this is #debianan IRC-Channel at freenode (freenode IRC service closed 2021-06-01)
0[00:00:00] <bigMouthCommie> running into disk space problems, randomly nuking log and cache files isnt helping. nuked wine out of /usr/lib, nuked files off of /home (separate partition) and bash still throws errors on tab-completion whining about disk space
1[00:00:16] <bigMouthCommie> any suggestions about how i can find the offending dir and clear up some space?
3[00:00:48] <bigMouthCommie> (still on debian 9 but the entropy is building so i'll be upgrading soon. i just need to be able to continue to use the machine a little while longer
69[00:42:28] *** Quits: izh_ (~izh@replaced-ip) (Remote host closed the connection)
70[00:42:30] <nvz> for some reason when I'm away from home I'm unable to connect to my server via mosh I get "mosh: Nothing received from server on UDP port 13380." but I am still able to connect over ssh
131[01:25:00] <zoredache> bigMouthCommie: kinda depends on what your actual end goal is. Those first two commands on that page would create a directory with a pretty bare-bones base install. If you needed to boot it, you would need to do more.
132[01:25:23] <bigMouthCommie> i want to install deb10 as a fresh install on a machine 2 miles away from me without leaving my chair
178[02:04:31] <uio1> Hi. On a thinkpad X61 here, XFCE, Buster. I set notification for 20% battery left and it just powered out at 8% with no notif. How can I fix this?
204[02:20:44] <detre> anyone know how to set a Linux server with lamp stack to respond to hostname in the browser address bar instead of the ip address on a local network?
210[02:25:49] <karlpinc> bigMouthCommie: There's an ssh option once you get the install going. But I'm not sure how you'd get it started. Of course if you're starting from a system that runs debootstrap then....
229[02:30:05] <karlpinc> detre: Sure. (It's probably running one of the dns servers mentioned above.)
230[02:30:24] <karlpinc> detre: The dns server does not have to be a separate box/vm/whatever.
231[02:31:45] <karlpinc> detre: There's also various "automatic" options that are supposed to make a network "easy". mdns and that autoconfig-ie thingie from apple come to mind.
232[02:32:11] <karlpinc> detre: IMO may as well run a dns server. YMMV.
233[02:32:17] *** Quits: silverballz (~hidden@replaced-ip) (Quit: One for all, all for One (2 Corinthians 5))
234[02:32:27] <detre> thanks for pointing me in the right direction
261[02:46:21] <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.
295[03:04:57] <RandomSerb> hello. I've a spare pc, just installed debian on it. Which software would you recommend to make this new server so that I can easily manage it as VPS's
296[03:05:50] <RandomSerb> as in, I want to be able to create VPSs for each of my projects. I mostly work with python/nginx.
313[03:14:39] <somiaj> if you just want multiple websites, would the multisite feature of nginx work?
314[03:14:51] *** Quits: frostschutz (~frostschu@replaced-ip) (Remote host closed the connection)
315[03:15:16] <somiaj> I personally use libvirt with kvm+qemu on my machine, and then all nginx is on the host is a ssl_proxy to the vms for each application I have on the server.
352[03:43:10] <mohammad-ghasemi> postfix: package is in a very bad inconsistent state; you should reinstall it before attempting a removal. But I can't remove it. Help!
353[03:43:18] <mohammad-ghasemi> and it gives and error about postfix, whenever I want to install any software.
364[03:49:25] <mohammad-ghasemi> themill, what is MTA?
365[03:50:12] <themill> !mta
366[03:50:12] <dpkg> A Message/Mail Transfer Agent (MTA) is a program that transfers e-mail in or out of a system, most often using SMTP. An MTA is usually accessed by calling /usr/sbin/sendmail and thus usually includes that binary. Therefore, in Debian, all MTAs (<exim4>, <postfix>, <qmail>, <ssmtp>, <nullmailer>, etc) mutually conflict. Ask me about <which mta>, <mta comparison> and <why an mta>. replaced-url
367[03:50:56] <mohammad-ghasemi> themill, no I just was doing the same thing as the linux video tutorial was doing. It installed postfix.
368[03:51:10] <somiaj> warsoul: what pakacge did you use to install cinnimon?
377[03:54:11] <somiaj> warsoul: try to remove the task-cinnimon-desktop and/or the cinnimon-desktop-enviorment meta packages, they should then allow any of the cinnimon parts you aren't using to also be automatically removed via the auto packages.
388[03:58:50] <mohammad-ghasemi> themill, it was just a practice of the video tutorial I was watching. In the video the guy installed it as an example to show how to use apt-get install command.
389[03:59:01] <themill> ugh, what a bad example to pick
390[03:59:14] <themill> mohammad-ghasemi: first step is probably "apt download postfix postfix-sqlite"
443[04:40:03] <bigMouthCommie> ImportError: Couldn't import Django. Are you sure it's installed and available on your PYTHONPATH environment variable? Did you forget to activate a virtual environment?
487[04:55:24] <bigMouthCommie> i installed python3-pip python3-django, cloned the git repo, cd'd to it, ran "python3 -m pip install pipenv" then "pipenv install" then the offending command
488[04:55:45] <themill> pipenv doesn't know about anything you installed with apt
490[04:56:01] <bigMouthCommie> ok. was i supposed to pip install django?
491[04:56:11] <themill> no, you should use --site-packages with pipenv
492[04:56:34] <themill> or you can install django inside the venv.
493[04:56:39] <warsoul> guys i installed debian 10 shrink i have a really nice pc dont know why is taking to long to start up debian
494[04:56:49] <warsoul> when i boot on windows it starts right away
495[04:57:07] <bigMouthCommie> themill: that seems really insightful, but its over my head
496[04:57:08] <themill> Bascially, this is yet another (bad) reinvention of package management from the python world... the 4th in the last decade
497[04:57:20] <bigMouthCommie> ok but i just want to run brutaldon
498[04:57:59] <bigMouthCommie> python3 -m pip install django gave me "Requirement already satisfied: django in /usr/lib/python3/dist-packages (1.10.7)"
499[04:58:02] <themill> bigMouthCommie: as I said, you either need to tell pipenv to use site packages or you need to install django inside the venv you have created
500[04:58:15] <somiaj> warsoul: systemd-analyze blame -- that should list the time of the services that take the longest to start.
501[04:58:20] <bigMouthCommie> i understand the english syntax, but i dont know the lines to type into a terminal
515[05:07:06] <vbn0rmvl> guys there is a bug in armitage the option "find exploits" not working with the newest update. Does anybody knows any solutions to fix that bug?
516[05:07:20] <warsoul> somiaj yes it has non-free
558[05:38:15] <somiaj> don't see any networking firmware missing. YOu may want to install new microcod, with an intel card, you'll want the intel-microcode package
585[05:42:43] <dpkg> [non-free] a component which contains software that does not comply with the <DFSG>. To add non-free packages to your packages index, ask me about <non-free sources>. To see which non-free packages are installed ask me about <non-free list>.
588[05:42:51] <dpkg> Edit /etc/apt/sources.list, ensure that the two main Debian mirror lines end with "main contrib non-free" rather than just "main", then «apt-get update». But bear in mind that you'll be installing <non-free> software. These may have onerous terms; check the licenses. See also <sources.list>.
589[05:43:02] <jh001> warsoul: do you have export numcpu="$(/usr/bin/lscpu | /usr/bin/pcregrep "^CPU.s.:" | /usr/bin/perl -nE'print /(\d+)$/')"
611[05:46:21] <somiaj> jh001: yea, in this case warsoul ran systemd-anaylize blame and got networking taking over a minute to start, which probably isn't entropy
613[05:46:43] <somiaj> but the easiest way to test if that is the case is not install haveged, but just move the mouse really fast and crazy like during boot, if that speeds it up, then you know that is the problem.
615[05:47:46] <user888> did you take a look in boot.log?
616[05:47:54] <jh001> somiaj: understood - it's a painless thing to try.
617[05:48:29] <warsoul> jh001 done
618[05:48:32] <warsoul> haveged installed
619[05:48:40] <somiaj> yea, I would just suggest the mose approach first to confirm heveged is needed, as it isn't something I would install if it isn't needed.
620[05:49:07] <awal1> what is the replacement for chromium-widevine?
621[05:49:12] <somiaj> (since it skipped my mind that buster was still having problems with this, as I thought it was mostly fixed with the CVE back when buster was testing)
622[05:49:27] <awal1> i am trying to vew sruff on netflix but i think widevine stuff are missing
626[05:49:39] <warsoul> somiaj how i install non-free
627[05:50:01] <warsoul> to be able to install microcode
628[05:50:14] <somiaj> awal1: I'm not sure, I personally just use google-chrome's package for widevine, as I find the scripts that rip it out of google chrome (same with pepperflash) don't seem to last relailably
629[05:51:02] <awal1> somiaj, I am using sid and widevine not available. hmmm, i know chrome works but i don't have it
630[05:51:11] <awal1> i may install it from official google repo
631[05:51:31] <awal1> but can it coexist with chromium in same session/user?
632[05:51:36] <annadane> warsoul, just add contrib non-free after main in the entires of your sources.list
636[05:52:44] <somiaj> awal1: I have both chromium and google-chrome installed, so that works, if you want to share profiles between the two, you may have to use google's chromium/chrome sharing for that.
637[05:52:45] <warsoul> how do i get to the source list
638[05:52:53] <annadane> as root, nano /etc/apt/sources.list
639[05:53:10] <awal1> somiaj, ok, thanks for the help, i will check that :)
640[05:53:12] <somiaj> but you either need the libary or to use google-chrome, and it seems there is not automatic tool to rip the libary out of google's package for ya
641[05:53:33] <annadane> make your changes, then control x, enter when it asks you what file name to save it under
642[05:53:33] <awal1> i may just install chrome, easier
643[05:53:38] <annadane> and then apt update
644[05:53:47] <awal1> it gets updates regularly and work fine?
645[05:54:11] <somiaj> yea, add the repo to your sources.list (or just download install the .deb and it will add it as part of the .deb)
646[05:54:26] <somiaj> then it will update with apt-upgrade like all your other packages when google makes a new package
683[06:03:25] <somiaj> I would ignore that, nouveau is already loading alot of firmawre correctly, sometimes the drivers look for older or firmware by other names until it finds the one on the system
739[06:28:51] <somiaj> warsoul mentioned something about signal strenght being way less in linux than windows, could be having trouble getting signal out to the network
740[06:29:10] *** Quits: bliv (~bliv@replaced-ip) (Remote host closed the connection)
741[06:29:22] <somiaj> I thought maybe firmware, but that all looked upto date. Maybe check the power settings on the wifi card to see if they are having trouble with power savinngs or something like that
742[06:30:01] <user888> yeah, it's simply not getting an IP. Are you able to try a wired connection?
790[06:53:16] <blackbart> My upgrade to buster seems to be hanging here: "update-initramfs: Generating /boot/initrd.img-4.9.0-4-amd64 \n W: mdadm: /etc/mdadm/mdadm.conf defines no arrays." Or perhaps update-initramfs supposed to take quite long?
791[06:53:35] <user888> damn. can you paste "ip -a" once again
799[06:57:49] <ensamvarg> Good morning. The new kernel in version 5.2 from Debian/testing halts on bootup on my HP ProLiant ML110 G6 with Xeon X3430 cpu. Anyone having problem with that kernel too? I already filed a bug report.
866[07:25:37] <user888> hmm not sure about that. can't really get into that now as I need to go. I would uncomment those lines we just changed and either disable the interface from the GUI if you have a desktop environment installed or try "ip link set eth0 down" if you only have a terminal.
870[07:27:50] <bigMouthCommie> themill: yo. i have been breaking my teeth on this for hours. i messaged the brutaldon dev and am going to do... something else for awhile. thanx for helping
944[08:08:41] <Haohmaru> it doesn't say whether i should run it while the linux on that device is running, or from a live image with the problematic device not mounted or so
946[08:10:47] <jelly> Haohmaru: I don't have experience with btrfs, so in general, unless you see anything about -c or -cc option in either fsck or fsck.btrfs manual, be wary of that advice
980[08:39:50] <Haohmaru> so what can i do? shut it down till i get a new HDD, then run a live image, dd or ddrescue the data onto the new one, and continue from there? can the btrfs partitions be "converted" to ext4 then?
1002[08:53:04] <Haohmaru> so then, what are my options? partition the new drive to ext4 and copy the files? i know some things might break, like fstab, but perhaps i would fix that.. what else could break?
1031[09:03:03] <Mo7Qt0> If you have any questions about backing up with tar, I have links on video in youtube how can create such backups, but in this video russian-language speaker
1032[09:03:30] *** Quits: pity (~pity@replaced-ip) (Remote host closed the connection)
1039[09:04:51] <alkisg> Haohmaru: maybe rsync instead of cp if you have failures while copying; and ask about "debsums" later to verify things got copied ok
1040[09:05:59] <alkisg> debsums -s => checks the packages files md5sums against the original md5sums; to see if there's corruption
1041[09:06:00] <Mo7Qt0> u can you can see video without sound and copy most actions :D
1048[09:09:28] <Mo7Qt0> but u can explain about grub there (replaced-url
1049[09:10:54] <Mo7Qt0> oh, can anybody help me with my problem?)
1050[09:10:57] <Mo7Qt0> Have a server with debian 9.9 and postfix+dovecot+amavis-new.
1051[09:11:11] <Mo7Qt0> The problem is that amavis trying to send mail to himself mailbox witch missing (amavis@replaced-ip) an attempt to send mail is carried out by a script /usr/sbin/amavisd-new-cronjob witch serves for training spamassasin.
1053[09:11:19] <Mo7Qt0> I tried to create a mailbox with name amavis, but after creating mailbox, mail stops going to recipients.
1054[09:11:21] <VicMackey> [debian-sid] Hello! I am trying to map a key in a Logitech USB keyboard. I got the vendor and product ID from 'lsusb' and produced the following line for /etc/udev/hwdb.d/90-custom-keyboard.hwdb:
1055[09:11:27] <Mo7Qt0> Not understandable - is spamassassin learning and how to fix this error?
1057[09:12:38] <VicMackey> evdev:input:b0003v046DpC31C* \n KEYBOARD_KEY_70065=b . But when I do #systemd-hwdb update I get ":2: Property expected, ignoring record with no properties". What am I doing wrong? Thanks in advance :9
1116[09:58:50] <tdn> What is the best way to have something (backup) run periodically on a laptop that is often without network and not on 24x7. Should I put it in /etc/rc.local and let it start 10 min after boot up? Any better idea? I want this to run as close to once a day as possible
1117[09:58:56] *** Quits: tryte (~tryte@replaced-ip) (Remote host closed the connection)
1124[10:01:41] <mia_> I've problems with latest debian , I tried to do a apt-get update; apt-get dist-upgrade but it hangs on /etc/kernel/postrm.d/zz-update-grub
1204[10:47:55] <diogenes_> aaii, can you: ping deb.debian.org
1205[10:48:09] <annadane> does anyone know offhand why after an install of a minimal MATE (marco + some things added manually) i can't right click on the desktop and there's no icons?
1206[10:48:14] <annadane> what do i have to install?
1207[10:48:26] <aaii> diogenes_, 64 bytes from 149.20.4.15: icmp_seq=1 ttl=40 time=284.469 ms
1258[11:09:16] <SonOFaza> virt-manager keeps saying it has no permissions. i tried various combinations of "chown -R user:libvirt" and "chmod -R 0770" without success. whats so special about these libvirt permissions?
1346[12:05:51] *** milos_ is now known as RandomSerb
1347[12:05:54] <RandomSerb> hello. I've a spare pc, just installed debian on it. Which software would you recommend to make this new server so that I can easily manage it as VPS's
1348[12:05:56] <RandomSerb> as in, I want to be able to create VPSs for each of my projects. I mostly work with python/nginx.
1349[12:06:16] *** Quits: pity (~pity@replaced-ip) (Remote host closed the connection)
1350[12:07:32] <RandomSerb> this said, I don't know how to start. I know I need something like vmware, but clueless as to how all that would go. As in, do I need an OS and which one below vmware, how to configure it, how to create ne machines, how to o assign local IPs to said machines, and , in the end, how to access them
1351[12:08:58] <mirazi_heket> debian8, anyone experienced this issue? Failed to list units: Activation of org.freedesktop.systemd1 timed out
1352[12:08:58] <mirazi_heket> i just typed 'systemctl'
1353[12:10:01] *** Quits: dselect (~dselect@replaced-ip) (Quit: ouch... that hurt)
1378[12:26:59] <Reventlov> Hello. Do you know how to test email integration of unattended upgrades? Right now, I think unattended upgrades are working, but as nothing has to be upgraded no email is sent, and I'm not sure mail is working
1379[12:28:41] *** Quits: madage (~madage@replaced-ip) (Remote host closed the connection)
1460[13:23:36] <ksk> Reventlov: try sending a mail (to root) - and take a look what happens? (like so maybE: "echo test" | mail -s "this is a test" root
1462[13:24:25] <ksk> I am not sure how unattended upgrades are implemented, but if, for example, any cronjob creates any output, this output will be mailed (to the cronjob running user, if I remember correctly)
1463[13:24:45] <ksk> as systemd is a thing though now, I am not sure if unattended upgrades maybe also switched to systemd-timers ;)
1472[13:31:04] <ksk> might be 65507 is the limit in ping/ the ICMP protocol (guessing wildly here..), and the 50k limit is something like "applied by your kernel/setup, because $REASONS"
1473[13:31:25] <Reventlov> ksk: well it works, so I guess that's ok
1474[13:31:58] <ksk> Reventlov: I do not know of an easy method to "make apt have an upgrad pending", so I think you are good then ;)
1475[13:32:33] *** Quits: BCMM (~BCMM@replaced-ip) (Remote host closed the connection)
1482[13:34:51] <uet6252> ghormoon: size 500000 may trigger a general sanity check while size 90000 may pass that one and the (still too big) packet triggers an error at socket level
1486[13:36:29] <uet6252> ghormoon: max. IP datagram size is 65536. minus 20 for the IP header, minus 8 for the ICMP header makes 65508 though, not 65507.
1515[13:51:17] <jelly> ghormoon: what's the use case sending such large packets with ping?
1516[13:52:34] <ghormoon> jelly: i was just trying how much i can send (to see if the repsonse time goes up) without having to open up any higher protocol (we had very slow responses on the webserver and this was first idea), so i was shooting up numbers to see what it will allow and it surprisedme that after few tries (going down) it told me the limit
1517[13:52:51] <uet6252> ghormoon: just checked to source. one check happens at option handling in main(), another two in ping4_run().
1518[13:53:15] <ghormoon> I'm just reafding in, one is in the case s: directly, righT?
1534[13:57:39] <ghormoon> it's just about being more user frinedly by telling the limit right away if the user is brain-off like me now and not even thinking about packet size limit
1535[13:59:02] *** debhelper sets mode: +l 1516
1536[13:59:44] <Habbie> yes, you'd file this in the debian bugtracker, or with iputils upstream
1537[14:00:02] <ghormoon> is there upstream? i did find only this repo
1544[14:02:46] <ghormoon> wel,, this is so minor thing it's better to file upstrream and it will arive to debian sometime. less work for debian maintainers
1611[15:00:29] <BCMM> colo-work: another problem with ubuntu is occasional attempts to make changes that are, basically, just an attempt to differentiate their product from the competition
1612[15:00:44] <BCMM> i.e. product identity > usability
1613[15:00:46] <colo-work> yeah, that's the much more frequently observed problem
1633[15:08:59] <jelly> EdePopede: if you have a more friendly alternative that factoid canbe updated.
1634[15:10:05] <EdePopede> jelly, i've already reached the point to call the web of the 90s userfriendly. browsers had a setting to not autoload images and most sites worked perfectly without that disease called javascript.
1652[15:14:02] <EdePopede> can't be conservative enough in using these tools without having firefox running wild after a few days somewhere. had to restart X yesterday and already killed the browser.
1660[15:16:41] <EdePopede> i was just thinking of preparing some profiles, zip them and recreating them in a wrapper script whenever i do something like going to imgur or so
1661[15:17:47] <EdePopede> i literally allow every single host temporarily, takes some time sometimes, but this also shows me what sites are ready for the drain. (same for cookies, OAuth anyone?)
1675[15:25:43] <EdePopede> sin(30°) = 0.5 ; it's just the same.
1676[15:25:47] <ksk> wwilliam: eh, man strace? :P
1677[15:26:06] <wwilliam> ah ok checking
1678[15:26:09] <wwilliam> thanks
1679[15:26:44] <ksk> "The name of each system call, its arguments and its return value are printed on standard error
1680[15:26:47] <ksk> "
1681[15:27:12] *** Quits: bqq (~quassel@replaced-ip) (Remote host closed the connection)
1682[15:27:13] <wwilliam> OK checking
1683[15:27:52] <ksk> wwilliam: that line basicly means something is using your systems locales, to do locales-magic. - if it was "something bad" you would normally see a more highlighted error so to say (like: "I can not open XY..." )
1684[15:28:22] <ksk> but the explanation in man strace seem to be quite nice, take a look :)
1692[15:33:23] <colo-work> does anyone of you have long-term experience (i.e., across multiple Debian releases) taking care of a redundant glusterfs setup, and can share a few impressions?
1693[15:35:11] <jelly> not even shortterm. openshift vendor/implementor was unable to log in and use glusterfs nodes on debian...
1779[16:19:16] <Soo_Slow> half-offtop question: what s the better (in terms of performance/speed) way to periodically verify integrity of large amount of varying sized files - via md5 or via crc32? or doesnt matter (difference isnt that noticable)?
1789[16:20:42] <Soo_Slow> Habbie, so md5 is the way to go? Or there are alternatives?
1790[16:20:45] <Habbie> uio, i bet it will show up in news on september 7
1791[16:20:54] <Habbie> Soo_Slow, sha1, sha2, there's a bunch
1792[16:20:59] <Jubei> Hello everyone. I am introducing some random gibberish into my apache2 virtual host config file and despite that apache2ctl configtest returns "Syntax OK". How can I troubleshoot htis?
1793[16:21:04] <Habbie> Soo_Slow, you could run 'openssl speed' to figure out how fast each of them are
1794[16:21:15] <uio> Habbie, Okay, it's more of log than a calendar.
1795[16:21:21] <jelly> vroom
1796[16:21:26] <Habbie> uio, that's my impression, yes
1797[16:22:15] <Soo_Slow> Habbie, but from which to choose? I mean - there are plenty of different hash algorithms. Which ones are viable to detect corruption?
1798[16:22:15] <uio> Cool. There are even Stretch point releases. Didn't know that.
1800[16:22:33] <Habbie> Soo_Slow, most of them are; sha256 appears to be the common choice these days
1801[16:22:58] <Soo_Slow> Habbie, but isnt it slower than md5?
1802[16:23:02] <jelly> preparing and testing debian-installer images is probably the most time consuming part of point releases. Just bumping some packages into the main repo seems less problematic
1803[16:23:08] <Habbie> Soo_Slow, it is
1804[16:23:12] <uio> Are there any risks associated with the point releases, or are they (unlike upgrades) more stable than what preceded?
1805[16:23:23] <Habbie> Soo_Slow, on my system, anyway
1806[16:23:27] <tds> Jubei: what file were you adding rubbish to, and do you see that file listed in `apache2ctl -S`? note that with the Debian scripts for httpd you might have a vhost in sites-available but not symlinked in sites-enabled, so it won't be read by httpd
1818[16:27:35] *** Quits: Error451 (~R@replaced-ip) (Remote host closed the connection)
1819[16:28:58] <uio> If I upgrade to 10.1, might I experience broken things or is it pretty safe?
1820[16:31:34] <Jubei> tds I was adding to my single virtual host config file (and it IS symlinked properly from sites-available)
1821[16:32:10] <Jubei> tds the people at #httpd say that my apache isn't finding any virtual hosts so it's defaulting to the server root directory replaced-url
1822[16:32:15] <vlt> uio: Depends on the version you're updating from.
1831[16:35:14] <uio> vlt, Sorry to press the question, but are they different from the upgrades? I don't want to repeat the problems I had with 9.9 to 10.0...
1832[16:36:06] <uio> I just learned Debian can run on the FreeBSD kernel. Cool.
1869[16:57:04] <rwp> tgunr, What you say is either impossible or a bug in init, for init to have a dead child process that it has not reaped. Are you sure the dead process is a child of init?
1879[16:59:30] *** Quits: P1ersson (~P1ersson@replaced-ip) (Remote host closed the connection)
1880[16:59:30] <dpkg> The <su> command changed in buster: it no longer overrides the PATH variable so commands normally available to root are not found. Put ALWAYS_SET_PATH yes in /etc/login.defs to get an approximation of the old behavior. See replaced-url
1881[16:59:40] <rwp> tgunr, Can you cut and paste the process list into a pastebin? Such as: ps -efH | nc termbin.com 9999
1883[17:00:57] <rwp> tgunr, 'su' simply switches your user but keeps your environment the same. Probably you should be using either 'su -i' or 'sudo' for such things instead.
1888[17:02:32] <rwp> tgunr, Sorry. To terse of output. Not enough information there for me. Not familiar with psfind output. However did that say it is locked in the 'D' state?
1895[17:04:09] <rwp> Most often uninterruptible D state sleep is due to a process waiting for I/O from an I/O device which is itself stuck and not rsponding.
1901[17:05:50] <rwp> At one time (perhaps not the way things are done now) the kernel would set up memory for a DMA action from a device driver. The kernel then waits for the DMA to complete. The kernel cannot kill the process and reclaim resources until the DMA completes. The process is listed in the D state. And if the DMA never completes then ...
1902[17:06:28] *** Quits: agosto (~paul@replaced-ip) (Ping timeout: 272 seconds)
1903[17:07:06] <tgunr> I was attempting to stop a zfs transfer using systemctl
1904[17:07:21] <nononsense41> jelly: if you happen to find something out about the bizarre installation image, kindly share it on forums.debian.org in a post titled "Something's not quite right with the installation image.." within the Help>Installation section
1905[17:07:34] <tgunr> the process was started by a cron timer
1932[17:15:25] <tgunr> only once scince 1980 have i rm /*
1933[17:15:27] <rwp> jhutchins_wk, Traditionally it was "su -" but time passes and I think "su -i" is the modern preferred syntax form of it. Also "sudo -i" might be better.
1939[17:17:56] <greycat> There is actually no standard definition for su, and its behavior is entirely system-specific. On Debian, the behavior has always been "change the PATH variable" up until buster. On HP-UX, it always changed PATH. On Red Hat and OpenBSD, it did not.
1940[17:18:05] <rwp> Ah! I conflated su and sudo. It is "sudo -i" or "su -l". Sorry for the misfire.
1941[17:18:59] <greycat> The alarmingly common suggestion to "just use su -" comes largely from Red Hatters, as far as I can tell. The unification of the Debian codebase with the Red Hat codebase has unsurprisingly led to an influx of Red Hat suggestions for Debian users. And yes, you can do that if you like.
1948[17:20:05] <rwp> Well... My finger memory of su and sudo was learned on HP-UX. And as we know HP-UX 10.20 was the ultimate release of it.
1949[17:20:16] <greycat> Yes, and su in HP-UX changes the PATH variable for you.
1950[17:21:15] <rwp> I would need to fire up either my gecko or my J5000 in order to check it. Too bad there isn't an easy way to emulate it in order to have a running system available as a VM in software!
1967[17:25:39] <rwp> It looks like the momentum is for "su -" rather than modernizing with "su -l". I will return to the traditional "su -" then when that is the topic then.
1968[17:25:44] <jhutchins_wk> Yeah, I think I got it from "Running Linux" by O'Rilley
1972[17:26:28] <greycat> I prefer modernizing with "sudo -s", but on systems where sudo is not installed, I have edited /etc/login.defs to restore the sane behavior.
1973[17:26:55] *** Quits: Megaf (uid155773@replaced-ip) (Quit: Connection closed for inactivity)
1976[17:28:58] <rwp> The problem with "sudo -s" is that it leaves $HOME set to the user. And then bad behaving programs (I am looking at you aptitude, but there are others) leave files behind owned by root.
1979[17:29:44] *** Quits: encod3 (~encod3@replaced-ip) (Remote host closed the connection)
1980[17:29:46] <greycat> FreeBSD's su doesn't change PATH. OpenBSD's su looks like it *does* change PATH now... I'll need to check older pages to see if this is a change.
1981[17:30:34] *** Quits: VadPerevad (~vadpereva@replaced-ip) (Remote host closed the connection)
1982[17:30:54] <greycat> Hmm, even back to obsd 4.0 it changes PATH for UID 0.
1987[17:32:59] <greycat> Here we go. OpenBSD changed from the FreeBSD/RedHat behavior (do not change PATH) to its current behavior (change PATH for UID 0) in OpenBSD 3.0 (December 2001).
1993[17:36:21] <rwp> I have always put /usr/sbin:/sbin into my PATH as a normal user. I think mostly for 'ping' back in the day. I forget now what commands were in sbin that I wanted in my normal environment now. Wasn't ping in one of the sbins back then? If not ping then at least one of the other useful networking commands.
1994[17:36:41] <greycat> ifconfig was probably one you wanted
2004[17:39:05] <greycat> modinfo is also OK to run as a regular user, but lives in /sbin
2005[17:39:10] <rwp> FTR I agree that /etc/login.defs now needs to have ALWAYS_SET_PATH yes for sane behavior. I had not really looked at the su PATH change until now. I had just noted that it had changed to not set PATH anymore.
2011[17:40:47] <greycat> but it probably won't get fixed
2012[17:41:29] <rwp> As a workaround someone could wrap su with a wrapper that sets PATH before calling su.
2013[17:42:11] <greycat> the wrapper would have to guess which user you're switching to, and only act when switching to root, which makes it nontrivial
2014[17:42:36] <greycat> unless you just put sbin-dirs in for *everyone*, which is acceptable to me
2015[17:43:08] <rwp> I always add the sbin-dirs to my normal user PATH. It's fine.
2016[17:43:13] <rwp> In face to face conversations I often proclaim that every Debian release is incrementally more buggy than the one before it. And then sigh deeply...
2017[17:45:14] <jhutchins_wk> rwp: Which is wrong.
2018[17:45:43] <jhutchins_wk> rwp: There have been several releases that were overall significantly better than the previous one.
2026[17:47:33] <rwp> I am adding notes about 'su' to my local Buster 10 errata file right now...
2027[17:47:57] <greycat> The interface name change thing is pretty horrible too.
2028[17:48:00] <jhutchins_wk> rwp: What was your first Debian release?
2029[17:48:03] <embden> jhutchins_wk: I have a small experience with Debian I used 8-9, 10th was unbearable for me, so now I am a mate guy. But I use it for work-in-progress things
2030[17:48:38] <rwp> The first Debian system I installed was Potato. But that was during the Woody release candidate time period and it was almost immediately upgraded to Woody.
2035[17:50:24] <embden> rwp: thanks, I thought it is a mistake in my case, so I reinstalled with netinst media and PATH there also doesn't include /usr/sbin for root.
2036[17:50:27] <lf94> Maybe debian is too large for its own good
2037[17:50:28] <rwp> Lamont grabbed me by the nose and dragged me through the Potato installation. On an HP x4000 workstation.
2038[17:50:33] <greycat> !buster su
2039[17:50:33] <dpkg> The <su> command changed in buster: it no longer overrides the PATH variable so commands normally available to root are not found. Put ALWAYS_SET_PATH yes in /etc/login.defs to get an approximation of the old behavior. See replaced-url
2040[17:51:02] *** Quits: mibo (~mibo@replaced-ip) (Remote host closed the connection)
2041[17:51:12] <greycat> !factinfo buster su
2042[17:51:12] <dpkg> buster su -- created by greycat <~wooledg@139.137.128.43> at Thu Jul 11 19:56:26 2019 (46 days); it has been requested 22 times, last by greycat, 39s ago.
2045[17:51:58] <greycat> does factinfo not show *changes* to the factoids?
2046[17:52:14] <uio> vlt Thanks!
2047[17:52:59] <embden> greycat: what is the reason to do so...I mean users switch to root to make their work more comfortable...why to exclude the directories from the PATH?
2048[17:53:13] <greycat> Did you miss the ENTIRE discussion we just had?
2051[17:54:02] <rwp> embden, The general consensus is that su *should* set PATH and that not doing so is buggy behavior introduced in Buster 10.
2052[17:54:06] *** zalt__ is now known as zalt
2053[17:54:38] <rwp> Which greycat has been saying all along. And I was convinced to that conclusion through the discussion just now.
2054[17:54:39] <greycat> If you literally want me to tell you the REASON they changed it: apparently they decided to use the same source code that Red Hat is using, instead of the source code they had been using up through Debian 9. And the Red Hat source code behaves in this way. And they didn't bother overriding it.
2056[17:54:45] <jhutchins_wk> embden: su is flexible, and different people use it differently. Originally (as far as Linxu goes) su did not chnage the environment, su - did. Because new users were constantly ignorant of that, a kludge was added that caused su for root to change the user environment. That has finally been corrected.
2057[17:54:58] <greycat> jhutchins_wk: lies.
2058[17:55:47] * jelly flexes su (to see if it breaks into s and u)
2059[17:55:51] <tgunr> so much easier to just login as root
2060[17:56:26] <jelly> tgunr: except you're actively prevented to log in as root into GUI
2061[17:56:27] <jhutchins_wk> greycat: Ya, well, you came down out of a different tree than I did.
2062[17:56:36] <jelly> ...in some cases
2063[17:56:42] <tgunr> dont need a gui as root
2064[17:56:46] <jhutchins_wk> jelly: Which is a good thing.
2065[17:56:55] <rwp> Never log into the graphical login manager and start X as root. Millions of lines of potential vulnerabilities in the code.
2066[17:57:07] <tgunr> been logging in as roott for almost 40 years
2067[17:57:30] <tgunr> YMMV
2068[17:57:31] <jhutchins_wk> If you NEED a GUI you probably shouldn't have root access.
2069[17:57:36] <rwp> Is X 40 years old now?
2070[17:57:39] <greycat> I log in as root all the time *on the console* or *on ssh*. Never through a graphical Display Manager.
2071[17:57:40] * tgunr nods
2072[17:57:41] <humpled> :þ
2073[17:57:58] <tgunr> Started with System III
2074[17:58:03] <humpled> but can't read the release notes
2079[17:59:33] <tgunr> running on a LSI-23 with 768KB RAM :)
2080[17:59:56] <karlpinc> I'm using an X client and server on different machines. I know there's other (qt?, etc.) protocols that do this but so far this has worked. I'm reluctant to try something new that might limit which applications I can use, depending on the application toolkit. Do I have options (aside from vnc) that give me an underlying protocol I can use with all my apps?
2082[18:00:17] <rwp> Remember that just because someone can do a thing does not mean that someone should do a thing. Sure it is possible to log into X as root. But one should never do it.
2083[18:00:35] <rwp> With great power comes great responsibility.
2085[18:00:47] <uio> rwp, Where does the security issue come from?
2086[18:01:03] <rwp> karlpinc, I did not discern a question in there. What did I miss?
2087[18:01:30] <rwp> uio, The security issue comes from literally millions of lines of code that was not written with root operation or security in mind.
2088[18:02:02] <embden> also, does somebody currently work on SELinux support in Debian?
2091[18:02:16] <uio> rwp, Probably a dumb question, but does X give more function? Is it not just a change in form?
2092[18:02:22] <jelly> embden: support is there. Policies are basically nonexistent.
2093[18:02:41] <rwp> uio, X give more function? More than what? What are you comparing it against?
2094[18:02:51] <jhutchins_wk> embden: Somebody does. It used to be one of the Federal security agencies - nsa? - these days it's mostly RedHat employes.
2096[18:03:40] *** iulian is now known as Guest67241
2097[18:03:44] <rwp> Every RHEL client I work with (CAD/EDA applications) installs RHEL and disables SELinux. Because the vendors don't keep up with it.
2098[18:03:45] <jelly> jhutchins_wk: redhat employees working on Debian?
2108[18:05:12] <uio> rwp, And how can display cause vulnerabilities?
2109[18:05:15] <Henry151> i screwed up super royally and i need halp XD replaced-url
2110[18:05:43] <jhutchins_wk> jelly: Not on Debian, on Linux.
2111[18:06:01] <jelly> jhutchins_wk: user did not ask for "on Linux"
2112[18:06:06] <uio> rwp, X doesn't have any additional power, so how could more go wrong?
2113[18:06:06] <Henry151> when i got to sudo apt install apt-transport-https , it said "you're about to do something potentially horrible, type "Yes, do as i say"
2114[18:06:07] <jhutchins_wk> jelly: systemd is a RH project, but it's used by debian.
2115[18:06:09] <Henry151> and i did it
2116[18:06:16] <Henry151> now, I have no apt
2117[18:06:37] <jelly> !pal Henry151
2118[18:06:38] * dpkg points at Henry151 and laughs uproariously
2120[18:06:50] <rwp> uio, If I knew of a specific instance I would be filing a bug rated security vulnerability and that specific thing would get fixed.
2121[18:06:54] <uio> jelly, lol
2122[18:07:07] <Henry151> lmao
2123[18:07:10] <rwp> uio, But literally millions of lines of un-audited code running as root should be sending chills down your spine.
2124[18:07:18] <Henry151> so I went and downloaded a .deb from replaced-url
2126[18:07:36] <Henry151> can someone halp me fix it all?
2127[18:07:51] <jelly> Henry151: which debian release is this supposed to be?
2128[18:07:59] <uio> rwp, I guess I just thought that all those lines only worked around display and not actually extra functions, so there shouldn't be increased risk.
2129[18:08:16] <Henry151> jelly: it's supposed to be stretch
2130[18:08:29] <Henry151> but if I have to update it to the newer version, I won't mind, now is as good a time as any
2131[18:08:29] <uio> rwp, But I don't really know what I'm talking about :)
2149[18:11:12] <Henry151> this overall debian installation, I have devoted like, weeks to it
2150[18:11:13] <jelly> and maybe next time heed the warning
2151[18:11:18] <Henry151> and it's a currently running media server
2152[18:11:21] <Henry151> with a bunch of users.
2153[18:11:29] <jelly> do you have a backup?
2154[18:11:29] <Henry151> it's still working as far as serving my media!
2155[18:11:32] <Henry151> nope
2156[18:11:37] <jelly> why not?
2157[18:11:40] <rwp> uio, That is actually the problem. You are not thinking about how it can go wrong. But neither were the authors thinking about how it could go wrong running as root either! That is itself the problem.
2162[18:12:24] <Henry151> i suppose i could and should have just backed up everything but the media
2163[18:12:30] <greycat> Henry151: you've definitely got some buster files from *somewhere*. If you didn't put buster sources in /etc/apt/sources.list then perhaps they are in /etc/apt/sources.list.d/* or something.
2164[18:12:49] <jelly> Henry151: at some point in the past you had enabled repos for a release NEWER than stretch (debian 9) and had many packages upgraded
2169[18:13:26] <uio> rwp, I mean, there has to be a lot more to it than not thinking there would be a problem. Drinking tea while logged in as root is, I assume, quite safe, but I've never thought about it much.
2170[18:13:37] <rwp> If so then I think you can download the correct apt for Debian 9 and downgrade back down to it.
2171[18:13:57] <jelly> rwp: they have libc6, dpkg and base-files from debian 10.
2172[18:13:58] <greycat> No, he can't downgrade. He could finish UPgrading to buster.
2173[18:14:04] <rwp> uio, Bring this up again later. Let's try to help Henry at the moment. :-)
2174[18:14:14] <jelly> and very likely lots more.
2175[18:14:22] <uio> rwp, Sure :) I guess it is a support channel after all.
2176[18:14:44] <jelly> Henry151: and you probably had apt from debian 10 as well
2177[18:15:00] <Henry151> I would like to upgrade
2178[18:15:12] <Henry151> but now that i ruined my apt i'm dying over here.
2179[18:15:16] <rwp> It's not clear to me from reading what exactly Henry151 did to get here. "sudo dpkg -i apt_1.4.9_amd64.deb" was that all? If so then I think none of those other things happened.
2180[18:15:43] <greycat> rwp: he's got base-files from buster. That did not happen with stretch sources and a hand-downloaded apt*.deb file.
2186[18:16:34] <greycat> either he pointed sources at buster, installed some crap, and then pointed them back to stretch... or he did a LOT more *.deb files than just one
2190[18:17:22] <Henry151> so, previously, i had changed my /etc/apt/sources.list from "stretch" to "stable" and ran "apt update" and "apt upgrade" and then, tried to go backwards, changed /etc/apt/sources.list back to saying "stretch" and done the apt update and apt upgrade again
2191[18:17:30] <Henry151> this was a month ago or so
2192[18:17:32] <greycat> Yeah, that won't work.
2193[18:17:37] <Henry151> everything was working fine after that operation
2194[18:18:01] <greycat> Your best bet at this point is to point sources back to buster (not "stable", as that is a moving target), and finish upgrading to buster.
2195[18:18:17] <rwp> After that operation of changing to 'stable' and upgrading from that point on you were running Buster 10. Since 'stable' is an alias for Buster 10 since the release to Stable of Buster 10.
2196[18:18:21] <jelly> there is no going backwards.
2199[18:19:17] <jelly> going forwards to buster seems the most prudent move. Of course it might break working software/services that is not part of Debian itself, and relies on Debian 9 components
2200[18:19:17] <rwp> And now that 'apt' is broken you will need to use either 'apt-get' or 'aptitude' which are separate binaries. Hopefully nothing in the libs behind are broken.
2203[18:19:41] *** h0n3ycl0ud is now known as ac1dh0n3ycl0ud
2204[18:19:49] <Henry151> when i try apt-get it says "apt-get: relocation error: /usr/lib/x86_64-linux-gnu/libapt-private.so.0.0: symbol _ZN3URIcvNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEB5cxx11Ev version APTPKG_5.0 not defined in file libapt-pkg.so.5.0 with link time reference"
2205[18:19:49] <jelly> /usr/bin/apt-get is part of apt package.
2206[18:20:15] <jelly> Henry151: pastebin the output of dpkg -l apt libapt\*
2265[18:34:40] <Henry151> so my previous sources.list had main contrib non-free, any reason to not do that here?
2266[18:35:24] <Henry151> and i imagine i should remove that plex source from /etc/apt/sources.list.d as well, at least for the moment until i have things fixed?
2267[18:35:46] <Henry151> and then i do an apt update?
2270[18:36:08] <Henry151> this all seems fairly obvious but i just want to make sure i'm following the best guidance possible because y'all know what you're doing and are giving me advice
2271[18:36:09] *** Quits: freedom (~freedom@replaced-ip) (Remote host closed the connection)
2279[18:36:52] <Henry151> and just a side note, I work from home and do customer support on the phone, so if i go silent for 5-10 minutes at a time here or there, it likely means i got a work call and had to take it.
2280[18:38:27] <jelly> Henry151: if you had reason to add contrib non-free in addition to main before (firmware?) the same reasons probably still apply
2289[18:42:23] <Henry151> i'm going to try to apt update now.
2290[18:42:45] <Henry151> seemed to work... try apt upgrade now i imagine?
2291[18:43:14] <Henry151> i'm just trying to be extra cautious and make sure to follow the guidance correctly here because i am afraid of screwing this up.
2321[18:56:08] <Henry151> now i just uncomment deb replaced-url
2322[18:57:01] <Henry151> from the /etc/apt/sources.list.d/plexmediaserver.list , and run apt update, apt install plexmediaserver ? thank you so much for rescuing me, #debian, i love you guys/gals.
2371[19:18:33] <elwisp> had to do hard reboot on my machine just now. Says i need to run fsck. fsck says i still have errors. i have tried backup superblock
2372[19:19:28] <karlpinc> elwisp: You probably want to reboot and add the following parameter to the kernel command line: fsck.mode=force
2375[19:20:10] <karlpinc> elwisp: man systemd-fsck
2376[19:20:13] <elwisp> i run debian on a desktp. And i have it encrypted. Right after typing in my password i get Resuming from hibernation. Which is wierd then just straign in to recovery journal get some buffer I/O error but i dont think the drive is toast. Its a quite new SSD Maybe a year old
2377[19:20:32] *** Quits: oish (~charlie@replaced-ip) (Remote host closed the connection)
2392[19:22:19] <elwisp> print_req_error critical medium error dev nvme0n1
2393[19:22:34] <karlpinc> elwisp: I always use -y, because it is never worth my time trying to figure out a perfect way to repair every broken block. (man fsck for what -y does first.)
2394[19:22:44] <greycat> !factinfo forcefsck
2395[19:22:44] <dpkg> forcefsck -- created by gsimmons <~gsimmons@gsimmons.org> at Wed Apr 15 07:11:10 2009 (3786 days); it has been requested 143 times, last by greycat, 12s ago.
2396[19:23:26] <elwisp> just get "unable to set superblock flags on /dev/mapper....."
2397[19:23:34] <karlpinc> greycat: ("always" as long as systemd started breaking things.)
2398[19:24:01] <elwisp> would changing the kernel command line really do any difference?
2399[19:24:05] <karlpinc> elwisp: I don't know if the intramfs has unencrypted the device yet.
2400[19:24:29] <karlpinc> elwisp: You might have to cryptsetup (or something) before fscking.
2401[19:24:34] <elwisp> hmm
2402[19:24:35] <elwisp> good point
2403[19:25:08] <greycat> dpkg, systemd-fsck is <reply>Under systemd, <forcefsck> no longer works. You can force all file systems to be checked by adding "fsck.mode=force" to your kernel parameters from the GRUB menu (see systemd-fsck(8)). There doesn't appear to be a way to force *just* the root file system to be checked.
2404[19:25:09] <dpkg> greycat: okay
2405[19:27:01] <elwisp> gotcha. i was running fsck on the wrong thing i thnk cryptsetup says it has setup the cruptcontainer however
2406[19:27:13] <elwisp> but it complains about the lvm volume
2407[19:27:27] <elwisp> i think i need to boot into a live environment to debug this
2411[19:28:38] <Henry151> thank you all again one more time. I have plex working now and fixed my debian install thanks to this great chatroom. I am going to idle in here and try my best to use my knowledge to help others as much as possible over the coming weeks, to pay it back. Thank you.
2422[19:35:20] <elwisp> got same error when running from live usb
2423[19:35:26] <elwisp> Hmm maybe this disk is toast
2424[19:36:04] <tgunr> not unusual, had 3 go bad on me this year all about a year old
2425[19:36:12] *** Quits: radkos (~radkos@replaced-ip) (Read error: No route to host)
2426[19:36:24] <elwisp> should have noticed something
2427[19:36:35] <elwisp> or maybe the crash was the moment it died
2428[19:36:59] <tgunr> what does smartctl say?
2429[19:37:07] <Henry151> tgunr: I went to their main page and downloaded the .deb and installed it with dpkg -i
2430[19:37:17] <rwp> In the past two weeks I have had one sata disk cable vibrate loose, one on motherboard disk controller fail hard, and one disk drive mostly fail, on three different systems, but all had 70,000+ hours on them.
2627[21:34:50] <ZaZaGX> i like debian so much. I'm going to name my first son Debs
2628[21:36:06] <hop> ZaZaGX: you are ware of the origin of the name debian?
2629[21:36:16] <ZaZaGX> yeah
2630[21:36:50] <hop> just checking o.O
2631[21:37:17] <karlpinc> ZaZaGX: The answer is, debian 10.1 will have no new features.
2632[21:37:19] <karlpinc> !stable
2633[21:37:19] <dpkg> [stable] The status of a Debian release when no packages will be added or version-bumped, and changes will only fix security issues and critical bugs. Packages can be removed in rare circumstances. The current stable version of Debian is Buster (10.x); ask me about <releases>. Security bugs are fixed in stable by backporting the fix to the stable version (ask me about <security backports>). replaced-url
2659[21:51:13] <touki> hi all. my laptop is heating with debian buster (it was not on stretch). i was told it might be the new Gnome, but the same happens with XFCE.
2723[22:03:46] <hop> that's ableism. but hey they are sexist, so who cares *g*
2724[22:04:05] *** Joins: toli (~toli@replaced-ip)
2725[22:04:15] <hop> .oO(say "any ideas" one more time)
2726[22:04:24] <greycat> !any ideas
2727[22:04:24] <dpkg> Yeah, I have an idea! But I'll need a sack full of rutabagas, a little K-Y jelly, and a handful of starving goats. And keep the NSA off my back for the next few hours!
2730[22:05:08] <ZaZaGX> have you tried ice or ice cream?
2731[22:05:11] *** Quits: iulian_ (~iulian@replaced-ip) (Remote host closed the connection)
2732[22:05:18] <ZaZaGX> it might cool it down faster
2733[22:05:22] <karlpinc> touki: Does the fan work? Is there a way to tell when the fan is supposed to start? (Yes, but I forgot all I once knew.)
2734[22:05:24] <hop> chiper than moving closer to the pole
2735[22:05:27] <jelly> touki: does the fan spin up?
2736[22:05:38] <jelly> can you hear it?
2737[22:05:41] <jhutchins_wk> touki: I'm sorry, it seems like it's silly hour here.
2738[22:06:25] <jhutchins_wk> touki: I think your best bet would be to make sure the power settings were reasonable and to use some tool like top/atop/htop/sysstat to figure out where the load is coming from.
2756[22:11:09] <touki> ssd is above 70°, while the rest is below 60.
2757[22:11:28] <jelly> can you, in fact, hear the fan?
2758[22:11:45] <jhutchins_wk> ... possibly feel the exhaust.
2759[22:12:08] <touki> jelly: not quite, and it is far from the maximum rate, according to sensors. this being said, before going xfce, I was hearing the fan massively.
2760[22:13:47] <touki> (it is a macbook air)
2761[22:14:00] <jhutchins_wk> touki: The sensors calibration is sometimes off, we are trying to determine whether the fan is infact still running, although I think the question of what's causing the load is more pertinent.
2762[22:14:05] <trek00> touki: usually I run gkrellm to check realtime system resources utilization
2770[22:16:44] <jhutchins_wk> trek00: The problem with GUI monitors like that is that they sometimes generate their own load that masks the actual problem.
2774[22:18:12] <trek00> jhutchins_wk: yes, actually gkrellm is very lightweight except for a single fork every 3 seconds, but the cpu utilization is really low
2775[22:19:17] <touki> it says: TH0A is around 72°, which I believe is SSD, checking.
2786[22:22:59] <trek00> touki: the cores may be ok, it depends by the model, but if ssd is consumer grade, it should not go above 50°C for long time and if it's idle, may be an hardware problem, the cables are ok?
2787[22:23:38] * karlpinc likes 'vmstat 1' for a quick overview of system load
2788[22:23:49] <touki> trek00: it is a Macbook laptop, so I can't tell.
2802[22:28:10] <touki> trek00: yes. so that may be a hardware / firmware problem? like the ssd would not have liked the last reformating or the new debian does not handle well mac ssd?
2803[22:28:13] <trek00> touki: 3 minutes after power on or after booting linux or starting xfce?
2804[22:28:53] <jhutchins_wk> Strange, if your system can't find the driver I wonder how you're bringing it up manually.
2805[22:29:15] <touki> I never waited 3 minutes between startup and logging in.
2806[22:29:31] <jhutchins_wk> trek00: He said it was fine under stretch, so it's not likely to be a hardware issue.
2807[22:29:42] <trek00> touki: please tell me the model so I can check some bugs i'm aware of
2808[22:29:59] <jhutchins_wk> We've suggested several useful tests, booting to an older kernel and running load monitors, no response on those.
2809[22:30:02] <trek00> jhutchins_wk: thanks i missed that part
2810[22:30:28] <touki> jhutchins_wk: ok. so I will try to boot on another kernel.
2811[22:30:40] *** Quits: jnavila (~jnavila@replaced-ip) (Remote host closed the connection)
2812[22:31:11] <touki> see you in a bit then
2813[22:31:21] *** Quits: touki (~jgrosman@replaced-ip) (Quit: Lost terminal)
2827[22:36:49] *** Lord_of_Life_ is now known as Lord_of_Life
2828[22:36:55] *** Quits: astrospace (~quassel@replaced-ip) (Remote host closed the connection)
2829[22:37:00] <BCMM> that only seems to cover some unofficial architectures, and i'm not sure why that is
2830[22:37:52] <trek00> BCMM: here hurd/bsd are missed
2831[22:38:21] <jhutchins_wk> BCMM: "Why" is "because nobody's doing the work".
2832[22:38:37] <trek00> uhm, and do you know where to find build errors of packages?
2833[22:39:02] *** debhelper sets mode: +l 1525
2834[22:39:16] <BCMM> jhutchins_wk: i didn't mean it as a complaint! i mean that I don't understand exactly why "work" is needed - i thought pacakges.debian.org automatically showed metadata for what's on the mirrors
2835[22:39:36] <trek00> i remember there was a page with the output of automatic builds
2837[22:39:56] <BCMM> is it to do with which architechtures are hosted on the ports infrastructure and which are hosted on the "normal" debian infrastructure?
2838[22:40:16] <jhutchins_wk> BCMM: Yeah, but somebody's got to set up the automation, which apparently hasn't happened.
2839[22:40:21] <trek00> found that, buildd.debian.org
2856[22:50:06] *** Guest4906 is now known as argus
2857[22:50:12] <cbthree> Can you install Debian 10 using a LIVE image, for example, debian-live-10.0.0-amd64-mate.iso after establishing a live session?
2858[22:50:36] *** argus is now known as Guest49211
2861[22:51:09] <cbthree> Can you install Debian 10 using a LIVE image, for example, debian-live-10.0.0-amd64-mate.iso after establishing a live session?
2862[22:51:21] *** Joins: wilson (~wilson@replaced-ip)
2863[22:51:45] <towo`> cbthree, yes you can
2864[22:52:55] *** Guest49211 is now known as argusbr
2873[22:54:33] <karlpinc> cbthree: (In other words, I forget.)
2874[22:54:56] <towo`> cbthree, the live media is using calamares as installer
2875[22:55:09] <towo`> i've never used a live debian by myself
2876[22:56:37] <touki> didn't suceed the reboot on older kernel yet. Do you have some documentation link? In the meantime I wask asked to run few other comments, vesides vmstat and iotop
2887[22:59:16] <nvz> cbthree: I would not recommend using the live to install, I'd recommend using the netinstall images
2888[22:59:53] <towo`> what's wrong with installing from live image?
2889[23:00:06] <touki> trek00: iotop shows about 0 everywhere, I installed gkrellm, which gave more or less the same values then lm_sensors / psensors.
2890[23:00:08] <trek00> touki: I was asking only because it is not mentioned in the manual replaced-url
2891[23:00:10] <cbthree> towo`, i know that when you boot up the Debian 10 Mate Live image you are presented with 5 options, one of them is the Graphical Debian Installer but if you choose the option Debian GNU/Linux Live you do not have the option to install Debian 10 once the live session is initiated
2893[23:00:12] *** Quits: jhutchins_wk (~jonathan@replaced-ip) (Quit: Lost terminal)
2894[23:00:26] <trek00> touki: sorry last message was for towo`
2895[23:00:27] <nvz> towo`: I did it a few months ago and it resulted in an install twice the normal size
2896[23:00:49] <trek00> touki: do you have stretch installed anymore?
2897[23:01:05] <nvz> towo`: it translations of all documentation and apps for every imaginable language, language support tools like terminals and input utilities for various languages.. etc
2898[23:01:18] <touki> trek00: nope, that's what I figured out. I had installed an LVM on stretch, so instead of upgrading I formatted everything.
2905[23:02:59] <trek00> nvz: disabling installation of recommends and tasks-* packages usually help reducing the installed syze
2906[23:03:41] <nvz> cbthree: no, they are the exact same installer, that work exactly the same way.. and there are netinstall images with firmware and normal install images with a full DE available on them
2907[23:03:42] <towo`> trek00, calamares has not such option
2908[23:03:50] <karlpinc> cbthree: So long as you have debian running you can use debootstrap to install onto a filesystem. But then you have to do manually a lot of what the installer guides you though. Partionining (encryping?) disks, making filesystems and so forth.
2909[23:04:05] <karlpinc> cbthree: The netinstall is eaiser.
2910[23:04:09] <karlpinc> !firmware images
2911[23:04:10] <dpkg> Unofficial <netinst> and DVD images containing non-free Debian <firmware> packages are available from replaced-url
2912[23:04:35] <karlpinc> cbthree: Always a good idea to use an image with non-free firmware built-in. See above. In case your hardware needs it.
2913[23:04:47] <trek00> touki: i think it's not that simple to install another kernel on your system, you need to build it from the source, instead you could try to make a live USB stick with an older debian version and check again if temperatures are different, if they are and this is a fresh buster install, i would open a bug report against the kernel
2918[23:06:45] <nvz> cbthree: the unofficial only means that it includes non-free firmware which is required by a lot of ethernet, wifi, bluetooth, and video adapters
2929[23:08:17] *** Quits: leandrovianna (~leandrovi@replaced-ip) (Remote host closed the connection)
2930[23:08:20] <trek00> touki: hope this can help you to diagnose the problem
2931[23:08:29] <ibrokeit> ok, so i suspended an openbsd install, pulled the disk, put in a Debian disk, and unsuspended. Then, I realized my error and instead of pulling the debian disk, I attempted a hard reset (not by pulling the plug though). now i am getting XZ-Compressed data is corrupt
2933[23:08:38] <nvz> the live is meant to be as inclusive as possible and without knowing how to manipulate a more advanced install, it will install more than 2GB of additional stuff
2934[23:09:02] <ibrokeit> i've mounted /boot and / with a life CD and not sure what is wrong with the kernel or partition, etc. Everything mounts alright. Note: using luks
2935[23:09:38] <ibrokeit> i'm on sid. i've done this before, years ago, but i forget how i fixed it.
2936[23:10:04] <trek00> ibrokeit: if you can chroot on the broken system, try to launch debsums to check if there are errors on files
2937[23:10:25] <nvz> ibrokeit: you're in the wrong channel if you're using sid, you should be on the OFTC network in #debian-next
2938[23:10:49] <nvz> ibrokeit: but as trek00 said, run debsums, reinstall any packages that are corrupted
2950[23:13:20] <ibrokeit> trek00, i initially presumed that, but that would mean that I have two broken kernel images, which i don't think could occur from this event
2952[23:14:04] <ibrokeit> i'm on debian 9, which i assumed was sid, and 10 being buster :$
2953[23:14:26] <nvz> ibrokeit: sid is the unstable branch, it has no version number it always has been and always will be called sid
2954[23:14:29] <dvs> !stretch
2955[23:14:30] <dpkg> Stretch is the codename for the current <oldstable> release, Debian 9, released 2017-06-17. "Stretch" is the rubber octopus in Toy Story 3, see replaced-url
2956[23:14:35] <trek00> ibrokeit: debsums will tell that, anyway where you get this error? after grub? kernel print something else?
2957[23:15:06] <greycat> It's conceivable that at the time you installed what would later become "debian 9", you were installing sid. If you did it while stretch was still being developed.
2958[23:15:07] <ibrokeit> trek00, after grub, before i would be asked to input the luks password
2959[23:15:18] <ibrokeit> greycat, that's the case
2966[23:17:00] <karlpinc> ibrokeit: Then it is also important to make sure that your /etc/apt/sources.list does not still list sid as a name, but instead says "stretch". Otherwise you'll just keep tracking sid and have crazyness happen because sid breaks.
2967[23:17:26] <karlpinc> ibrokeit: (And then "apt update" to make the sources.list changes take effect.)
2969[23:18:01] <karlpinc> ibrokeit: If it's been sid all along and you've been updating, well, there may well be lots-o-trouble.
2970[23:18:12] <ibrokeit> karlpinc, k i'm on 10/buster, not stretch
2971[23:19:02] <trek00> ibrokeit: if you don't have debsums installed you could manually check with (cd /; md5sum -c /var/lib/dpkg/info/linux-image-4.19.0-5-amd64.md5sums | grep -v OK$)