11[00:13:15] <cybrNaut> right.. and in fact i was thinking
maybe i don't need to set a partition as active/bootable if
it's a GPT table
12[00:13:41] <somiaj> that flag did very little even in mbr
days for linux
13[00:14:23] <somiaj> basically that flag was the partition the
mbr would try to boot second if there wasn't a boot loader on
the main mbr, which there usually was (most install grub on /dev/sda
vs /dev/sda3 for instance)
16[00:16:35] <ComCat> IMPORTANT NOTICE: Hi Ops. Please update
the topic to point to the new server (libera.chat). Thank you.
Current situation: crown prince of korea and co-founder of MtGox
(bitcoin exit-scam)
replaced-url
17[00:16:40] <ComCat> moved to Libera.chat. Sponsors of
Freenode are beginning to pull out, it is unknown for how long the
freenode will be stable. Most channels, projects and users are now
migrating to Libera. Libera is a registered non-profit organization
with mitigations against this disaster happening ever again. Leaked
chats showing disgusting intents and former devs that have worked
with Mr.Lee before says that he's disorganized, dishonest and
that they wish to
18[00:16:45] <ComCat> never work with him again (he managed to
scam freenode out of our hands and he lied to the staffers). All
officiall freenode channels are now currently under complete
censorship.
replaced-url
19[00:16:50] <ComCat> situation if the topic is set already.
Spread this message too.
28[00:20:18] <cybrNaut> would be nice if they would say *where*
Libera.chat is registered. freenode is in the UK, under UK law.
what's the new jurisdiction
29[00:20:55] <petn-randall> Duh, Liberia.
30[00:21:01] <petn-randall> ;)
31[00:21:20] <sussudio> i think i saw sweden mentioned.
32[00:21:22] <jelly> I don't like when people mix truths
and lies to promote an agenda
34[00:21:43] <jelly> even if I might agree with 90% that idiot
wrote
35[00:21:46] <sussudio> jelly: sounds like politics.
36[00:22:08] <somiaj> Although intersting, just a reminder that
debian's irc network is irc.oftc.net = irc.debian.org, and if
you are intersted about things you should join #libera on their
network to talk about such things.
62[00:50:31] <ryouma> i just got "grub-install: error:
attempt to install to encrypted disk without cryptodisk enabled. Set
`GRUB_ENABLE_CRYPTODISK=y' in file
`/etc/default/grub'." what can cause this error? i have a
plaintext boot partition and an encrypted root. why does grub think
the DISK is encrypted?
88[01:07:00] <ryouma> jelly: thanks for confirming. i found a
thing on the web that hinted at that. i am trying to move root to
another partition on the same drive. for some reason /boot is not
mounted, so grub must not have found it.
92[01:10:54] <ryouma> i will try booting to make sure
everything works, then mount newroot and newroot/boot, populate the
latter, and then do the dpkg-reconfigure grub-pc again.
93[01:11:15] <ryouma> or maybe i don't need the rebooting
step,hmm
94[01:11:25] *** Quits: Numero-6 (~Numero-6@replaced-ip) (Quit: << - Qui etes vous ? - Je suis le nouveau numero 2 -
Qui est le numero 1 ? - Vous etes le numero 6 - Je ne suis pas un
numero ! Je suis un homme libre!! >>)
98[01:12:50] <terr> Somiaj, my admin set up multiboot for me.
350 GB drive. About 100 in partition 1, windows 7 pro. About 100 in
partition 2, windows 10 pro. These will be C & D both NTFS.
About 100 partition 3 for ext4 (probably) and Debian. Looks pretty
good to me. I can use dd and drop this on a terabyte drive and the
rest is an extended partition for anything
100[01:13:50] <somiaj> I don't do much with disk imaging
(mostly cause I don't mess with windows that much). I
peronsally prefer to just crate a new partition table/file system
and then copy things, though unsure how well this works with NTFS
101[01:14:24] <somiaj> But if you dd the smaller drive onto the
bigger drive, you'll have to extend the partition table, then
you can use the reamining space as a new partition (or extend
partitions)
117[01:25:18] <terr> Over the weekend I should be able to get to
a Debian install. Finally. I was ready to do this almost a year ago
and had to drop the project.
128[01:38:52] <solrize> i need to temporarily create a single
directory with about 500,000 files in it. with ext4 this is quite
painful, does anyone know if xfs can handle it? e.g. by using a
hashed dir structure instead of a linear one. it would be on an SSD
based virtual disk if that matters
155[02:09:19] <solrize> hmm actually seems kind of cpu bound in
the unarchiver so maybe it is ok after all
156[02:09:37] <somiaj> are these files larege or small?
157[02:09:42] <solrize> a few MB each
158[02:09:45] <solrize> varies
159[02:09:55] <solrize> about 80GB total uncompressed
160[02:10:03] <solrize> from maybe 100K to 10M depending
161[02:10:16] <somiaj> is xfs working better, or could have just
been just due to the large number of small things, the cpu has been
the bottle neck even with ext4?
163[02:10:58] <solrize> i think it was slower with ext4 but not
sure, i gave up pretty quickly and did some hack to move files out
of the target directory while they were being created, and that
helped a lot but was a nuisance
164[02:11:03] <PMT> What are the half a million files, out of
curiosity?
165[02:11:14] <dondelelcaro> solrize: ah; well, there's the
theoretical performance limit (something like for a in seq 1 500000;
touch $a; done; ), and then there's the actual performance that
you'll get while moving the files
166[02:11:29] <dondelelcaro> but dir_index fixed most of the bad
performance issues
167[02:11:29] <solrize> moving files = just overwriting
directory entries
168[02:11:34] <solrize> what is dir_index?
169[02:11:41] <PMT> an ext[34] feature.
170[02:11:44] <dondelelcaro> !dir_index
171[02:12:13] <dondelelcaro> hrm; no factoid
172[02:12:13] <solrize> the 500k files are a fanfiction archive,
replaced-url
173[02:12:13] <solrize> oh is there something i have to do to
enable dir_index?
174[02:12:13] <PMT> oh that archive
175[02:12:13] <dondelelcaro> nope; it's enabled by default
176[02:12:15] <solrize> nice
177[02:12:31] <dondelelcaro> [at least, assuming this is ext4
and relatively recently created]
178[02:12:40] <solrize> pmt yeah i convered it all to text files
and am trying to deduplicate it and upload to archive.org
179[02:12:40] <dondelelcaro> (like past 10 years)
180[02:12:47] <solrize> yeah this is on hetzner cloud
232[03:00:22] <solrize> it takes about 1 minute to move 10k
files from the old directory to a new one
233[03:00:42] <solrize> so i have a script that monitors the
size of the unarchiving directory and when it's about 11000, it
makes a new dir and moves 10k out
234[03:01:00] <solrize> it's all pretty lame but this
really speeds it up a lot
235[03:01:35] <solrize> i was kind of tempted to spin up a 128gb
ram machine and do this on a ramdisk but $$$
236[03:02:19] <solrize> especially at hetzner, their big
machines are dedicated cpu and expensive
237[03:03:33] <solrize> i guess i could have done this a bit
differently and saved some time at the expense of more hassle, but
this should finish extracting in 90 minutes or so
238[03:03:43] <solrize> without the file move script it would be
5+ hours
250[03:12:04] <solrize> hmm that's not about a single dir,
but interesting it says xfs is slow and uses more ram
251[03:12:14] <solrize> oh well this is a pretty temporary thing
252[03:12:57] <PMT> also RAM is cheap now compared to 2010 :P
253[03:13:38] <solrize> i have to recreate a few of the split
(10k file) archives that i trashed by accident, and want to make a
de-duplicated archive which uncompressed is about half the size of
the original
254[03:14:02] *** Quits: Filohuhum (~filohuhum@replaced-ip) (Remote host closed the connection)
306[03:55:07] <PMT> cybrNaut: A) as remarked, these are
basically optional in nearly every circumstance, and B)
replaced-url
307[03:57:39] <somiaj> are the partitions mounted?
308[03:58:03] <somiaj> oh wait, nevermind, I guess gpt has a lot
more fancy partition type names.
309[03:58:08] <cybrNaut> sgdisk --list-types has only boot
partitions for BSD, ONIE, PowerPC, & BIOS
310[03:58:22] <PMT> it has basically arbitrary GUIDs for them.
311[03:58:25] <ryouma> in my cursory attempts, i found sgdisk to
be insufficiently scriptable, at least in that you can't
trivially store it and then apply it to a differently sized disk
312[03:58:59] *** Joins: nova (~novasenco@replaced-ip)
335[04:14:13] <cybrNaut> "8303 Linux x86 root (/) 8304
Linux x86-64 root (/)" <= so there's even a need for
architecture distinction? that has to be a screw up
340[04:17:58] <PMT> cybrNaut: the thing I linked above explains
it, the notion was being able to arrange the partitions in the right
places purely based on architectural hints and the typing
341[04:22:17] <cybrNaut> fuck me.. sgdisk also can't handle
writing to sector 34. that seems to be a unique capability of
parted, which has absurdly broken documentation
342[04:23:11] <cybrNaut> unless cfdisk can handle sector 34,
i'm stuck with parted
343[04:23:26] <PMT> I don't think it can, but I also
didn't think parted could, so.
344[04:23:29] *** Quits: catman370 (~catman@replaced-ip) (Quit: See you later..)
350[04:26:39] <cybrNaut> it's funny that parted handles the
job, then when I do a "sfdisk -d" it correctly dumps the
layout, but when I feed that dump back into sfdisk it gives
"out of range"
361[04:36:32] <cybrNaut> it's most likely sloppy. They want
to protect users from shooting themselves in the foot, but they
didn't do the extra work of writing the logic to check whether
sector 34 to 2047 is spanned by a BIOS BOOT partition without, and
without overstepping. So they said "fuck it.. call it a
day"
363[04:38:41] <cybrNaut> another|: that would be rejected
364[04:39:14] <cybrNaut> this is my sample code: sgdisk --clear
--new=1:34:2047 -c 1:"BIOS boot" --typecode=1:$(sgdisk
--list-types | sed -ne 's/.*\(....\).bios.*/\1/gip')
/dev/sdb
365[04:39:44] <another|> hmm.. works pretty well in my script.
notice the order. first creating partition 2, then 1
366[04:39:46] <cybrNaut> i get "Information: Moved
requested sector from 34 to 2048 in order to align on 2048-sector
boundaries. Could not create partition 1 from 34 to 2047"
367[04:40:46] <another|> might also work with setting -a
368[04:42:10] <ryouma> i occasionally partition new drives. i
have an old computer and will get drives that might be external or
internal at some point or might be used on a new computer. if i want
to put enough partitions at the beginning to deal with various mbr
gpt bios uefi whatevers, i.e. enough partitions that i can use for
whatever purposes are needed for that stuff, what is the number of
such partitions that i would need?
370[04:43:08] <cybrNaut> wtf.. the docs don't say the order
of args matters. There's no reason that should work
differently. The fact that we are forced to specify partition number
for every option implies that arg order is moot. But indeed, when i
specify partition 2 before 1, sgdisk accepts it
371[04:45:49] <cybrNaut> another|: how did you know to specify
partition 2 before 1?
372[04:46:40] <PMT> The only comments in parted's source
about the restriction point to a Microsoft KB about having changed
the default alignment of the first partition starting in Vista, but
not why "1 MB" was chosen.
373[04:47:53] <another|> cybrNaut: figured it out. `sgdisk
--clear -a 1 -n 1:34:2047 -t 1:ef02 -c 1:"BIOS boot"
/dev..`
374[04:48:31] <another|> seems the order matters. specify
alignment before `-n`
375[04:49:08] <another|> cybrNaut: trial and error ;)
376[04:50:14] <cybrNaut> ah, so i could perhaps do -a 1 -n
1:34:2047 ... -a 2048 -n 2:0:0 ...
377[04:52:34] <cybrNaut> yeah, that works. but it's still
unclear why flipping the order should make a difference.. that
should fail on the basis of alignment as well
378[04:53:06] *** Quits: catman370 (~catman@replaced-ip) (Quit: See you later..)
396[05:14:32] <cybrNaut> PMT: i just noticed the docs of sfdisk
actually say "It is recommended to save the layout of your
devices. sfdisk supports two ways." .. "Use the --dump
option to save a description of the device layout to a text
file." .. "This can later be restored by: sfdisk /dev/sda
< sda.dump"
425[06:05:12] <PyR3X> listening on a tcp port, I'm unable
to send data outbound after a connection is established, however, I
can send new packets outbound without issue -- what could be causing
this?
429[06:08:08] <ryouma> rephrasing: i have an old computer and
will get drives that might be external or internal at some point and
will also be used on a new computer. i want to put enough 1gb
partitions at the beginning to deal with various mbr gpt bios uefi
whatevers (gpt stuff, bios boot partition, whatever is needed for
uefi, etc. -- i am ignorant), but idk what things are needed. to be
safe, do i need 1, 2, 3, 4 such partitions?
433[06:11:07] <somiaj> ryouma: UFI is actually fairly easy, with
a GTP partition you need one vfat EFI partition, I wuld say
256-512MB is fine, and then all other paritions can be whatever you
want them to be
435[06:11:41] <somiaj> cybrNaut has some very special use case,
such as trying to get legacy boot working on GPT partitions, which
requires a bit more work.
436[06:12:13] <somiaj> at least I think that is party of that,
also they want encryption and a bunch of other things, but depending
on your use case, UEFI + GPT is fairly straight forward and easy,
nothing really special to deal with.
438[06:12:36] <somiaj> I personally suggest use MBR for legacy
boot, and GPT for UEFI just to make life easier.
439[06:13:05] <ryouma> my current computer is bios. idk what my
new computer is, but it might be uefi. i think i will start with the
drive being external.
440[06:14:22] <somiaj> most computers in the last at least 5
years will support uefi and legacy, and I suggest uefi, I find it
much nicer
441[06:14:30] <ryouma> this is a much older computer
442[06:14:59] <somiaj> even up to 10 years you can still get
uefi support, I think I switched to uefi about 10 years ago, but
somewhere (my memeory is bad) it got to be hit and miss
443[06:15:01] <ryouma> the booting software says nothing about
uefi at least
444[06:15:33] <sney> I discovered today that my current desktop,
with a motherboard made in 2012, has uefi support. I guess it's
debian that didn't have support back then because the install
is bios, but uefi certainly isn't new anymore.
445[06:15:34] <ryouma> this is like a 2008 computer
446[06:15:48] <somiaj> I was be really conservaitve with 5 years
for safety, but yea if you only have legacy boot, go with mbr
447[06:16:03] <somiaj> I think the biggest limitation you get is
2TB partition limits, but you can have multiple partitions that size
448[06:16:23] <ryouma> oh great i thought it was the whole drive
449[06:16:29] <somiaj> as I would avoid trying to do legacy mbr
boot on a gpt partition table if you can, it is possible but imo not
worth the effort.
450[06:16:45] <somiaj> I think it is per primary partition (so
maybe only 4 of those)
451[06:16:55] <somiaj> Unsure if logical paritions can each be
2tb or not
452[06:17:03] <ryouma> but i'd want to put an extra
partition for that vfat for later conversino to gpt or so? and add
space at end for the extra gpt partition thingie?
453[06:17:06] <sney> lvm probably bypasses it too
454[06:17:56] <somiaj> I don't think you can really convert
that easilly, or at least I haven't seen much on this. You will
have to rebuild the partition table / filesystems if you want to
switch
455[06:18:08] <ryouma> ok
456[06:18:27] <ryouma> so just do mbr and then will it run on an
uefi computer?
457[06:19:03] <somiaj> sney: I think what tooke me the longest
to switch to uefi was wanting to also switch to gpt partition
tables, so eventally had to get a live system to copy over my old
install
458[06:19:31] *** Quits: kish` (~oracle@replaced-ip) (Remote host closed the connection)
459[06:19:33] <somiaj> just took a bit of work because I
didn't want to install fresh, instead keep upgrading my install
from like 2006 when I swithced to amd64 from i386
460[06:19:46] <sney> ryouma: depending on age, the uefi computer
will most likely have a bios/legacy/csm mode that you can use.
463[06:20:07] <sney> some new machines have gotten rid of CSM
but I think it's just the cutting edge stuff.
464[06:20:08] <ryouma> t need extra partitions then
465[06:20:18] <ryouma> csm?
466[06:20:45] <somiaj> ryouma: yea, I think you are over
thinking it. In general legacy boot + mbr, or uefi + gpt, and
currently the majority of machines support both.
467[06:20:48] <sney> "compatibility support module",
the bios/legacy uefi thing.
481[06:24:04] *** Joins: ruslan (~ruslan@replaced-ip)
482[06:24:05] <sney> it shouldn't
483[06:24:55] <somiaj> Yea, mbr partition table support is not
going anywhere, it is only legacy boot that may be slowly phased
out, but I don't see it happeneing yet
498[06:32:51] <somiaj> outoftime: and you still get that error
even after you install mypaint-data?
499[06:32:58] <outoftime> somiaj: I have installed mypaint, it
just looks like mypaint-data didn't get installled
automatically
500[06:33:50] <outoftime> I just wanted to report a possible
problem.
501[06:34:23] <somiaj> You do have a few thrid party repos, this
can due to dependency trees make things harder for apt to solve
certain trees
502[06:34:58] <somiaj> the fact that install mypaint-data was
able to fix it is where I would stop.
503[06:35:35] <somiaj> Personally I would try to reproduce this
in a fresh debian buster chroot that has none of your third party
repos, because with the info given it is hard to say if this is
reproudciable using only debian sources or what the conflict that
confused apt was.
505[06:35:50] <somiaj> That is if you cared enough to see if
this issue effects debian with only debian sources.
506[06:36:22] <outoftime> I have live but with stretch on it
507[06:37:00] <somiaj> chroots are quick to build with
debootstrap, I might have one floating around.
508[06:37:23] <outoftime> somiaj: yes, I have just remembered
that too
509[06:38:01] <somiaj> fwiw, on my buster vm I have no issue
installing mypaint, but there are lots of libs it depends on
510[06:38:02] *** debhelper sets mode: +l 931
511[06:38:28] <somiaj> maybe you had some issue with one of
those libs and some of the third party sources, but installing
mypaint-data directly was enough to figure out how to deal with it
512[06:39:00] <somiaj> Anyways to me this is more an issue with
your third party sources than debian, but the sources there seem
reasonable, just sometimes apt will have a harder time with
dependency trees as a result.
513[06:40:45] <somiaj> If you wanted to share the log of
installing mypaint-data in /var/log/apt/history.log, I might be able
to figure out which libary
519[06:47:47] <somiaj> Remove: gimp:amd64 (2.10.8-2),
libmypaint-1.3-0:amd64 (1.3.0-2.1), libmypaint-common:amd64
(1.3.0-2.1) -- one of those was probably the issue, but not sure why
520[06:48:13] *** Quits: darunesh (~darunesh@replaced-ip) (Remote host closed the connection)
525[06:51:45] <somiaj> outoftime: you might want to notice that
gimp was removed, looks like you can't have mypaint and gimp
installed at the same time due to this conflict. I'm unsure why
this conflict exists, but it appears
replaced-url
534[06:55:00] <somiaj> for some reason (not quite sure) the
debian packages mypaint and libmypaint cannot both be installed at
the same time, so you have to choose between mypaint or gimp
535[06:55:24] <somiaj> apt just got confused because of this
conflict and that is why you had trouble.
536[06:55:49] <somiaj> yea, seems this is jsut an issue in
buster, bullseye won't have this conflict anymore
537[06:55:50] <outoftime> "This bug has been fixed in
Debian Testing and Unstable with new upstream
538[06:55:55] <somiaj> Yup
539[06:55:57] <outoftime> releases." this is pleasant to
know
540[06:56:47] <somiaj> I'm still a bit surprsied that apt
couldn't figure out to remove libmypaint and gimp without
installing mypaint-data manually, but sometimes conflicts like that
apt can't decide what is best when they are a bit down the
dependecy tree
541[07:01:13] *** Quits: darunesh (~darunesh@replaced-ip) (Remote host closed the connection)
692[10:29:18] <hegemoOn> so are you moving to libera ?
693[10:29:30] <hegemoOn> when debian 11 will be officially out ?
694[10:29:35] <jelly> !wwbr
695[10:29:35] <dpkg> Now Debian "Bullseye" is the
current testing branch as of 2019-07-06 and it will be released
"when it's ready."
696[10:30:05] <hegemoOn> end of may was a target isn't it ?
697[10:30:12] <jelly> it was not
698[10:30:21] <jelly> Debian does not have timed releases
699[10:31:33] <jelly> if you want to guess, the release is ready
when the number of release-critical bugs falls to zero
700[10:31:35] <jelly> !rc bugs
701[10:31:36] <dpkg> Release-Critical bugs are Debian bugs with
critical, grave or serious severities, preventing the next release
of Debian. See the graph at
replaced-url
703[10:32:58] *** Quits: eri (uid469694@replaced-ip) ()
704[10:33:11] <jelly> hegemoOn, it's too early to say what
will remain of Debian unofficial presence on freenode; the OFFICIAL
#debian is on irc.debian.org (= irc.oftc.net) and has been there for
15 years now.
705[10:33:13] *** Quits: n4dir (~n4dir@replaced-ip) (Remote host closed the connection)
706[10:34:08] <jelly> so hop over to OFTC for your debian needs,
that's where all the developers have been hanging out anyway
707[10:34:59] <GreenBean> OFTC was the first time everybody got
pissed off at Freenode
708[10:35:00] <Iridos> 15 years! Seems like yesterday
709[10:35:07] <GreenBean> no it doesn't lol
710[10:35:25] <GreenBean> this time is different though
711[10:35:49] <GreenBean> Honestly OFTC and Libera.Chat should
just merge
727[10:40:19] <ksk> soo, I went from screen to tmux some time
ago, what is the adequate replacement for Irssi? Should it be
replaced at all? (Gotta admit it does work, and you can get basicly
anything you want into it, but most of times I have to join #irssi
to do anything advanced, so there is that..)
728[10:40:25] <H4ndy> I always tell myself that I need to take
care of logs but haven't once needed to search them for
somewthing in like 20 years of IRC usage
756[10:56:18] <EdePopede> i've noticed some shifting issues
with Gomuks in xterm+screen, but not in plain screen. i bet
it's something with the UTF-8 it uses, but maybe just ANSI
sequences. what about irssi (and maybe weechat)?
770[11:03:15] <EdePopede> didn't check its sources, no idea
what it actually uses. only i never had similar issues with vim or
mc or aptitude or whatever
771[11:03:20] *** Quits: endstille (~endstille@replaced-ip) (Quit: I'll be back.)
772[11:03:33] <EdePopede> mauview? sounds like a start.
784[11:12:22] <jsync> I extended my var directory with a 4tb hdd
mounted in an external case, & I then mounted it in the internal
machine, & now it can't boot the os, not actually.
839[12:48:58] *** Quits: LucaTM (~LucaTM@replaced-ip) (Quit: To infinity and beyond...)
840[12:55:14] <oxek> this doesn't have a CVE assigned, but
is it fixed in rxvt in buster?
replaced-url
841[12:55:20] <EmleyMoor> There seems to be an increase in
traffic to the Internet through my Debian-based router since a
little before 01:00 UTC today... what's a good way to see where
that is coming from within my network?
842[12:56:25] <petn-randall> EmleyMoor: wireshark? Or tshark.
865[13:11:54] <petn-randall> funabashi: If you set a root
password during installation, your primary user does not get root
privileges. If you don't set the root password, your primary
user is added to sudo and sudo is installed.
907[14:29:54] <jelly> dob1, so until we figure out where things
are going with freenode and libera.chat, OFTC is the safe, sane,
secure place to /join #debian
929[14:57:04] <jelly> I'm seeing many migrate to libera,
some to OFTC, some deliberating, occasional "we're on
Matrix now" craziness
930[14:58:22] <jelly> #debian and #debian-offtopic is in the
third group, and it would REALLY be the easiest thing for us to
finish the bloody migration to OFTC that started 15 years ago
931[14:58:32] <jelly> but I'll note we did set up a #debian
presence on libera.chat preemptively.
932[15:00:04] *** Quits: n4dir (~n4dir@replaced-ip) (Remote host closed the connection)
948[15:11:01] <jelly> moving to OFTC would have likely been a
more complex deal due to channel policy differences, and not easily
done on a short notice
952[15:12:02] <oxek> jelly: are you aware of any nice writeup
somewhere that would document the differences between freenode &
oftc, with respect to channel policies and other things?
953[15:12:27] <jelly> sadly no
954[15:12:45] <oxek> yeah, I too tried looking for such a
document but didn't find anything
1025[16:26:55] <urgodfather> hello all, im struggling to create a
pseudo interface. i've tried both a virtual and a dummy with no
prevail. will someone pass some insight on this?
1027[16:28:03] <PMT> Information like precisely what you tried,
what you want to accomplish, and what happened instead of working
would probably be helpful.
1031[16:32:44] <urgodfather> both create an interface as it
should but not delivering the desired results. i need to replicate a
physical interface as close as possible. i am aware that i will also
need to create routing policies for traffic.
1040[16:38:48] <urgodfather> i need a second interface for a
software application. i've tried a virtual, a dummy, and even a
vpn interface. none fit the bill. surely this can be done. the
interface doesnt require a physical connection, only the capability
to be bound to within the software layer.
1041[16:39:34] <PMT> You're still not explaining what
it's not doing.
1062[16:47:29] <urgodfather> the goal will be out of scope for
their chain of support, thus, my asking for general guidance or a
different approach i may have overlooked
1079[16:59:04] <urgodfather> since this is hosted, i cannot get
beyond the "external" network. it cant create an
"internal" b/c there isnt one. to pacify this, im
assimulating the second interface so that i can create that internal
network. once done, i can address routing in iptables as needed
1162[18:14:49] <jsync> I guess it has something to do with where
the 4tb hdd was mounted when I extended the hdd. Maybe it could be
fixed in bios before boot?
1163[18:15:13] *** Quits: idhugo (~idhugo@replaced-ip) (Remote host closed the connection)
1194[18:26:05] <cybrNaut> "$ symbolizes a command to be
entered in the user's current system, while # refers to a
command entered in the Debian chroot."
1195[18:26:24] <jsync> PMT, Yeah, this actually is the first
reboot.
1196[18:26:40] <cybrNaut> that's in the 1st paragraph
1280[19:18:40] <jelly> if file command found something that looks
like lvm, then you're lucky and might just have to recreate the
partition in the partition table
1300[19:42:09] <somiaj> cybrNaut: no that won't have any
effect, a repo will contain multiple releases so a release file is
needed. A cd image won't so a release file isn't
necessarly needed. It might be that deboostrap won't work from
a cdimage (I've never tried or seen any docs on this)
1315[19:48:04] <PMT> There's apt-cdrom, but I specifically
meant a handler for URIs starting with 'cdrom:'
1316[19:48:16] <PMT> which I presume is what apt-cdrom adds.
1317[19:48:57] <cybrNaut> apparently someone figured that users
will check the ISO hash & sigs, and so Release.gpg isn't
needed. But they didn't anticipate debootstrap use cases
1331[20:01:23] <cybrNaut> it's hard to trust these docs
since they botched telling users when to use chroot. Are there
better debootstrap docs anywhere?
1332[20:01:57] <cybrNaut> i found the wiki but it's also
quite rough and minimal:
replaced-url
1339[20:03:32] <somiaj> Though I just use a mirror
1340[20:04:09] <somiaj> also where do you think you should be
chrooting? What did the install guide get incorrect there?
1341[20:04:25] <rulezzz> hello there I cant configure my usb
webcam whent I run chesse and try to switch cameras betwen the
laptop and usb it show a black window
1342[20:05:18] <cybrNaut> a debian install on a dir that's
not a mountpoint would result in an unbootable installation. The
files need to be in root (/) in the most typical use cases
1343[20:05:45] <somiaj> sure, but that isn't the only use
case of debootstrap, I'm sure you can adjust.
1344[20:06:00] <cybrNaut> somiaj: the doc says: "$
symbolizes a command to be entered in the user's current
system, while # refers to a command entered in the Debian
chroot."
1345[20:06:19] <cybrNaut> every single command on the page is
prefixed with "#"
1349[20:07:48] <somiaj> what? there is no chroot going on there,
it is saying the first thing you ahve to do is create the
filesystems you want to isntall on.
1350[20:08:11] <cybrNaut> somiaj: note the "#"
1351[20:08:29] <cybrNaut> somiaj: the doc says: "$
symbolizes a command to be entered in the user's current
system, while # refers to a command entered in the Debian
chroot."
1352[20:08:49] <sussudio> "niston2" is spamming me in
PM.
1353[20:08:54] <somiaj> OH sorry, I missread that, I was thinking
# = root, $ = user, so yea file a bug.
1354[20:09:06] <sney> sussudio: /umode +R
1355[20:09:08] <sussudio> and 14 other fuckers.
1356[20:09:10] <tuxd3v> PMT, breaks with a message, I suspect
from qemu-user-static saying it was not possible to alocate memory(
commpage memmory )
1390[20:20:50] <cybrNaut> the debootstrap manpage says this
command will "Complete the bootstrapping process", but the
debian.org doc says it's only needed if the "target
architecture is different from the host" => debootstrap
--second-stage
1391[20:21:23] <tuxd3v> PMT, that is not the exact message, but
its close
1392[20:21:30] <cybrNaut> if my host architecture and target
architecture are the same, should i skip the --second-stage?
1393[20:21:46] *** Quits: earthundead (~earthunde@replaced-ip) (Remote host closed the connection)
1394[20:21:59] <tuxd3v> cybrNaut, yes you should skip
1408[20:27:10] <PMT> jsync: I would probably recommend very
carefully manually recreating the partition table, not using an
installer prompt to do it, as installers usually go on to clobber
the contents of the new partitions too.
1409[20:27:17] <rulezzz> input: HP HD Webcam [Fixed]: HP HD Web
as /devices/pci0000:00/0000:00:14.0/usb1/1-4/1-4:1.0/input/input59
1418[20:30:44] <PMT> There are a whole bunch of partitioning
utilities. parted and fdisk are two examples. I do not know what
options are present on the live disk you're running from.
1460[20:48:51] <PMT> jsync: well, people keep telling you to
recreate the partition table, and you won't do that, so now
it's time for alternate strategies.
1481[20:54:17] *** Quits: cosimone (~user@replaced-ip) (Remote host closed the connection)
1482[20:54:21] <jelly> it shouldn't be too hard to confirm
"there are no files under /var right now because it's a
separate filesystem and it's not mounted"
1531[21:37:33] <cybrNaut> how does whether it's modular
affect creating files in /dev? I'm trying to get through the
"D.3.4.1. Create device files" section of the debootstrap
guide (replaced-url
1532[21:37:59] <somiaj> for the most part that is a bit dated,
udev will handel /dev as a tmpfs
1534[21:38:20] <cybrNaut> i know udev is normally used, so i want
to stick with that
1535[21:38:55] <somiaj> MAKEDEV generic should be enough, if udev
is used, the actualy contents in /dev are ignored and a tmpfs is
mounted overthem by udev
1536[21:39:48] <somiaj> note it also isn't unreasonable to
just bind mount /dev from outside the chroot to inside the chroot so
you ahve access to the devices, mostly needed if you want to mount
other filesystems or install grub
1545[21:42:51] <somiaj> the issue with a manual install is there
is a lot of variety about what one needs to do, so it is hard to put
a single step by step guide, but more a set of things that you
should consider and adjust to fit your needs
1549[21:44:11] <cybrNaut> ideally the guide should at least show
what the installer normally does. If someone needs to deviate on a
step, they will already know that.
1587[22:16:05] <jelly> I think I've had that situation maybe
twice since udev became a thing, so that's since 2004ish? when
were lenny and etchnhalf released
1591[22:17:51] <sney> I've used mknod to manually create a
stubborn device node a single-digit number of times since then, just
enough to remember the syntax without looking it up
1602[22:25:14] <jelly> jsync, there's no way to reduce a fs
and the LV under without data loss unless you first restore
connectivity to all the PV parts and get the VG and LV online
1604[22:25:52] <jelly> jsync, so you have basically two options:
either fix the PV access issues, or accept data loss are restore
contents of /var from backups
1709[23:22:05] <dpkg> If you have a question, just ask! For
example: "I have a problem with ___; I'm running Debian
version ___. When I try to do ___ I get the following output ___. I
expected it to do ___." Don't ask if you can ask, if
anyone uses it, or pick one person to ask. We're all
volunteers; make it easy for us to help you. If you don't get
an answer try a few hours later or on
replaced-url
1710[23:22:06] <jelly> nope, sorry, all out of help
1711[23:22:41] <abrotman> I'm all out of help, I'm so
lost without you ..
1712[23:23:35] *** Quits: badsektur (~badsektor@replaced-ip) (Remote host closed the connection)
1713[23:23:42] <rulezzz> or how can I disable one webcam
1714[23:24:26] <abrotman> Does it use the same kernel module? you
could block the module/device from loading
1717[23:25:11] <rulezzz> I think taht the two webcams use the
same kernel module
1718[23:25:34] *** Quits: Haudegen (~quassel@replaced-ip) (Quit: No Ping reply in 180 seconds.)
1719[23:25:56] <abrotman> you may still be able to block the
device
1720[23:25:59] <somiaj> maxrazer: I think you might still need to
manually run xmodmap, the place to run it can very depending on your
xsession,
replaced-url
1721[23:26:20] <somiaj> maxrazer: adding commands to a file just
allow you to run xmodmap filename
1723[23:27:05] <maxrazer> So, you still have to call xmodmap.
That needs to be called from somewhere. I saw a forum post about
somebody trying to do that and it not working from chron or other
place
1724[23:27:54] <somiaj> maxrazer: what xession do you use (as in
how do you startx (a display manager) and if so which session do you
choose?)
1741[23:30:13] <maxrazer> Yes, I put exec i3 in there.
1742[23:30:14] <somiaj> maxrazer: sounds like you could setup a
~/.xsession file to run xmodmap then exec the i3wm wm
1743[23:30:33] <somiaj> maxrazer: okay before that line add your
xmodmap line or anyother commands you may wish to run in your
session
1744[23:30:37] <maxrazer> Ok, yes that might work.
1745[23:32:31] <jhutchins> !webcam
1746[23:32:31] <dpkg> For webcam device support in Debian, ask me
about <gspca>, <uvcvideo>. See also <quickcam>,
<ov51x-jpeg>, <ov511>, <m560x-driver>,
<w9968cf>, <pwc>, <sn9c20x>.
replaced-url
1749[23:35:01] <maxrazer> I'm not sure if I have to specify
.xmodmaprc in the .xsession call, or if it will default to that. I
created a file .xmodmaprc in $HOME with the expression and called
"xmodmap" before "exec i3" in my .xsession in
$HOME.
1750[23:35:39] <maxrazer> I guess I'll log out and see if it
works.
1772[23:43:52] <somiaj> velix: I don't know if acl's
can do this, but this really won't stop a user from running a
binary, they could download a copy of the binary, put it in $HOME
and run it anyways
1773[23:44:08] <velix> somiaj: Ok. I thought AppArmour or such
might help.
1775[23:44:16] <maxrazer> No, I need the file name. It did not
work without it.
1776[23:44:20] <velix> somiaj: I'll try ACLs. Good idea.
1777[23:44:45] <somiaj> velix: but again that really isn't
that much of a secuirty step, since as I said the user could
manually download the binary (or if they can read it, make a copy
and run the copy).
1778[23:44:55] <velix> somiaj: Sure. I just was curious :D
1779[23:44:55] <somiaj> velix: hence why I wanted to know what it
is actually wanted to achive.
1791[23:47:04] <PMT> rulezzz: did you follow all parts of the
suggestions in the question you linked about this
1792[23:47:13] <somiaj> maxrazer: yea, I figured you needed the
filename, xmodmap would just print info if you didn't. You
could have just tested that in a terminal too
1793[23:47:14] <velix> somiaj: disallow curl, wget ;) just joking
1794[23:47:39] <cybrNaut> i have a line in /etc/modules:
"loop" what is that for? is that needed to do "mount
-o loop"?
1795[23:48:36] <somiaj> yea looks like that modules if for
loopback devices
1796[23:48:39] <somiaj> block devices
1797[23:49:47] *** Quits: NikopolSohru (~NikopolSo@replaced-ip) (Remote host closed the connection)
1798[23:50:04] *** Quits: dvs (~hibbard@replaced-ip) (Remote host closed the connection)