6[00:04:51] <somiaj> tramplefoot: yea, all debian packages will put documtation in /usr/share/doc/packagename/ as files, and you can just go read it using your prefered tool
8[00:05:57] <phogg> Is there an easy way to get a list of all currently-installed packages for which there is no installation candidate from a repo? What would be reported by apt-cache policy as being from /var/lib/dpkg/status?
9[00:06:09] <somiaj> phogg: aptitude search ~o
10[00:06:15] <phogg> thanks
11[00:06:35] <somiaj> we call those obsolete packages, and aptitude can search for them (the ~o is a shortcut for that)
12[00:08:43] <phogg> What about if there are installation candidates, but not for the specific version currently installed? I'm trying to resolve a situation where I had packages from testing, but then dropped testing as a source. I'm trying to figure out which related deps remain.
23[00:20:17] <phogg> if not don't worry about it, I can always apply a little script-fu to the problem
24[00:21:12] <somiaj> !not available
25[00:21:12] <dpkg> To get a list of packages you have installed now, that are not available from any repository in your sources.list: aptitude search '?narrow(?not(?archive("^[^n][^o].*$")),?version(CURRENT))'
26[00:21:29] <somiaj> I think that works, it compares the version in the archive to the current version
27[00:23:18] <phogg> hmm nice, better than the shell monster I was contemplating
28[00:23:52] *** Quits: isostatic (uid224824@replaced-ip) (Quit: Connection closed for inactivity)
31[00:24:39] <somiaj> aptitude has super cow powers
32[00:24:44] <phogg> it covers both older (out of date) packages e.g. from prior releases and too-new packages, but I can pretty easily filter from there.
33[00:24:51] <phogg> you can tell by the mooing
34[00:25:35] <phogg> it's too bad aptitude's search features aren't more widely known, they do solve a lot of problems
43[00:31:24] <phogg> I think that's a universal constant
44[00:32:32] <oxek> I forget which one is supposed to be the actual file, and which one is supposed to be the link to it.
45[00:32:39] *** Quits: dvs (~hibbard@replaced-ip) (Remote host closed the connection)
46[00:32:58] <oxek> phogg: is there some common pattern I am missing?
47[00:33:06] <phogg> no, it's counter-intuitive for sure
48[00:33:57] <oxek> every time I have to run 'ln', I have to open the manpage first...
49[00:34:07] <phogg> the thing to remember is that the target (/path/to/x in your example) isn't a file (and doesn't even need to exist), it's a string.
50[00:34:46] <oxek> and that would create a broken link if /path/to/x does not exist, right?
51[00:34:49] <phogg> yes
52[00:35:16] <phogg> the last arg (the link name) is the file that you create, it will point to whatever target
53[00:35:39] <phogg> or if its a dir a file named after the basename of each target will be created in that dir pointing at the full target string
60[00:37:32] <oxek> I meant it targets both the file and the symlink
61[00:38:05] <phogg> be clear "target" is the string the symlink points at (this is how it is described in the man page). It is never the symlink file itself.
62[00:38:27] <oxek> you're right, I'm just confused
63[00:38:38] <oxek> thank you
64[00:38:55] <phogg> confusion about ln argument order is normal
65[00:39:05] <phogg> I know it pretty well but will still sometimes call it backwards
85[01:04:55] <phogg> somiaj: this gets me what I was looking for: aptitude search -F '%p %v %V %O' '?narrow(?not(?archive("^[^n][^o].*$")),?version(CURRENT))' | while IFS=' ' read -r pkg ver candidate origin; do if [[ $ver == "$candidate" ]] && [[ $origin == '(installed locally)' ]]; then echo "$pkg"; fi; done
86[01:05:04] *** Quits: werneta (~werneta@replaced-ip) (Remote host closed the connection)
87[01:05:26] <phogg> I couldn't see how to get aptitude search to do the ?version(CURRENT) equal to ?version(TARGET) itself, but there might be a way
100[01:27:34] <somiaj> and if you really want to get funky, these also work with install/remove and aptitude, so you can script a lot of stuff (though it can also be dangerous, outside of using search I avoid wildcard installs with aptitude)
101[01:28:08] <somiaj> the only one I often use is aptitude remove ~o (on systems I'm 100% stable) and there isn't anything in aptitude search ~o I want to keep, simplifies typing out all the package names
107[01:39:13] <tramplefoot> xfce4-weather-plugin on buster doesn't work anymore because the weather data provider met.no has changed their api again. the testing package doesn't seem to have this problem, but i don't want to mix testing and stable packages together. i want to create a private backport, but how possible is backporting when the main dependency issue is libc?
114[01:42:38] <dpkg> First, check for a backport on <debian-backports>. If unavailable: 1) Add a deb-src line for sid (not a deb line!); ask me about <deb-src sid> 2) enable debian-backports (see <bdo>) 3) apt update; apt install build-essential; apt build-dep packagename 4) apt -b source packagename 5) dpkg -i packagename-ver.deb To change compilation options, see <package recompile>; for versions newer than sid see <uupdate>.
115[01:43:21] <somiaj> Provided the package is mostly standalone, I think a backport from testing would be fine, debhelper-compat is in backports, so you should be able to build it on stable just fine.
130[02:00:00] <Heart_And_Chops> ok. maybe a step back. how do i give a non root user access to a usb device? i have a small script i wrote in c++ and when i run it as my regular user it fails, if i run it with sudo it works fine.
131[02:00:25] <mutante> Heart_And_Chops: grep usb /var/log/messages ; lsusb -v
150[02:02:27] <somiaj> I dno't think there is /dev/usb, but I am not familar with scanners, if we could track down the device actually being used, that would help solve this problem.
151[02:02:30] <Heart_And_Chops> no not in lsblk. it shows in usb-devices or lsusb
172[02:07:46] <somiaj> and you should do root.group and add yourself to the group, and then update udev to apply the permsision correctly
173[02:08:20] <Heart_And_Chops> somiaj: that sounds like a better solution but you've lost me a bit, can you walk me through that please?
174[02:08:34] <mutante> that page has info on finding the scanner, the device ID and also privileges for users
175[02:08:45] <mutante> i think all the stuff for a regular scanner would apply here
176[02:08:48] <tramplefoot> is this an nfc card scanner or like a optical scanner
177[02:08:54] <tramplefoot> a nfc**
178[02:08:55] <somiaj> Heart_And_Chops: I would have linked the page mutante did, it gives some info about how to configure udev
179[02:09:01] *** debhelper sets mode: +l 1135
180[02:09:23] <Heart_And_Chops> it does a bunch fo stuff, its got a special driver with an SDK. it scans the RFID chip in a passport, also does IR, UV and regular light photos with a camera.
181[02:09:56] <tramplefoot> if has an optical component, sane would probably do the trick
182[02:10:01] <mutante> ooh.. that sounds like more fun
183[02:10:05] <mutante> multiple scanners in one, kind of
184[02:10:22] <somiaj> tramplefoot: I think here the issue is not accessing the device, but assiging it correct permissions, they already have a tool/script that works with it
185[02:10:37] <tramplefoot> oh right
186[02:10:49] <somiaj> so sane might not do much, they have tracked down the device, now just need to setup a udev rule to make the permissions persistant
202[02:22:23] <somiaj> that I don't really know, I'd have to research the syntax, but you need to be able to idetinfiy the usb device by some unique feature, then configure the options. Though udev is farily powerful, sot he docs are fairly extensive.
203[02:22:47] <somiaj> maybe someone with experience with udev rules would be able to help (or you will have to manually change the ownership each time you boot/plug it in
204[02:24:30] <somiaj> also put the file in /etc/udev/rules.d/.... (not /lib -- my mistake there, /lib is for system stuff from dpkg/.deb packages, and /etc/ is for user files)
223[02:40:03] *** Quits: ttill (~ttill@replaced-ip) (Remote host closed the connection)
224[02:40:14] <Heart_And_Chops> zamnedix / somiaj / mutante : thank you, i got the udev rules working and it seems to persist a reboot now. very very nice. i like. <3
243[03:04:15] <oxek> but debian packaging for bash removes that dot
244[03:04:19] <oxek> why is that?
245[03:06:43] <n4dir> wild guessing would be that the dot refers to pwd, so you don't accidently run an executable in there
246[03:06:44] <EmleyMoor> oxek: Not sure if this is relevant but . in the PATH is a potential security issue as a potentially dangerous executable in the current directory could be started in error.
320[04:41:11] <brachamh> hello, trying to get Evolution (on Debian Buster) to use my Nextcloud contacts via CardDAV. It seems to sign in but then doesn't show any addressbooks
345[05:13:22] <lenswipe> im aware that a reddit post is hardly rock solid evidence, but still
346[05:13:54] <sponix> lenswipe: and there is a reason you are using a flatpak instead of a regular system package ?
347[05:15:03] <somiaj> you wouldn't happen to need a newer version of flatpack for that feature?
348[05:15:17] <lenswipe> sponix, 1. The system package is extremely old and brings with it a version of jack that has broken my audio more times than I can count 2. Someone in here insisted it was a better way to load software (actually for OBS, not Ardrour but the point remains)
349[05:16:08] <lenswipe> somiaj, So....flatpak exists so that you can easily install newer versions of applications than are in the system repos without breaking things...except that package manager itself needs updating
350[05:16:14] <lenswipe> this is like an ice-t meme
351[05:16:23] <lenswipe> my bad. icecube
352[05:16:47] <somiaj> lenswipe: I understand that, but isn't there a package/binary that installs these packs?
353[05:16:58] <lenswipe> yeah, it's called flatpak
354[05:18:31] <somiaj> that is what I'm talking about, it could be that the flatpak binary is part if what is needed for this feature and the one you have may not have this feature -- just a guess
407[07:12:01] <gry> Hi, ratrace spoke with me a few days and said to "[use] VBoxManage [to] convert from vdi to raw image for qemu to use". anyone here have experience with this? where is the correct button?
445[08:30:31] <ratrace> gry: I don't know lxde, and you could google for a power management plugin or something, but also systemd-logind is capable of that iirc, check out HandleLidSwitch option of /etc/systemd/logind.conf
480[09:20:23] <gry> I think 'ignore' worked. Thank you, ratrace
481[09:20:31] <gry> (for lid close)
482[09:21:09] <gry> in virt-manager, I start game 'minetest', it has issues with mouse. in virtualbox the same issue is resolved by disabling 'mouse integration'. how can the same setting be toggled in virt-manager ?
483[09:21:19] <gry> it is debian host and debian guest
485[09:26:23] <ratrace> gry: I don't think qemu-kvm has that ability unless you install special support in the guest OS. In case of spice-based display (default I think) that'd be `spice-vdagent`
486[09:26:38] <ratrace> gry: by the way, gaming in VM is going to be terrible without GPU passthrough
487[09:27:43] <diogenes_> gry, you might get a better performance with vmware player.
488[09:29:17] *** Quits: yonder (~yonder@replaced-ip) (Remote host closed the connection)
489[09:33:40] <gry> ratrace: installing spice-vdagent did not help
509[09:58:08] <gry> i installed that package on the guest and restarted the guest; in minetest the mouse keeps looking 'down into the ground' and i can not navigate in the game; in virtualbox when i enable mouse integration this behaviour becomes normal
515[10:02:23] *** Quits: bn_work (uid268505@replaced-ip) (Quit: Connection closed for inactivity)
516[10:04:41] <ratrace> gry: no idea really. I do have a gaming VM (qemu-kvm based) but with PCI passthrough (GPU and a USB pci lane for mouse and kbd).
519[10:06:26] <ratrace> gry: by default with qemu and spice, when you click on the window it should grab the mouse, and if it doesn't try pressing ctrl+alt+g keys simultaneously. with captured mouse, the guest OS should behave as if it was normal local hardware
520[10:06:45] *** Quits: st-gourichon-fid (~Stephane@replaced-ip) (Remote host closed the connection)
521[10:07:19] <ratrace> with "integration", the mouse is not captured but roaming freely in and out of the VM window. makes for a more seamless desktop experience in some cases (Eg. where the VM window represents one application)
524[10:08:08] <ratrace> btw, pressing ctr+alt+g again will "uncapture" mouse (and keyboard) again
525[10:09:07] <gry> ratrace: in vitualbox, it is either 'not integrated, yes captured' or 'not ingetrated, not captured' or 'integrated'. the latter setting appears to work the best
536[10:14:27] <gry> it is too much for me to understand, i don't know the protocol
537[10:14:52] <ratrace> maybe that's red herring and it's the _spice_ interface that's giving you issues. maybe you can change that to vnc? I don't really use virt-manager so I don't know if and where is the config option for that
538[10:16:34] *** Quits: _aeris_ (~aeris@replaced-ip) (Remote host closed the connection)
581[10:55:26] *** Quits: Anderson69s (~Anderson6@replaced-ip) (Remote host closed the connection)
582[10:57:42] <ratrace> footnote, but /etc/mailname is only _default_. Postfix doesn't have to use it. In fact it's better it doesn't because mailname fragments configuration (by adding yet another place where you need to configure one of the local hostnames)
603[11:09:37] <tehuty> ratrace: It is not a crossposting it is an different questiion which I ask in #debian. It's is related to the defaults set in this OS.
618[11:17:27] <dpkg> Use «hostname foo» to set the hostname, $EDITOR /etc/hostname to set it for the next boot (create /etc/hostname if it does not exist) and $EDITOR /etc/hosts to set up local translations for a FQDN. See also 'man 1 hostname', or ask me about <mailname>. replaced-url
620[11:19:57] <tehuty> hostname -f gives me already my fqdn. hostname is set to my correct host name.
621[11:20:31] <hegemoOn> is there any GNU standard explaing why for exemple gcc -v output in stderr instead of stdout ?
622[11:20:43] <hegemoOn> *explaining*
623[11:20:56] <tehuty> /etc/hosts is set to fqdn like this 1.2.3.4 service.domain.com service
624[11:20:57] <jelly> CyberManifest, that question is asking about an older Ubuntu release; it is not a good idea to assume optimal solutions will be the same on different distros; the accepted answer says "setting the domain" where it only sets the search domain for the resolver (ns lookups)
625[11:21:11] *** Quits: szorfein_ (~ninja@replaced-ip) (Remote host closed the connection)
626[11:21:15] <jelly> tehuty, what's the actual line from /etc/hosts?
627[11:21:32] <jelly> and actual output of hostname -f?
635[11:23:00] <jelly> I ask for actual output because that way it's easier to help you.
636[11:23:02] *** Quits: horner__ (~horner@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
637[11:23:50] <jelly> if you're changing some parts of it but not other, and you don't tell us that, it makes it hard to compare certain things and see if things are set up correctly
639[11:24:27] <tehuty> I see but you have to trust me it's runing like that currently
640[11:25:06] <jelly> then I don't have a lot more free help to offer, sorry
641[11:25:22] <tehuty> should give dnsdomainname gives me my domainname, but domainnamen nothing.
642[11:25:34] <CyberManifest> jelly: solutions should be cross compliant IMHO not to confuse the Linux standards
643[11:26:08] <tehuty> */usr/bin/domainname
644[11:26:50] <jelly> CyberManifest, except I don't KNOW what standards apply, but I AM here in #debian having a bit of know-how about Debian and its best practices. Not other distros.
645[11:27:19] <CyberManifest> jelly: Linux is Linux
646[11:27:50] <jelly> CyberManifest, in general I'd agree with that ideal, but in practice every distro has its own nuisances
647[11:28:14] <CyberManifest> jelly: and hence the headaches
648[11:28:17] <CyberManifest> and the egos
649[11:28:43] <ratrace> isn't "domainname" only for NIS? like, it's irrelevant that it says (none) if you don't use NIS
650[11:29:06] <jelly> tehuty, domainname returns "(none)" on all the debian installs that I tried
651[11:29:26] <jelly> some of them are mail servers w/ Postfix
652[11:29:50] <ratrace> it's hostname and hostname -f that matters here (for postfix). /etc/hostname is typically nonfqdn, and entries in /etc/hosts are the ones that define fqdn, because fqdn depends on the interface used, thus the reverse IP lookup via hosts
654[11:31:31] <jelly> and distros disagree even on those simple things. On redhat/fedora/centos, hostname command and /etc/hostname will typically be the whole fqdn and tools will expect that.
656[11:32:17] <tehuty> ratrace: in this case everything is correctly set. but does domainname nothing and posconf -d hostname shows a wrong domainname part?
657[11:32:17] <ratrace> right
658[11:32:19] <jelly> on debian and probably at least some derivatives, you want only short name in /etc/hostname, and a matching ip + fqdn + short entry in /etc/hosts
659[11:32:26] <ratrace> tehuty: domainname is for NIS
660[11:32:46] <jelly> CyberManifest, so linux isn't quite linux
666[11:34:32] <jelly> CyberManifest, you can't seriously argue that redhat and debian, both 25 years old or so, have intentionally obfuscated things; it's just historical baggage for both and it's hard to change things.
667[11:34:43] <ratrace> tehuty: maybe you wanted `hostname -d` to get the non-fq domain part?
674[11:36:31] <jelly> like, actual fucntionality, and not just their expectations of command outputs
675[11:36:32] <tehuty> ratrace: didn't thing about that, I thought this thing would read the same, like the name says domainname. Every seems to setup right, I was counfused bei postconf -d which reads a wrong domainname?
676[11:36:53] <jelly> tehuty, postconf -d is NOT the actual value used.
677[11:37:00] <jelly> run postconf without -d
678[11:37:30] <ratrace> tehuty: where you you see /usr/bin/domainname in relation to postfix?
679[11:37:45] *** zalt_ is now known as zalt
680[11:37:45] <ratrace> yeah -d is defaults, and -n is for explicit overrides
681[11:37:57] <tehuty> I just wonder if postfix could read that setting from system defaults and how it would be set correctly so that postfix reads the right setting.
682[11:38:10] <jelly> tehuty, it doesn't matter.
683[11:38:58] <jelly> the value actually used by postfix is shown with "postconf myhostname" and not "postconf -d myhostname"
684[11:39:17] <tehuty> Yes when I have set this setting in postfix it doesn't matter. true.
685[11:39:45] <jelly> and it's set by default when postix package is installed in Debian
699[11:48:48] <CyberManifest> jelly: yes, but same principle; just fyi I've dabbled with Linux since RedHat Apollo, so I'm familiar with the time sequence
700[11:49:35] <jelly> and honestly linux distros look a lot more similar to each other compared to differences in all the unix OSes they mostly replaced
780[12:53:17] <jelly> dka, are you running linux or windows or what right now
781[12:53:58] *** Quits: Psil0Cybin (Psil0Cybin@replaced-ip) (Quit: Well its 2am, time to go home and spend some quality time with the kids.)
782[12:54:50] <dka> I am running linux debian
783[12:54:59] <jelly> oh, you're on debian. Then just do what diogenes said, make sure the usb stick isn't mount before writing to it
784[12:55:16] <dka> ok
785[12:55:22] <ratrace> dka: are you asking about copying the ISO onto a stick, or makign your own bootable stick, with modifications from that ISO as source?
786[12:55:39] <ratrace> (keyword: with modifications)
787[12:55:58] <dka> I don't want to edit any iso, I got firmware-10.5.0-amd64-DVD-1.iso and it's fine
788[12:55:59] <jelly> lsblk will show you disks/block devices, you can figure out which sdX your USB stick it by size, probably
789[12:56:16] <dka> I only want to make it bootable so I can format my server and install debian with it
790[12:56:23] <ratrace> dka: then cp it is. alternatively some people prefer dd .
791[12:56:28] <jelly> it's bootable from usb by default.
792[12:56:48] <jelly> if your server can boot from usb storage you'll be fine
796[12:57:20] <dpkg> You can install Debian from a USB stick/thumbdrive/pen drive/key on x86 systems, as long as your system's BIOS can boot from USB. Details are in the Installation Guide, see replaced-url
812[13:12:47] <tomreyn> just a suggestion: pointing to /dev/disk/by-id/ may help novices identify the correct target more safely (than saying "/dev/sdx")
836[13:40:31] <dka> I have Partition(s) 1 on /dev/sda have been written, but we have been unable to inform the kernel of the change, probably because it/they are in use. As a result, the old partition(s) will remain in use. You should reboot now before making further changes.
840[13:41:39] <diogenes_> Psil0Cybin, i only buy HPs and they have always been reliable and work out of the box with linux, just don't buy a brand new machine and expect everything to work.
842[13:43:51] <tomreyn> diogenes_: FWIW, I can't confirm 'out of the box' with a recent but not new hp laptop model. but it'd work good enough with some modifications.
843[13:44:52] <tomreyn> this is a 'convertible', business-oriented, firmware overloaded, model, though.
934[16:09:42] <oxek> you don't know how many partitions on /dev/sda are mounted, and you'd need to unmount them all, and you should probably cp the .iso to the block device directly as opposed to a partition on the block device, so specify /dev/sda instead of /dev/sda1
937[16:12:46] <oxek> also, unless you're on an nvme ssd, /dev/sda is likely pointing to an actual harddrive, so be careful with that - you might probably need /dev/sdb or similar
938[16:13:52] <oxek> that's why we have /dev/disk/by-uuid and similar instead
939[16:14:20] <dka> it seems like it worked
940[16:14:25] <dka> I will try to boot from it now
941[16:14:52] <ratrace> it'll "work" in that it won't error out, but you can't boot that stick, from those exact commands
942[16:14:56] <oxek> I'd be surprised if cp to /dev/sda1 will make it bootable
943[16:15:27] <ratrace> dka: so like oxek said, assuming sda is the USB stick and not something else, you must cp to the device itself (sda), not its first partition (sda1)
944[16:15:49] <dka> so how did i get the files on the usb stick with that exact cmd
946[16:16:55] <ratrace> that particular command would only copy the ISO contents to the 1st partition, bit-by-bit, not as a file being copied to a filesystem
978[16:44:19] <ratrace> dka: the ISO itself is a "snapshot" of entire USB contents, partitioning included, so you must copy it to the whole device itself, not a partition on it.
980[16:45:22] <ratrace> dka: so if /dev/sda is the correct device (the USB), that's where you `cp` the ISO to. It's just that 'sda' is quite an uncommon designation for a device you didn't boot from
1017[17:29:49] <nkuttler> oxek: i'd guess it inherits, SHELL is dash for me as login shell
1018[17:30:45] <oxek> I booted with init=/bin/dash, and $SHELL is empty. Then I started bash and $SHELL is /bin/bash, and then I started dash from bash, and $SHELL is again empty
1019[17:30:49] <oxek> something is strange
1020[17:31:09] *** Quits: libertycity (~libertyci@replaced-ip) (Remote host closed the connection)
1021[17:31:13] <sney> sounds like that's not a variable that dash sets.
1022[17:31:56] <nkuttler> like i said, when used as login shell i get /bin/dash as SHELL
1023[17:32:10] <nkuttler> (this is on unstable though)
1122[19:50:26] <oxek> karlpinc: I wouldn't know how. I think I'll just live with it, it's not great a deal, just means I can't rely on $SHELL actually being a reliable indicator of what the shell is.
1223[21:50:03] <Psil0Cybin> hey i need help i am able to install fedora core workstation on my HP new laptop and it works perfectly fien but i really want to run DEbian instead
1224[21:50:13] <Psil0Cybin> what is the best ISO that i can compare to the workstation from fedora?
1225[21:51:08] <petn-randall> Psil0Cybin: No idea what Fedora Core Workstation is, but have you tried installing with the firmware installer?
1226[21:51:12] <petn-randall> !firmware installer
1227[21:51:12] <dpkg> Debian-Installer is able to load additional <firmware>, by including it within installation media or supplying on removable media (e.g. USB stick, floppy). See replaced-url
1228[21:52:33] <Psil0Cybin> petn-randall, so what do i know what i am missing from the Fedora ISO file that Debian clearly is missing
1229[21:52:49] <Psil0Cybin> like how do i figure out the firmware image for the specific laptop because fedora core works out of the box but again i really just want debian since i am so used to it
1238[21:59:23] <jmcnaught> Psil0Cybin: the unofficial firmware ISO simply includes all the firmwares that are legally allowed to be distributed. You can run a command like "dmesg | grep -i firmware" to see which firmwares you might need that are missing.
1239[22:00:37] <jmcnaught> Psil0Cybin: there's a chapter of the install manual with more information about firmwares during installation: replaced-url
1240[22:00:42] <oxek> is there any public firmware that is not legal to distribute?
1241[22:01:01] <another> !firmware images
1242[22:01:01] <dpkg> Unofficial <netinst> and DVD images containing non-free Debian <firmware> packages are available from replaced-url
1244[22:02:01] <jmcnaught> oxek: there's a package b43-fwcutter that extracts the firmware from a file it downloads during installation because the firmware itself cannot be redistributed
1245[22:02:47] <jmcnaught> there might be others like it but that's one I know about.
1265[22:21:36] *** omero8213 is now known as S3xyL1nux
1266[22:29:46] <jmcnaught> f8e4: here's another one: ^search^replace on the command line. For example you could run "systemctl start foo" then on the next line "^start^status" would run the previous command substituting 'start' with 'status' resulting in "systemctl status foo"
1293[22:54:29] <karlpinc> jmcnaught, f8e4: I always use the emacs key bindings. Control-p for previous line, Alt-backspace for delete the last word, type "status"...
1332[23:36:39] <dka> I am trying to install debian on a ssd that alreayd have debian install
1333[23:36:59] <dka> so I went into the partitionning of the disk and when I have finished , i got this error
1334[23:37:45] <dka> Error informing the kernel about modifications to partition /dev/sda7 -- device or resource busy. this means linux wont know about ny changes you made to /dev/sda7 until you reboot -- so you shouldn't mount it or use it in any way before rebooting Error !!!
1335[23:37:58] <dka> I can ignore, cancel or go back
1336[23:38:04] <dka> what should I do now?
1337[23:38:07] *** Quits: Vizva (~Vizva@replaced-ip) (Remote host closed the connection)
1338[23:38:54] <jhutchins> dka: which relese? Which installer? What is your partitioning?
1339[23:38:57] <jhutchins> !paste
1340[23:38:57] <dpkg> Do not paste more than 2 lines to this channel. Instead, use for text: replaced-url
1341[23:39:01] *** debhelper sets mode: +l 1151
1342[23:39:03] <dka> i downloaded the latest release
1382[23:47:12] <jhutchins> dka: You're really not giving us to work with here. Tell us what you've done so far.
1383[23:48:28] <alexrelis[m]> Is there any way to enable multi-factor authentication with PAM with three factors to authenticate (password, google-authenticator, and biometric), but can still let you in when you can only provide two out of the three authentication methods? I wanted to just have password and google-authenticator, but I'm afraid I'll get locked out of my system if my phone runs out of battery.
1390[23:56:30] <alexrelis[m]> cluonbeam: Yeah, but I'm also scared of losing that, lol. If it had it so that I had 3 options to authenticate and only needed 2 I wouldn't be in danger of losing access to my system. But if I had a fingerprint to scan just in case my phone is out of battery or lost then it wouldn't be so bad.
1392[23:57:31] <cluonbeam> alexrelis[m]: keychain, man. That's where mine stays. If I lose my yubikey, I've also lost my OTHER keys, so that's not going to happen.
1393[23:57:55] <mutante> alexrelis[m]: you can definitely stack the methods and for each individual auth method you can define if it's "required" or "sufficient" or "optional". see "man pam.d" for the details what it means. but basically it's the difference between "ignore failure or not". What i don't see though is a real "any 2 out of 3".
1394[23:58:25] <cluonbeam> You could put two sufficient auth options in a row. The preferred first, and the backup second.
1395[23:58:38] <cluonbeam> If the preferred fails, it falls to the second, and if that succeeds, you're good.
1396[23:58:41] <Ede|Popede> ,file
1397[23:58:41] *** Quits: FuzzyByte (~weechat@replaced-ip) (Remote host closed the connection)
1398[23:58:42] <judd> (file <pattern> [--arch <amd64>] [--release <stable>] [--regex | --exact]) -- Returns packages that include files matching <pattern> which, by default, is interpreted as a glob (see glob(7)). If --regex is given, the pattern is treated as a extended regex (see regex(7); note not PCRE!). If --exact is given, the exact filename is required. The current stable release and amd64 are searched by
1399[23:58:43] <judd> default.
1400[23:58:57] <alexrelis[m]> cluonbeam: That sounds sufficient enough. Thanks.
1401[23:59:00] <cluonbeam> alexrelis[m]: What I would suggest is a yubikey as primary, and g-auth token as secondary. If you lose your yubikey, you're good. If your phone dies, you're good.
1402[23:59:01] <alexrelis[m]> mutante: And thanks as well.