cross-platform stuff there, but also has <!nocheck> after each
line
218[01:58:31] <syrius> i ended up just compiling sources
separately and installing them toa temp dir and updating
PATH/LD_LIBRARY_PATH for the time being until a new deb is pushed to
testing
219[01:58:54] <syrius> i can drop the env vars when it's
fixed. this will be fine for now. i just didn't understand why
it needed all those cross-platform deps
220[01:59:30] *** Quits: krabador (~krabador@replaced-ip) (Remote host closed the connection)
221[01:59:51] <ChmEarl> syrius, the x-platform deps is bad
enough, but during the build they might use qemu-debootstrap during
x-compile, which takes long time
222[02:00:01] <syrius> i saw the refs in debian/control.in -
don't see a debian/control.main.in or
debian/templates/control.main.in
223[02:00:10] <syrius> uhg
224[02:00:32] <syrius> so i can just drop lines in that file
though as far as the deps are concerned?
225[02:00:49] <baconicsynergy> I have a question about the
recent speculative execution errors in Intel CPUs. Does a complete
fix require the non-free repository w/ non-free intel microcode, or
is there a fix within the main security updates repository? I really
don't want to enable non-free, but I want to be secure. Does
anyone have any input on this situation?
243[02:08:49] <bennypr0fane> It says the used Hash: SHA512, but
that one is too short to be sha512. Also, if I get the sha256 sum
for this files, it matches
417[05:09:15] *** Quits: Uberius (~Uberius@replaced-ip) (Remote host closed the connection)
418[05:10:40] <cmyers> I am desperately seeking help installing
debian. I've confirmed my thumb drive is correctly
"burned" with the debian 9.5.0 xfce cd1 image (sha1sum
matches, works on other machine), but both text and graphical
install don't work. Graphical install gives a blank screen,
text install gives a weird unreadably tiny screen (like 1cm wide). I
have a 5min video link I can share if it helps. The machine is a bra
419[05:10:45] <cmyers> nd new system76 oryx pro that came
preinstalled with ubuntu (and boots fine).
420[05:11:23] <cmyers> I've filed a support ticket with
system76 and they seem clueless and insist it should work and my usb
drive must be burned wrong or something
456[05:46:03] <io_elephant> I have a Kingston HyperX 3.1 USB. I
am having some weird performance issues with it. Lets say I unplug
it and plug it back in. Here is the kernel log:
replaced-url
457[05:46:09] <io_elephant> reran it and got 800 MB/sec. Is the
kernel loading the correct driver?
512[07:03:46] <iratie> I made a stretch USB installer from the
latest ISO, ran the installer from win10 and got as far as the
restart and OS selection screen where I select 'continue
installation' but when I select that, windows bootloader
interrupts, restarts and says windows failed to load. any ideas?
517[07:05:35] <klys> what software have you installed debian to
usb with?
518[07:06:08] <iratie> just extracted the ISO to the usb and ran
it. when i boot from usb, it starts the installer but then fails to
read the USB midway through the install
519[07:06:39] <klys> can you dd it onto the usb disk? winimage?
520[07:07:04] <iratie> do I need to? it loads the installer from
boot and runs it from windows, I'm not sure that'd help me
521[07:07:24] <klys> wait, you had said you were having a boot
problem.
522[07:07:40] <iratie> yeah. windows boot manager interrupts the
install
558[07:55:29] <m0rd3cai> does anyone know how to locate where a
folder would be being referenced as $SNAP_COMMON. the install
documentation refers to that, not the specific directory
593[08:32:25] <blingrang> Hi, just a quick question that I
couldn't find the answer to on the deb stretch wiki. I see that
systemd is the installed service manager. Does this mean sysvinit
has been completely dropped unike Jessie?
673[10:17:47] <martin-_-_> the hdd is from a nas, and the
connected hdd is from the same nas system, so the partitions and
folder structure is the same
674[10:18:27] <martin-_-_> the intern main hdd has: /dev/hdc1 on
/ and /dev/c/c on /c type ext3
(rw,noatime,acl,user_xattr,usrquota,grpquota) [c]
689[10:24:21] <martin-_-_> jasabella thanks I need to install it
690[10:24:39] <joftr> martin: so /dev/sda1 is mount on
/USB/USB_HDD_2_1 and that mount point now shows a "c"
directory ... isn't that what you are searching for?
691[10:24:58] *** Quits: louvre (~louvre@replaced-ip) (Remote host closed the connection)
692[10:25:18] *** Joins: Immanuel (~Manu@replaced-ip)
703[10:30:16] <joftr> martin: that's new for me, I've
never seen such subdivision of a normal partition like sda1 into
kind of "subpartitions" your fdisk output does not show
any such subpartitions
708[10:33:39] <xingu> martin-_-_: a complete shot in the dark
but how about this for a theory; /dev/c is a lvm vg that lives on
/dev/sda5
709[10:34:11] <martin-_-_> the /dev/hdc1/ with its 2gb looks
like it would be the system Linux partiontn and /dev/hdc3 or
/dev/hdc5 looks like it would be the media space?
710[10:34:27] <martin-_-_> I have to google for lvm vg
711[10:34:56] <martin-_-_> aha yes that could be the case xingu
720[10:41:03] <martin-_-_> vgscan tells me Found volume group
"c" using metadata type lvm2
721[10:41:31] <xingu> martin-_-_: lvm manages the backing store
by itself (it has its own block allocator), so I feel that it's
unlikely you'll recover something on the underlying device
(sda5) that is rational from ext3's perspective without it
722[10:41:47] <xingu> martin-_-_: did /dev/c/ populate?
723[10:42:19] <martin-_-_> yes but I had this already
830[12:13:41] <frikinz> Hi. On Stretch, quite some unicode chars
are wrongly displayed, even in xterm (printf '\xFF\x0F'
displays a question mark). Any idea what to check or what the
problem could be?
831[12:14:22] <frikinz> Reminds me of unicode headaches, but
that was 10 years ago :)
849[12:27:09] <frikinz> Locales looks fine so I guess font yes,
but I can't remember changing anything. Fonts listed in xorg
log are ok. I use kde as DE, I could try xfce
850[12:27:33] <frikinz> let me change to en_US.UTF-8 for a try
851[12:27:57] *** Quits: sulit (~mutou@replaced-ip) (Remote host closed the connection)
852[12:28:03] <ksk> nah, its more about do your fonts support
this kind of sign afaik. and as I said, I can neither display it
(using mosh on a remote session)
853[12:28:46] <ksk> like for example, even if the UTF-8 locle
does support asian type of letters, you also need a font that
supports it.
854[12:29:01] <jasabella> yeah looks like font issues
855[12:29:06] <ksk> changing the DE would be a little overkill.
878[12:35:47] *** Quits: petan (~g@replaced-ip) (Remote host closed the connection)
879[12:36:20] <rocketmagnet> i've troubles with
notify-send, when i type in notify-send "Welcome" i get no
output, in my taskbar i've the widget to display messages but
no messages appear, i also tried to reinstall notify-send and the
other packages but nothing happens :X
880[12:36:54] *** Quits: sulit (~mutou@replaced-ip) (Remote host closed the connection)
882[12:38:56] <BCMM> rocketmagnet: it would probably help to
know which DE or taskbar you're using
883[12:39:20] <rocketmagnet> i'm using mate with latest
stable debian
884[12:40:02] <rocketmagnet> when i go to the settings where i
can setup the notifications and also see an example how it's
working, but notify-send does nothing
893[12:44:57] <rocketmagnet> it's a shell script, is it
enouth when i trypen in the path + the name of the shell script with
the .sh at the end or without ?
895[12:49:06] <frikinz> jasabella: just tried fonts-roboto, the
same. And I haven't found a monospace variant in that stretch
package. Do you confirm using that font successfully on debian?
896[12:49:41] <jasabella> yes, but i'm using en_GB.UTF-8
locale
1005[15:14:45] *** Quits: coolmadmax (~coolmadma@replaced-ip) (Remote host closed the connection)
1006[15:17:25] <joftr> bipul: ok, it seems like it should, but
does host1 have those 2 .1 IPs assigned (to these bridges) - check
w/ "ip addr" on host1 ......... and do you have any
firewalling going on - on host1 or the nodes? ........ I think the
fact that the error is "destination port unreachable"
suggest that routing is actually working, but something different is
messing with the packets
1019[15:45:16] <ksk> Im using Xfce, can I switch alt-tab to
something else? (as its running on windows in a virtualbox VM)
1020[15:46:58] *** Quits: dina16 (~d-ina@replaced-ip) (Remote host closed the connection)
1021[15:47:30] <joftr> bipol: so now we at least know why you are
seeing "dest port unreachable", for some reason the
packets from 123.13 hit that filter/FORWARD fallback rule, which es
rejecting everything w/ exactly that icmp message
1025[15:52:26] <joftr> bipol: ah, right ... should have seen that
right away: the order of the rules is important, that's why
from 122.x to 123.y it is working, but not the other way around
1026[15:53:42] <bipul> My name is bipul :(
1027[15:53:55] <joftr> bipul: the packets from 123.x are going
past the first 3 filter/FORWARD rules and unfortunately hit the 4.
rule, which result in the ICMP message we get
1028[15:54:02] <bipul> okay :)
1029[15:54:41] <joftr> bipul: sorry (I should use another IRC
client, which automatically attaches those name prefixes, I guess)
1030[15:54:42] <bipul> could you hint any link or documentation
to learn more about this iptables rules/
1034[15:55:40] <joftr> bipul: not from the top of my head, but
why aren't you just change the order in your firewall script?
what do you use to setup these rules?
1035[15:56:43] <bipul> I have not setup these rule :( infect i
have setup this node to learn IPTABLES chain rules.
1036[15:56:59] <joftr> bipul: I guess these rules are generated
by a simple script, doesn't look like generated from something
like shorewall
1037[15:57:16] <bipul> Yes, it might be generated from libvirt
1038[15:57:34] <bipul> I have not setup or configured any
iptables rules.
1040[16:00:13] <joftr> bipul: unfortunately I'm not familiar
with libvirt w/ iptables rule generation ... as a very first step
you could change the order of the rules manually
1045[16:01:36] <joftr> bipul: node3 is news, until now we were
talking about host1, node1 and node2
1046[16:02:24] <bipul> I'm sorry. Yes.
1047[16:02:56] <bipul> I would like to make node1 act as a router
when i send ICMP request to node2.
1048[16:03:17] <joftr> bipul: currently, I'd say host1 is
your router, which is not only routing but also NATing .... so any
123.x host will think packets from 122.y are coming from 123.1 (=
host1)
1049[16:03:34] <bipul> exactly.
1050[16:05:33] <joftr> bipul: but the order of the rules in
filter/FORWARD is "incorrect", for one direction ....
I'd put rules 6 and 7 right after 1 und 2
1073[16:15:21] <joftr> bipul: kind of the rule of thumb: first
have the "specific rules", which allow stuff, then the
less specific which still ACCEPT (like the -i virbr0 -o virbr0) and
finally the disallow everything else stuff (regarding virbr*, like
the REJECT)
1074[16:16:15] <bipul> Oh okay got it. let me note it down.
1096[16:45:58] <candyban> I am trying to preseed my strech
install, but I seem unable to get past the keyboard selection. Does
anyone have a working preseed file which will not prompt for the
keyboard layout?
1133[17:21:44] *** Quits: Achylles (~Achylles@replaced-ip) (Remote host closed the connection)
1134[17:21:48] <candyban> mspe: I have
replaced-url
1135[17:23:05] <candyban> mspe: I have "d-i
keyboard-configuration/xkb-keymap select us" ... or should I
use "keyboard-configuration keyboard-configuration/xkb-keymap
select us" instead?
1149[17:34:43] <cybercrypto> NCS_One: Yes. why? having any
issues?
1150[17:35:10] <cybercrypto> correct. dist-upgrade should do :)
1151[17:35:22] <annadane> for point releases upgrades are simple.
for upgrades between releases (for example, 9 -> 10) there are
release notes and such
1152[17:36:30] <annadane> also, if you have internet access you
do not have to wait (and should not, due to security updates) for
the next point release to upgrade your system
1153[17:37:22] <NCS_One> cybercrypto: Debian has Linux 3.16 that
was released August 3, 2014
1154[17:37:34] <annadane> then you might be on jessie
1155[17:37:47] <NCS_One> cybercrypto: also Firefox 45 that also
is very old and with a lot of issues
1156[17:38:01] <annadane> cat /etc/debian_version
1157[17:38:15] <mspe> candyban: I did my preseed file by
selecting lines from a log after a manual install, so I think
it's the right owner, yes
1183[17:47:38] <annadane> basically with stable you can check for
updates every week or 2 weeks or whatever, the most important stuff
is on the security mailing list
1204[18:00:58] <annadane> NCS_One, apparently if you want to use
the proprietary nvidia driver you'd need to use the
nvidia-legacy-340xx-driver package or if you want to use the free
implementation then use xserver-xorg-video-nouveau as it says
1205[18:01:23] <annadane> there are reasons for and against using
both
1206[18:01:32] <annadane> what's your preferred desktop
environment/window manager?
1207[18:01:55] <Something1> NCS_One, you install the package and
configure it as per the wiki link above. You can specify security
only (default) and it will update those automatically. You can let
it send you e-mail about it too, if I recall correctly.
1208[18:02:00] <NCS_One> xfce
1209[18:02:27] <NCS_One> annadane: ^
1210[18:02:39] <annadane> then nouveau is probably fine. what you
can do prior to the upgrade is (as root) update-alternatives
--config glx and select the mesa option
1211[18:02:46] <annadane> (which also requires a reboot)
1212[18:03:11] *** Quits: rtmataeu34 (~energyd@replaced-ip) (Remote host closed the connection)
1213[18:03:12] <annadane> basically the situation is that nouveau
is buggy because nvidia doesn't work with them, but nvidia is
buggy in other ways
1214[18:03:40] <annadane> i use nouveau with xfce and don't
have crashes. when i used KDE however, it would always break
1215[18:04:23] <NCS_One> annadane: but I already doing the
upgrade
1217[18:05:27] <annadane> i guess doing it after the upgrade
finishes it also fine, and you can have that reboot suffice for the
upgrade and the switch to nouveau
1226[18:11:16] <annadane> i'm not sure what to do here. my
intuitive feeling is to switch to mesa, reboot, and then continue
with the rest of the upgrade, but o
1227[18:11:21] <annadane> but i'm not sure if that's
wise or not*
1235[18:18:08] <at0m> i guess dpkg didnt come that far then..
1236[18:18:46] <at0m> i most often apt -f install when install
fails, and personally that hasn't gotten me into trouble yet.
but if annadane doesn't know, then who am i
1237[18:19:05] <annadane> please. i guess half the time
1258[18:25:21] <annadane> ok, here, i have an idea
1259[18:25:27] <at0m> "unsupported devices are present in
the machine"
1260[18:25:29] <Gerowen> Any particular reason the Debian
installer would fail because "No common CDROM drive was
found" even though I'm installing from a USB stick?
1261[18:25:42] <at0m> NCS_One: does nvidia-detect tell you these
are the right drivers?
1262[18:25:55] <annadane> at0m, it says their card isn't
supported by the nvidia-driver package
1268[18:27:46] <at0m> annadane: i think NCS_One has to install
nvidia-legacy-340xx-driver, not nvidia-legacy-384
1269[18:28:32] <annadane> right
1270[18:28:33] <at0m> NCS_One: i would purge all nvidia packages,
then install nvidia-legacy-340xx-driver
1271[18:28:48] <annadane> well
1272[18:28:52] <annadane> hang on though
1273[18:29:02] *** sekizo is now known as schizo
1274[18:29:17] <annadane> if they purge all nvidia packages and
are on nvidia currently then it will muck up their GUI and unless
they're semi-proficient with irssi won't be able to reach
us for support
1275[18:29:21] <NCS_One> at0m:nvidia-legacy-340xx-driver or
xserver-xorg-video-nouveau right?
1276[18:29:32] <annadane> at the very least switch to nouveau
before doing that
1277[18:29:55] <annadane> so we have to get the sequence right
here as it involves multiple steps
1292[18:33:59] <annadane> 1) update-alternatives --config glx,
switch to the option that says mesa - if not available, apt install
xserver-xorg-video-nouveau; reboot 2) purge nvidia packages (it
should be apt purge nvidia. - note the .) 3) apt install
nvidia-legacy-340xx-driver
1293[18:34:04] <NCS_One> for me either driver is ok since it
works ok
1321[18:46:42] <LikWidChz> Well I'll throw my question out
there and see what I can snag.
1322[18:47:03] <LikWidChz> So trying to install samba on kali,
installed it from apt-get, hosed the config file and thought what
the hell ill just remove it and install it again, however it kept my
hosed config, ive rm'ed the directory and when installing it
agian it doesn't create it. Could someone school me a tad on
what the issue is? hopefully its ovbious to you folk
1326[18:47:55] <BCMM> oh sorry didn't see that was said
already
1327[18:48:43] <annadane> LikWidChz, if kali is anything like
debian you can apt purge to get rid of config files unless it's
saved as a dot file in your home directory, but again, really,
it's better to ask kali support
1328[18:49:06] <LikWidChz> so something like apt-get purge samba?
1329[18:49:11] <annadane> probably
1330[18:49:27] <LikWidChz> worth a shot, how is purge different
then remove?
1331[18:49:53] <annadane> as i said, gets rid of config files
1332[18:50:19] <LikWidChz> OOH ooh ohh gotcha gotcha, you did say
that.
1336[18:51:41] <NCS_One> and it didn't went well :(
1337[18:52:04] <annadane> dot files in your home directory can be
viewed by ls -a and they can also have configuration stuff which you
can selectively delete to have programs erase their data/preferences
and such... carefully. but you knew that because people running kali
should know what they're doing
1338[18:52:24] <annadane> !whynotpentest
1339[18:52:30] <annadane> uh
1340[18:52:32] <annadane> !whynotpentesting
1341[18:52:32] <dpkg> For some reasons why Kali/Parrot/Other
based-on-Debian penetration testing distributions shouldn't be
used as a home desktop, see
replaced-url
1343[18:52:47] *** Quits: vbutnaru (~vbutnaru@replaced-ip) (Remote host closed the connection)
1344[18:52:54] <NCS_One> can I change to mesa-diverted or should
I install the nouveau?
1345[18:53:01] <annadane> mesa-diverted is nouveau
1346[18:53:10] <NCS_One> ok
1347[18:53:20] <annadane> hence why i said, "switch to the
option that says mesa"
1348[18:53:50] <NCS_One> I'm on tty and getting the same
errors
1349[18:53:52] <annadane> LikWidChz, and apt purge - if kali is
anything like debian - will not remove dot files
1350[18:53:58] <LikWidChz> Yeah kali is on a laptop I have for
screwing around. works good, wanted samba to easily files from
windows -- nix without dicking around
1351[18:54:20] <LikWidChz> annadane, was looking in /etc for a
samba or sa... something .conf file and didnt see anything
1352[18:54:23] <annadane> NCS_One, yeah well first of all get on
nouveau, select the mesa option and reboot and then we can continue
working on it
1353[18:54:37] <NCS_One> I did it
1354[18:54:46] <annadane> okay
1355[18:54:57] <annadane> did you install the legacy driver from
nvidia?
1356[18:55:07] <NCS_One> no
1357[18:55:40] <NCS_One> I'm on tty with X
1358[18:55:49] <annadane> i suppose that'd be the next
logical thing to try
1359[18:55:50] <NCS_One> s/with/without
1360[18:56:08] <annadane> LikWidChz, fair enough, it's just
that you wouldn't believe the amount of "hi i'm on
kali linux teach me how to be a hacker"
1375[18:58:24] <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.
1376[18:58:29] <annadane> not sure if it's specific to sid,
though
1377[18:58:46] <rozie> probably is
1378[18:58:59] <LikWidChz> annadane, I 100% completely
understand. Ive used the distros before and futzed around with them,
but I know in a linux channel asking people for that kind of advice
isnt exactly a great idea. I actually learned linux based off
wanting to crack WEP back in 2005 so its funny now i'm learning
some of the basics which is always funny. Start with compiling a
kernel and end with how to install an application HAHA.
1379[18:59:07] <rozie> and yes, I tried to join and it's
invite only
1380[18:59:16] <annadane> see the factoid - OFTC
1381[19:00:26] <rozie> !oftc
1382[19:00:27] <dpkg> OFTC is the Open and Free Technology
Community, a support/collaboration service. They have an IRC
network: irc.oftc.net. You may be connected to OFTC's network.
replaced-url
1383[19:00:40] *** Quits: LucaTM (~LucaTM@replaced-ip) (Quit: To infinity and beyond...)
1384[19:00:46] <annadane> NCS_One, also i'm happy you
managed to still connect to irc, i should have mentioned how to
before you did all this in case something went wrong, that's my
fault
1385[19:01:01] <NCS_One> annadane: can't install it
1386[19:01:25] <rozie> BTW any idea why I'm banned on
#debian on IRCnet?
1387[19:01:37] <annadane> no idea. tor, vpn?
1388[19:01:44] <rozie> wasn't even active there
1389[19:02:03] <NCS_One> annadane: is there a way I can paste to
debian.paste.org from tty?
1390[19:02:08] <rozie> no, not using Tor nor VPN on this IP
1391[19:02:08] <LikWidChz> annadane, I ran apt-get purge samba,
apt purge samba "didn't find anything" rm -r the
/etc/samba directory, then ran apt-get install samba, it installs
the app but then the /etc/samba directory doesn't exist.. So
any ideas? or is my method -100 points for me..
1402[19:06:22] <annadane> if apt purge samba didn't find
anything then maybe the package isn't called samba
1403[19:06:27] <LikWidChz> this might sound dumb, wonder if I
just created that directory after I purged the install then
installed it again? maybe for whatever dumb reason the install
doesn't create a directory..
1404[19:06:29] <annadane> look through your apt logs to see what
the package name actually is
1405[19:06:58] <annadane> i have no idea what kali does but
it's possible i suppose to have empty/non existent configs
1406[19:07:03] <LikWidChz> its possible ive been hosing and
unhosing the incorrect thing.
1407[19:07:29] <annadane> maybe. anyway, as said, there's
two places to look for configs, /etc and dot files, for the dot
files you can ls -a | grep -i samba
1409[19:08:23] <annadane> LikWidChz, what's happening now
though is there's someone with a debian question who needs help
and it's getting scrolled off screen so at this point i'd
just take it to kali support if my solutions haven't worked
1414[19:11:12] <annadane> NCS_One, at the very least, what does
apt --fix-broken install say? pipe it to termbin like you just did
but don't automatically press Y yet
1426[19:19:39] <at0m> i was trying to get you away from 384,
which doesnt support your card.. did you remove all nvidia packages
after update-glx switching to mesa?
1427[19:19:55] <at0m> (maybe it's different for
nvidia-alternative, idk)
1428[19:20:11] <at0m> (maybe driver support is different for
nvidia-alternative, idk)
1429[19:20:13] <NCS_One> at0m: you mean apt purge nvidia.?
1430[19:20:15] <annadane> yeah, i'd definitely just remove
all the nvidia stuff after having switched to mesa/nouveau
1431[19:20:21] <at0m> NCS_One: yup
1432[19:20:39] <NCS_One> gives error
1433[19:20:39] <annadane> (i'm surprised switchint to
nouveau forced you into a tty, though)
1434[19:20:46] *** Quits: mete (~mete@replaced-ip) (*.net *.split)
1509[19:38:04] <at0m> jhutchins: NCS_One doesn't need to if
he only formats sda1 and leaves sda6 untouched
1510[19:38:26] <at0m> then assign sda6 to /home in fstab even
after install and all is swell
1511[19:38:35] <jhutchins> NCS_One: Yes, you can do an install
over an existing one, and as long as you skip formatting the
partitions you won't overwrite /home, but you want a backup
anyway, because "oops".
1534[19:52:09] <inquisitiv3> Is there any numbers of how many
times there's package breakages in testing and sid? I'm
considering running one of them on my machine, and want to make an
informed decision.
1590[19:57:35] <annadane> the general answer is, depends on what
your machine is for, and how much time you want to spend fixing it
when it breaks. anecdotally, people have had few problems for years
but you can of course run into issues. if you need the computer to
work then the issues can affect your work.
replaced-url
1591[19:57:45] <annadane> and it should not be used on servers
1593[19:58:06] <annadane> there are ways to get newer packages in
debian stable, like backports, compiling your own packages,
downloading tarballs, etc
1598[20:00:03] <annadane> the current Debian Project Leader
recommends unstable over testing for users who don't want to
use stable which is, i think, a testament to its solidity
1601[20:00:48] <annadane> if you have a rescue medium and have
backups of your files and have access to another computer to
troubleshoot then it's probably fine
1625[20:03:55] <inquisitiv3> annadane: Thanks! I've read the
FAQ entry earlier, but I haven't seen any actual data on it. Do
you have a link to there Chris recommends unstable? I know he talked
about it on an episode of Destination Linux (podcast), but I
don't think he said unstable > testing.
1626[20:04:41] <annadane> i can find a reddit link or something
1628[20:05:44] <inquisitiv3> ntd: I think it depends on the type
of the release actually. I've read in various threads on Reddit
(no official source) that Ubuntu LTS is based on testing, and the
interim releases is based on unstable. A small detail.
1632[20:08:42] <kirk781> Shouldn't LTS be based on stable?
1633[20:09:11] <annadane> not the dpl but
replaced-url
1634[20:10:03] <annadane> anyway, one reason to potentially
prefer sid over testing: 1) packages can get removed from testing if
they don't meet certain criteria 2) things get immediately
fixed in sid and take longer in testing 3) package transitions can
have delays in testing
1635[20:10:36] <annadane> i guess that's 3 reasons
1636[20:10:50] <ntd> trusty and xenial were both ahead of deb
stable when released
1644[20:13:12] <annadane> i guess "work" is relative
because yes there are only releases once every 2 years and there are
packages uploaded with bug fixes that stable users don't get
1645[20:13:23] <annadane> lack of significant crashes, anyway
1646[20:13:32] <annadane> though as i said anecdotally people
have run it just fine
1649[20:14:51] <inquisitiv3> annadane: Yeah. It's just that
I subscribe to the belief that packages should get *all* bugfixes.
And if the alternative is unstable I'll run that instead.
1650[20:14:54] <annadane> i mostly run stable as i currently only
have the one computer; i'd probably run sid otherwise with the
other one running stable as backup
1651[20:15:57] <inquisitiv3> It would be nice if Debian had
interim releases like Ubuntu (6 or 9 months) that made newer
packages available, while still having stability.
1652[20:16:43] <annadane> more people running unstable/testing =
more people filing bugs so i'm not entirely on board with the
debian community's shunning of users who want not-stable
1653[20:17:16] <annadane> unless it's "i want x latest
package", then you *can* get it usually in stable and you take
risk upon yourself by updating to something which at minimum
requires more maintenance
1654[20:17:29] *** Quits: krabador (~krabador@replaced-ip) (Remote host closed the connection)
1662[20:20:11] <okamis> Hello, updated debian and also to vlc
version 3.*. Vlc has stopped working. I get audio but no window at
all to interact with
1663[20:20:37] <annadane> okamis, stable?
1664[20:21:59] <inquisitiv3> annadane: Regularly it seems that
Debian folks in general (at least them that hangs out @ Reddit) is a
little to conservative. Noting wrong wanting a stable OS, but
there's limits.
1681[20:26:45] <annadane> inquisitiv3, one of the things which
can turn people off is if they're stuck in an infinite boot
loop or can't access x, it may be rare but in theory it can
happen and you need to be prepared for it
1684[20:28:21] *** Quits: okamis (~okamis_@replaced-ip) (Remote host closed the connection)
1685[20:28:57] <jhutchins> inquisitiv3: New people have always
complained about the fact that newer packages exist but take a long
time to work the way into stable distributions. RedHat's even
slower.
1686[20:29:44] <jhutchins> inquisitiv3: It's funny, but the
same people don't seem to pay any attention to Windows software
versions. Maybe because they're not public.
1697[20:38:30] <cybrNaut> some QA needed.. someone should
probably remove the "agenda.app" package- it's
defunct and undocumented =>
replaced-url
1698[20:39:23] <Habbie> cybrNaut, might be good to file that in
the debian bug tracker
1699[20:40:40] <Habbie> cybrNaut, wait
1700[20:40:49] <Habbie> cybrNaut, on your url: Agenda is a
calender application. This application is defunct. It's
homepage and download link are down. You can use SimpleAgenda.app
instead.
1701[20:41:06] <Habbie> cybrNaut, in the debian agenda.app
changelog at
replaced-url
1709[20:42:20] <inquisitiv3> jhutchins: What do you mean
regarding the Windows software versions? The difference between
Debian and Windows is that in Debian everything is frozen, while in
Windows its just the OS that is frozen. And Windows have a much
shorter release span since 10.
1710[20:43:04] <cybrNaut> Habbie: does that mean the version
claimed to be working is in the repo?
1823[22:09:49] <dpkg> If you don't know enough about Debian
to do this yourself, you won't know enough about Debian to fix
the problems you *will* encounter. Please spend some months or years
learning about Debian first. Sorry.
1827[22:09:55] <dpkg> Habbie: have you tried
replaced-url
1828[22:09:56] <annadane> i'm stupid.
1829[22:09:58] <NCS_One> hi
1830[22:10:12] <annadane> so then you just need the one line
1831[22:10:15] <annadane> sorry
1832[22:10:16] <Habbie> annadane, so i'd say the
instructions just map, but 'sid/main amd64' looks
different to me
1833[22:10:18] <NCS_One> annadane: I was able to upgrade it
1834[22:10:24] <annadane> good news
1835[22:10:25] <Habbie> then again my sid just says 'sid
main'
1836[22:10:52] <NCS_One> annadane: what commands should I use to
check if all went ok?
1837[22:11:06] <annadane> dina16, btw, what did you mean by
server? i hope you're not running sid on a server
1838[22:11:35] <dina16> i fix it all okayy
1839[22:11:38] <dina16> thnk's
1840[22:11:42] <Habbie> dina16, good!
1841[22:12:22] <Habbie> so, just for me to get things clear, sid
is forever unstable, but testing cycles through names? (and those
names become 'stable' when testing cycles to the next
name?)
1842[22:12:24] <annadane> NCS_One, er... i guess you ought to be
able to check if it's installed with dpkg -l
nvidia-340xx-legacy-driver or whatever the package is called
1843[22:12:33] <annadane> Habbie, yeah
1844[22:12:39] <Habbie> thanks
1845[22:12:59] <Habbie> do stable and testing cycle at the exact
same time or is there some delay?
1846[22:13:11] <Habbie> (this is useless curiosity now)
1847[22:13:14] <annadane> anyway, IIRC you had like 250+ packages
that needed to be upgraded, right? you can check if they're all
updated via apt update
1848[22:13:46] <annadane> well, stable releases once every 2
years, testing gets packages from sid within 2-10 days
1849[22:14:01] <Habbie> that i get
1850[22:14:02] *** debhelper sets mode: +l 1073
1851[22:14:06] <Habbie> so every 2 years, stable 'changes
names'
1852[22:14:10] <Habbie> does testing change name on that same
day?
1853[22:14:13] <annadane> yep
1854[22:14:17] <Habbie> thanks
1855[22:14:23] <dina16> i used sid like Rolling release
1856[22:14:25] <annadane> buster is testing now but then when
buster becomes stable testing will be bullseye
1857[22:14:28] <Habbie> do we know what comes after buster?
1858[22:14:30] <Habbie> ah
1859[22:15:41] <annadane> this is also why we recommend people
not to put "stable" in their sources list and to only use
release names like stretch etc because when buster is the new stable
...
1862[22:16:16] <Habbie> how about putting 'testing' in
sources.list? that's like unstable with a small bit of QA?
1863[22:16:26] <Habbie> which, i guess, slows down a lot near the
two year cycle
1864[22:16:37] <annadane> testing will always track testing, so
if that's what you want then fine, as long as you're aware
of that
1865[22:16:54] <Habbie> i feel testing would be a pretty decent
development VM
1866[22:17:02] <annadane> sid will always be unstable but stable
and testing will change based on releases
1867[22:17:36] <Habbie> i gave fedora rawhide a shot as a
development VM for a while, but every time i left it alone for a
month it would need so much time to run updates
1868[22:17:43] <Habbie> i imagine testing and sid would be
similar
1869[22:18:07] <NCS_One> annadane: I'm using the nouveau
driver
1870[22:18:23] <annadane> ok, so it's up to you to decide
whether you want nouveau or nvidia
1871[22:18:43] <annadane> if you want nvidia then
update-alternatives --config glx and switch to nvidia
1872[22:18:44] <annadane> and reboot
1873[22:18:51] <NCS_One> nouveau is working ok, I'm going to
keep it
1874[22:19:04] <annadane> yeah, until it crashes - i literally
just had one earlier with xfce
1875[22:19:29] <annadane> but it probably also depends on your
computer
1876[22:20:18] <annadane> anyway, weren't you the one with
many packages to upgrade? you should probably upgrade the system
1877[22:20:22] <NCS_One> annadane: are there any commands I can
use to check if the upgrade was a full success?
1878[22:20:36] <annadane> yeah, type apt update and see if
there's anything it still needs to install
1879[22:20:37] <NCS_One> I already upgraded it
1880[22:21:41] <annadane> Habbie, note also "bit of QA"
also can mean significant delays
1881[22:21:43] <NCS_One> annadane: "All packages are up to
date." :)
1882[22:21:46] <Habbie> annadane, yes
1883[22:21:51] <annadane> there are distros with actual rolling
releases and debian isn't... exactly one of them
1884[22:22:05] <Habbie> annadane, i always wonder how unstable
compares to distros that advertise as rolling
1886[22:22:40] <annadane> NCS_One, and i hope you rebooted
afterwards, due to (minimum) kernel upgrades
1887[22:22:41] <Habbie> 3.1.7 is a good read
1888[22:22:47] <annadane> now, keep on top of your updates
1889[22:23:14] <annadane> oh sorry yeah i guess i meant to link
3.1.7
1890[22:23:17] <Habbie> np
1891[22:23:19] <Habbie> i read both :)
1892[22:23:52] <annadane> so, yeah. tempting to think of testing
as being "stable is too old, unstable is crashy, i'll use
testing! perfect solution!" but that's not really the case
1893[22:23:57] <NCS_One> annadane: thanks for you help
1894[22:24:12] <annadane> some people even recommend not using
testing on an internet-facing machine at all - it also gets the
slowest security updates
1895[22:24:17] <annadane> different opinions from people i guess
1896[22:24:21] <Habbie> ah yes
1897[22:24:31] <Habbie> as unstable will get updates 'pretty
darn soon' which includes security fixes
1898[22:24:45] <Habbie> but the trickle down to testing might be
stuck due to RC tickets
1899[22:24:54] <annadane> (unstable isn't GENERALLY crashy
but yes it has happened and you need to be prepared)
1900[22:26:17] <petn-randall> I wouldn't recommend anything
but stable on machines serving things to the internet. Even sid can
get security updates late if the newer, bug-free version fails to
build from source.
1902[22:26:55] <Habbie> petn-randall, that makes sense
1903[22:26:57] <annadane> right. there *are* distros which do
actually specialize in rolling, i guess the two most known being
arch and opensuse tumbleweed
1904[22:27:04] <annadane> also a "fork" of sid called
siduction
1905[22:27:21] <annadane> oh, nixos too i suppose
1906[22:27:39] <Habbie> annadane, forgive my continued ignorance
while you keep answering awesomely - i always thought opensuse
tumbleweed was just like fedora rawhide or debian unstable, what am
i missing?
1907[22:27:52] <annadane> oh i can't really answer questions
on that
1908[22:27:53] <Habbie> well nixos is special in that you get
some control over versions, right?
1909[22:27:54] <annadane> i have no idea
1910[22:27:56] <Habbie> annadane, alright
1911[22:28:05] <annadane> yeah you can install multiple versions
of packages at once
1912[22:28:13] <Habbie> i do know about arch, i have a few
coworkers who swear by it and will occasionally swear at it
1913[22:28:33] <Habbie> but thanks to them (and thanks to me
being on macOS with brew which also is 'rolling'), things
happening in debian rarely surprise us ;)
1914[22:28:41] <annadane> stable in my mind is the perfect mix of
everything, you really don't need new packages except for a few
exceptions and you can get those through various means
1915[22:28:53] <Habbie> yes, full agree on that
1916[22:28:57] <annadane> i should mention, nvidia makes rolling
anything significantly harder. screw them.
1972[23:25:48] <yahal> hi I've got a little problem with
ufw, after installing programms from jessie-backports and disable
the ufw, I cannot enable it without getting a python trackback
failure
replaced-url
2024[23:52:51] <yahal> petn-randall: the pid the traceback error
revealed is not activ right now, could it bit, that there is a
pid-file which causing this error?
2025[23:54:07] <petn-randall> yahal: Could be, but it
shouldn't produce a traceback in that case.
2027[23:54:31] <petn-randall> yahal: Btw, jessie is only
supported through jessie-lts, you might want to consider upgrading
at some point.
2028[23:55:29] <annadane> rydare, so uh... my best guess is
either remove the packages or try downgrading to testing but not all
functionality is even included in experimental, i might try removing
and then installing from testing
2029[23:55:37] <annadane> best guess, someone may know more
2030[23:55:46] <yahal> I thinking about to try a system upgrade
to stretch, but I am not sure it is supported by my hoster. They
allow only jessie as latest debian to be preinstalled
2031[23:56:08] <rydare> annadane: thank you for your help. i just
need to find out which packages are from the repository
2032[23:56:28] <rydare> this would really be more easy if i could
get an xsession going lol