[00:21] zorcon: I had the issue recently on a 144hz monitor - I don't think I ever found a good way. [00:22] It looks like Gnome has issues with 144hz - mouse looks fine too. That was on 19.04. [00:32] Hello. I am trying to figure out why lines in my crontab do not seem to work. Here is an example in my user's crontab: * * * * * /usr/bin/notify-send -t 10000 "This is a test." [00:32] It doesn't appear to do anything. [00:35] LaurentDumont: Okay darn. Yeah I tried different GPU drivers, no luck. Also it seems like certain windows can be laggier than others. Unfortunate, bc there's no reason my hardware shouldn't be able to handle it [00:57] The wonderful upgrade problems, complete fail all around.. [00:57] Fresh install doesn't work either, I set to autologin .. it get login screen .. login: well just back to login screen [00:59] open a console and update even though it should have installed updates during install ... now the real fun begins, it's spamming dmesg errors actively across the second terminal, it has some strange errors in the terminal rendering which i'm not even sure how it managed that.. and the best part yet: my logged in users X session randomly decides to display over the active terminal [00:59] _how_ ? [01:00] I'm at a complete loss here, can't use any part of it ... maybe kill lightdm or gdm or whatever it decided to use for this version while i cant see what i'm doing? [01:00] than ctrl+f1 back and hope it went to the desktop and then replace ldm etc? [01:00] ffs [01:07] hi all! i'm trying to install my lamp stack on 19.10 desktop and i'm running into a few issues [01:08] first of all, i'm experimenting with the moinmoin wiki. i seem to have followed all the steps in the guide, but localhost/mywiki seems to be outputting python code..... [01:09] I'm not sure this is really the right place for that, unless you're somehow certain the packages are broken and it's not a configuration error [01:10] does the base webserver work if you're not trying to run moinmoin ? [01:11] this channel doesn't support the "ubuntu server guide"? [01:12] Ah yes i'm sure it does, don't mind me.. thought you meant it was steps on moinmoins guide [01:13] :) [01:13] mike802: can you link to the guide you used? [01:14] https://help.ubuntu.com/lts/serverguide/moinmoin.html [01:15] do you have mod_cgi enabled? [01:16] apt install mod_cgi? [01:16] it should be in the core apache2 package, but may not be enabled [01:17] does `apache2ctl -M` list it? [01:17] you can `a2enmod cgi` to enable it if not, then restart apache [01:17] hmm, so it says "unable to locate package mod_cig" [01:18] that sounds like you tried to apt install it? [01:18] yeah.... no luck.... [01:19] yes, it's part of the apache2 package as i said [01:19] ok.. [01:23] mike802: any luck? [01:24] i was able to enable module cgi, then retarted, but no luck locating mod_cgi [01:24] if you enabled the cgi module, that's mod_cgi :) [01:26] amazing! thank you! [01:27] ok, so i'm also having an issue with phpmyadmin. it says, "package 'phpmyadmin' has no installation candidate [01:30] what version of ubuntu are you running? [01:30] not sure about the ubuntu side, but in debian at least, phpmyadmin isn't available in buster [01:30] ,v phpmyadmin [01:30] 19.10 desktop [01:30] aww, no bot here :( [01:30] it worked just find with 18.x server [01:30] !info phpmyadmin [01:31] Package phpmyadmin does not exist in eoan [01:31] ah, thanks! [01:31] yeah, i tried looking up the package location, but no luck! [01:31] i think... [01:31] yeah, the package is gone as of 19.10 [01:32] alright, i'll see if i can do all the stuff i need by hand! thanks anyway! [01:33] looks like there's a ppa, though that does include the note "Do not use these packages on production! Wait until they are released through the official ubuntu-channels" [01:33] Hello! I'm interested in using ssmtp to set up a script to pipe the results of ifconfig to my gmail account. Question, would I need to set up an SMTP server on my localhost, or can I just configure mail or ssmtp to shoot it off without one? [01:35] sentin3l: I haven't used ssmtp but assuming it's more or less like msmtp, you should be able to use it and certainly don't have to configure a full mail server. [01:35] sentin3l: You can point it at google's smtp servers and just send it to yourself using your gmail address [01:48] phew, so msmtp isn't letting me without setting up a .conf ... which, when I go to do, the system doesn't know where its at [02:21] hello, need some help with sending mail from command line to my gmail account :) [02:22] namely, the output from ifconfig === zbenjamin is now known as Guest39577 === zbenjamin_ is now known as zbenjamin [02:48] So... strange thing. Superscript 2 seems to get from my web app to the db and out to a yaml file, but subscript 2 does not. [02:49] My first guess is that MySQL isn't using a 4-byte character encoding, because I think I've heard that in the past. But that doesn't explain why the subscript 2 can be stored and displayed in the web app. [02:49] So any thoughts about why it isn't displaying correctly in the terminal? [03:20] hello mustmodify [03:20] Hi [03:20] what do you see in the terminal instead of the correct symbol? [03:21] a box. [03:22] there is 'showconsolefont' command in terminal, which may help with troubleshooting [03:22] interesting. "Couldn't get a file descriptor referring to the console." [03:23] (I access this machine via SSH.) [03:23] Anyway, I'm going to see if the character's there and will be read on the other end. If not, then I'll dig further. [03:23] I mean, I'd rather see it. But if it works, that's not enough of a thing to be a blocker at this point. [03:24] what terminal is it? [03:24] and what font [03:25] well, I'm using putty and Courier New [03:25] that's a good thought... that font likely just doesn't have that character? [03:25] Though... [03:26] Seems like a popular enough one you'd want to include it. [03:44] try another font [03:44] Bitstream Unicode is good [03:48] will do [03:49] yes [04:30] Take care ya'll [06:12] @leden Just wanted to say thank you for your help the other day. Changing the location of mirror worked. Thank you! :D === mIk3_09 is now known as mIk3_08 === NickG365_ is now known as NickG365 [07:08] i want to use dead windows key for desired folder or application . [07:18] anyone know why the installer keeps freezing after 50%? [07:20] epitamizor: what version? what flavour? [07:20] 19.10 desktop version [07:21] it boots to the live cd and works fine but the install keeps stopping at 50% [07:21] I had no issues installing that over the weekend. lubuntu? xubuntu? 32 bit? 64 bit? [07:22] did you md5sum check the download to ensure that you have a good download? [07:22] i tried selecting the minimal radio button and even that doesn't finish installing [07:22] md5sum from the livecd? [07:23] no -- did you md5sum check the download that you used to created the livecd? [07:23] no [07:24] i heard that doesn't work because the blocks are not the same when it gets written to the media usb [07:24] so md5sum nameoffile.iso [07:24] the usb is bigger than the iso [07:24] correct [07:24] but you are checking the downloaded image [07:24] oh, the iso used to copy matches, but from the iso to the usb it changes [07:26] I'm not sure if my system is too new for the drivers or what,. I tried selecting proprietary drivers for install and same issue occurs [07:26] if you go here: https://ubuntu.com/download/desktop/thank-you?country=RU&version=19.10&architecture=amd64 it gives information on you to verify the download [07:27] oh this is a sha356sum [07:27] sha256 [07:27] usually when you get errors like you describe, maybe something got scrambled in the download -- it happens [07:30] sha256 matches for iso [07:30] epitamizor, then choose option just without updates that will install system from livedvd [07:32] epitamizor: how did you make the dvd? [07:32] blue1, copied to usb [07:32] epitamizor: what programme did you use? [07:32] leden, that option didn't work either [07:33] dd, etcher, dd-rescue? [07:37] dd [07:38] what was the command line you used? [07:39] dd bs=4M ubuntu64.iso /dev/sdd1 [07:40] well I use something like: dd if=file.iso of=/dev/sdd bs=4M [07:41] i have never written to the first partition, always the device name -- [07:41] that should work however [07:41] so doesn't it need format before dd copy? [07:42] I usually do that, yes. otherwise it could be undefined. that said, dd is an image copy, so it will overwrite anything that's there. [07:42] maybe there is still extra stuff on usb that is not being zero'd [07:42] perhaps -- couldn't hurt -- [07:44] you might consider brasso [07:46] brasero [07:46] i have had issues sometimes with dd not working correctly. [07:47] whats weird is the system can boot to the livecd and the applications and environment works fine,but the installer doesn't finish [07:47] there are directions here: https://ubuntu-mate.org/raspberry-pi/ search for: Making a microSDHC orĀ USB -- that has always worked for me. [07:48] i have seen dd do some crazy stuff. [07:48] that uses dd-rescue [07:49] easiest, would be brasero, next dd-rescue, and lastly dd -- okay I am off to bed. almost 1 am here. === AC3J_ is now known as AC3J [07:50] nytol -- sorry I couldn't be of further assistance [08:08] hello, I have headless ubuntu and I'm running it on a work PC that requires all connections to go through proxy, on Windows the proxy works by specifying url to pac (proxy auto configuration) file, without any credentials. The file basically has 2 if rules (each ends with `return ("...")`) and a global return: `return ("PROXY proxy.mycompany.internaldomain:8080");`, how to teach apt to use that proxy? [08:11] pomeha: man apt.conf has some proxy info [08:15] fructose: only ftp_proxy is described, is that what apt actually uses? I thought it used http [08:16] pomeha: that's described in a separate manpage: apt-transport-http(1) [08:16] (proxy settings are there) [08:26] rbasak: thanks [08:27] so it doesn't seem to work with pac [08:30] pomeha: no but it looks like you should be able to write an equivalent configuration [08:30] THe problem with pac is that it requires a javascript interpreter [08:30] Not something that you want apt to be dependent on :) [08:31] rbasak: however https://askubuntu.com/questions/349702/ [08:31] err [08:31] Not sure why that might be [08:32] apt works well with regular HTTP proxy servers [08:32] I use that function all the time [08:32] rbasak: I meant that `man apt-transport-http` has a section OPTIONS > Automatic Proxy Configuration [08:32] Oh [08:32] Yes but that's not pac [08:32] it's apt's own language AFAICT? [08:33] oh [08:33] alright then [08:33] problem is that our proxy is not regular, it is also ns proxy [08:34] You mean DNS [08:34] yeah [08:34] I mean dns proxy or something [08:34] not sure [08:34] If using an HTTP proxy, apt won't need DNS directly I don't think [08:35] thing is some domains don't get resolved to IP when I ping them, yet when I open that very same host via http - the proxy works out and the connection works! [08:35] hell, I don't even know how to debug such networking connections :/ [08:35] wireshark only shows traffic to the proxy [08:35] and its unreadable :/ [08:37] pomeha: do you have something specific that isn't working here? [08:38] I've tried to specify `Acquire::http::proxy "http://proxy.mycompany.internaldomain:8080";`, but `sudo apt update` shows errors like: `Err:3 http://security.ubuntu.com/ubuntu bionic-security Release Could not handshake: An unexpected TLS packet was received. [IP: 10.45.2.200 8080]` [08:41] and `E: The respository 'http://security.ubuntu.com/ubuntu bionic-security Release' no longer has a Release file. N: Updating from such a repository can't be done securely, and is therefor disabled by default.` [08:45] what is mean by pseudo terminal allocation in SSH? What does "-t" option mean in SSH? [08:47] geekPanther: it is explained in the manpage. [08:54] flog, I have read that but I cant understand what pseudo terminal mean? [08:55] What does it mean? [08:59] a device that acts like a terminal === peter-bittner1 is now known as peter-bittner [09:00] commands often behave differently when they detect that stdout is a terminal [09:00] so it creates such a device on the remote end, and forwards everything written to it to the terminal on your local end === peter-bittner1 is now known as peter-bittner [09:11] I just did done installing ubuntu, but things don't look crystal clear [09:11] what t odo [09:15] drogas: could be the display resolution does not match the resolution of your physical display [09:15] drogas: Or what do you mean by "things don't look crystal clear"? [09:16] Anyone have a Dell Precision 7540 with Ubuntu? [09:16] drogas: are you chatting from the computer right now? [09:18] yes I am tatertots it just looks a bit blurry coming from windows. things on firefox looked sharp [09:18] I have the resolution set to the maximum [09:20] it almost feels like I am on remote connection. Do you get what I mean? [09:23] geekPanther: if you are running a command on a remote machine which requires a terminal, like "less" or "tmux", or something which pops up a menu in the terminal, that command will fail if you try to run it directly over ssh [09:23] geekPanther: "ssh remote_machine screen -dr" will fail without "ssh -t" option [09:24] geekPanther: if you "ssh remote_machine" to get a shell, then you can run these commands in that shell. but the -t option to ssh is required to make it work in one step. [09:25] it might be because the drivers are not installed [09:26] drogas: you might check and make sure [09:26] it says NVA0 on the "about " in settings [09:27] and lspci lists only intel it seems [09:27] I have a nvidia quadro graphics card [09:27] I am guessing these are not used on linux [09:28] drogas: in terminal> inxi -Fxxprzc0|nc termbin.com 9999 [09:28] drogas: if you don't have it, it'll tell you how to get it...then run the command above and share the url/link here [09:33] running that command returns no url [09:37] drogas: in terminal> inxi -Fxxprzc0|nc 5.39.93.71 9999 [09:37] drogas: try that [09:38] Hi! I'm on ubuntu 18.04.3. Every other dns request is failing and when it fails the failure is cached. I know the dns on my router works flawlessly. Any ideas what it could be? [09:58] not sure what is going gonna put this on hold for now [09:58] on another note [09:59] my conda base gets activated for any new terminal I open [09:59] bashrc does not contain any conda activate command [10:14] how can it be, that nslookup succeeds, but ping says "Name or service not known"? [10:15] ping uses nsswitch. nslookup does not. [10:15] Try "getent hosts " for a closer emulation to what ping is doing [10:16] And check /etc/hosts, nsswitch.conf, etc [10:16] rbasak: interesting. thank you. [10:17] my nsswithc.conf has this line: hosts: files myhostname mdns4_minimal [NOTFOUND=return] dns mdns4 [10:20] rbasak: mmh.. getent hosts succeeds, while ping still fails [10:20] (to resolve) [10:34] Anybody still using Ubuntu 14.04? Do you mind testing, if this works there? : "sudo add-apt-repository ppa:jarnos/ppa-purge && sudo apt-get update && sudo apt-get install ppa-purge" [10:50] jarnos, Ubuntu 14.04 LTS is EOL & thus off-topic here. Ubuntu 14.04 ESM is supported by Canonical through Ubuntu Advantage, but still off-topic here. [11:00] it seems the latest ubuntu package for firefox-trunk is broken. can't seem to click anything in the window [11:01] obiwland: that might be an IPv6 issue then? Try ping -4 [11:01] Oh, no, sorry. Resolve. Not ping. [11:03] rbasak: the problem is gone since i disabled systemd-resolved. I now use unbound. [11:03] rbasak: but thank you anyway [11:03] rbasak: will try it, should the problem reoccur [11:04] This is how i did it: https://askubuntu.com/questions/907246/how-to-disable-systemd-resolved-in-ubuntu [11:04] obiwland: note: that's not a supported configuration [11:04] obiwland: i don't care. it works and i understand it better than the systemd mess [11:05] obiwland: see: https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1745463/comments/2 [11:05] Launchpad bug 1745463 in systemd (Ubuntu) "Disabling systemd-resolved breaks dhclient resolvconf integration" [Undecided,New] [11:06] Sure, that's your choice [11:06] obiwland: thanks for mentioning it [11:06] I'm just pointing out that it's not recommended in case others thing it's a good idea [11:06] I think it's more likely that you're masking an underlying networking issue, FWIW [11:06] rbasak: that might very well be case [11:07] rbasak: and i'd like to solve it eventually, but for now i've got some work to do ;) [11:07] it's a typical monday here xD [11:08] :) === parys15 is now known as parys [11:22] hello, I have an issue with OpenVPN: i managed to configure it well enough so when I connect to it I can actually connect to the machine hosting the vpn server. However, everything else on the network times out when I try to connect; any ideas? Ubuntu 18.04.3 LTS [11:23] !vpn | fritzroy15 [11:23] fritzroy15: For more information on vpn please refer to https://wiki.ubuntu.com/VPN === the_ant1 is now known as the_ant [11:26] lotuspsychje that is the biggest non-answer I have ever gotten to a question; not only does the wiki article not come even close to helping (which is always the case), but this particular one hasn't been updated in 5 years; grats' [11:27] the last version of ubuntu listed is 10.04 lmao [11:28] fritzroy15: im just widening your options, the wiki has edited in 2015.. [11:53] here's a comparison between the old and new routing tables: https://gist.github.com/lupupaulsv/2ac476a47c50f0b95b3316c8f586f34d [12:02] Hiyas all [12:03] Hi there. [12:03] So I'm trying some containers... [12:04] And my package of interest is: https://packages.ubuntu.com/search?keywords=libaom-dev&searchon=names `libaoim dev` I have Bionic version of ubuntu and according to this link this package should be in the `disco` package... but when I do `apt install disco` I get Unable to locate package disco [12:04] `libaom-dev` [12:05] Any ideas how can I install this package for Bionic version? [12:07] I am having login issues, takes 4 minutes to authenticate on login and sudo. Ideas? [12:08] frojnd, the command is sudo do-release-upgrade, but upgrade your existing packages first [12:08] I noticed pam_fingerprint-gui lines in my journalctl output [12:10] also noticed many kernel: usb 1-4.1: device descriptor read/64, error -110 errors, but only usb device connected is my wireless mouse dongle [12:15] BluesKaj: # do-release-upgrade bash: do-release-upgrade: command not found [12:16] frojnd: are you sure you're running ubuntu? [12:17] Ben64: yes.. in docker why? [12:17] fritzroy15: What routes do you push from the server? [12:17] frojnd: ah, that's probably why [12:18] fritzroy15: Do you have e.g. push "route 0.0.0.0 0.0.0.0" or similar? [12:18] docker messes with stuff [12:18] hm, looking to buy a pcie-based usb-c/usb3 card for my PC ... any suggestions with the best speeds and "just works" kudos for ubuntu? [12:19] frojnd, I thought you were trying upgrade to Bionic release [12:19] BluesKaj: no... release is fine I'm on Bionic ...at least seen from logs... [12:20] I just need `libaom-dev` package [12:20] My question is how can I install `libaom-dev` package on current setup - Bionic [12:20] package doesn't exist in bionic [12:21] frojnd: no, your question is, how do you install libaom-dev in a docker container. [12:21] frojnd: docker containers are not full OS's [12:22] Pricey: it should go through "172.31.0.0/20 dev eth0 proto kernel scope link src 172.31.9.74" on the VPN machine; but for some reason it doesn't [12:22] frojnd, sudo apt install libaom-dev in the terminal [12:24] oh, doesn't exist in Bionic, my mistake [12:24] BluesKaj: that package does not exist in bionic [12:24] Does it exist in newer versions? [12:24] frojnd, maybe there's a ppa for libaom-dev [12:25] frojnd: why do you think you need it exactly? [12:25] leftyfb: for video encoding dependency for some other project... [12:25] frojnd, I'm tesw\ting Focal 20.04 and it's in the repos [12:25] frojnd: what other project? [12:25] testing' [12:25] leftyfb: cobalt [12:25] frojnd: does it tell you a specific library file it needs that is missing? [12:26] Not sure... am setting up env first [12:26] frojnd: it sounds like you should be doing this in a full OS (VM/lxd container). Not docker [12:27] leftyfb: I don't think so... I Need more ppl fiddling with cobalt... [12:27] more man power to say at least [12:27] docker would be perfect fit once setup [12:28] frojnd: maybe contact cobalt for support? [12:28] I've contact ubuntu if there is a way to install that package in Bionic :) will check ppa first. Thanks [12:31] okay I think I see the problem [12:31] the issue is not my screen [12:31] but I think how the fonts are rendered [12:31] hence why it looks blurry [12:31] drogas: can you screenshot that please [12:34] lotuspsychje https://imgur.com/a/4UBYcDI [12:34] not sure if you can see it from screenshots [12:35] but there's definitely a blur around the fonts, and they look low-res. [12:36] drogas: is your graphics card driver installed properly? [12:36] yes apparently when I tried to install the nvidia ones it said it is already installed [12:37] drogas: wich card chipset and driver please? [12:38] I'm getting "error: attempt to read or write outside of disk 'hd0'" and I tried the steps mentioned in https://askubuntu.com/questions/397485/what-to-do-when-i-get-an-attempt-to-read-or-write-outside-of-disk-hd0-error but when i do "insmod normal" I get the same error again as output [12:38] nvidia quadro but the suggested driver was nvidia 430 I believe [12:38] which is the one that is intalled [12:38] can anyone please explain what the error means and how i can fix it? [12:38] drogas: you have your screen issue outside your browser too? [12:39] lotuspsychje yes it seems [12:39] drogas: perhaps try a driver switch as a test? [12:40] can't even pullup settings....keeps crashing now smh [12:43] (gnome-control-center:32083): GLib-CRITICAL **: 12:42:49.052: g_strsplit: assertion 'string != NULL' failedSegmentation fault (core dumped) [12:43] great [12:52] I'm getting "error: attempt to read or write outside of disk 'hd0'" and I tried the steps mentioned in https://askubuntu.com/questions/397485/what-to-do-when-i-get-an-attempt-to-read-or-write-outside-of-... but when i do "insmod normal" I get the same error again as output [12:53] i entered the bios now and i'm seeing in the boot priority order 3 hard disks even tho i only have one [12:56] Hi. I am doing `apt install -y texlive-full' in my Dockerfile but it then hangs because it wants me to select a geographical region and waits for keyboard input... (ubuntu 18) [13:01] dostoyevsky: use DEBIAN_FRONTEND=noninteractive apt-get install -y texlive-full === leden_ is now known as leden [13:11] ioria, finally figured out why a 4 min login. fingerprint-gui was installed somehow and it was hanging up. [13:21] is there a command that shows me which packages were installed on a system (without the automatically installed depencies) [13:21] +? [13:29] apt-mark showmanual [13:38] fritzroy15: Ok, but in your openvpn configuration (on server or client) have you specified that something should be routed? e.g. by using the push config I pasted above? [13:39] Cheez: thanks!! [13:46] is anybody here familiar with BlueZ? [13:46] !ask | nklap95 [13:46] nklap95: Please don't ask to ask a question, simply ask the question (all on ONE line and in the channel, so that others can read and follow it easily). If anyone knows the answer they will most likely reply. :-) See also !patience [13:48] I'm trying to use the BlueZ DBus mesh API but the interface org.bluez.mesh is not there. What can I do to access the mesh api? [13:51] > python3-tk:amd64 // File "/usr/lib/python3.6/sre_compile.py", line 17, in // assert _sre.MAGIC == MAGIC, "SRE module mismatch" // AssertionError: SRE module mismatch // dpkg: error processing package python3-tk:amd64 (--configure) // installed python3-tk:amd64 package post-installation script subprocess returned error exit status 1 [13:51] this is the reason why my `apt install -y texlive-full' still fails.. [13:52] whoa [13:52] (ubuntu 18) [13:52] I have some custom apt repositories, maybe they broke it? [13:52] dostoyevsky: we dont support compiling and neither external ppa's [13:53] Maybe I just need to change the order inthe dockerfile and install texlive before adding other repositories.. [13:53] dostoyevsky: very likely [13:56] https://www.amazon.com/Touchscreen-Ultrabook-Screenpad-i9-8950HK-Professional/dp/B07V827G5B/ref=pd_sbs_147_10?_encoding=UTF8&pd_rd_i=B07V827G5B&pd_rd_r=aa6b1f0e-452f-4ef8-a910-055198cdf772&pd_rd_w=oREWw&pd_rd_wg=1GlUf&pf_rd_p=5873ae95-9063-4a23-9b7e-eafa738c2269&pf_rd_r=Q0C97VE6KHP8CX1TGN1X&psc=1&refRID=Q0C97VE6KHP8CX1TGN1X [13:56] https://www.amazon.com/Touchscreen-Ultrabook-Screenpad-i9-8950HK-Professional/dp/B07V827G5B/ref=pd_sbs_147_10?_encoding=UTF8&pd_rd_i=B07V827G5B&pd_rd_r=aa6b1f0e-452f-4ef8-a910-055198cdf772&pd_rd_w=oREWw&pd_rd_wg=1GlUf&pf_rd_p=5873ae95-9063-4a23-9b7e-eafa738c2269&pf_rd_r=Q0C97VE6KHP8CX1TGN1X&psc=1&refRID=Q0C97VE6KHP8CX1TGN1X [13:56] has anyone already experimented to use linux on this kind of laptop ? [13:57] or this one ? https://www.amazon.fr/Gigabyte-Aero-15-OLED-Sa-7FR5130SH/dp/B07W84XX9Z/ref=pd_sbs_147_6/258-8182924-2689729?_encoding=UTF8&pd_rd_i=B07W84XX9Z&pd_rd_r=c77a85c9-b2a1-433a-b123-0f0bbd0e4531&pd_rd_w=YoUcF&pd_rd_wg=qA046&pf_rd_p=a9f5d7c2-08ef-42e9-9709-9c2a12683b28&pf_rd_r=WXGG8WVTFH3VXFHMW0ZM&psc=1&refRID=WXGG8WVTFH3VXFHMW0ZM [13:57] hi there! i am trying to boot up a old laptop with a ubuntu 16.04.6 i386 live usb-stick [13:57] jeromelanteri: Please stop that [13:57] jeromelanteri: ubuntu installs on alot of several brands, i dont think asus would be a problem [13:57] it runs right into a kernel panic [13:57] lotuspsychje, screnpad recognized also ? [13:58] is there a way to scroll up so i could see the cause for the panic? [13:59] jeromelanteri: i didnt test that myself yet [13:59] any link about screenpad driver for Asus ? [14:00] well jeromelanteri , there are people talking about that ASUS in general on ask Ubuntu and reddit. You'll do better to get info from somewhere like that versus this support channel, until you at least have the device... run into an issue and are in need of support [14:00] jeromelanteri: https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/1846090 [14:00] Launchpad bug 1846090 in linux (Ubuntu) "Asus UX580GD (GDX1505:00 27C6:01F1) Touchpad Screenpad support" [Undecided,Confirmed] [14:01] !info bluez-tools | nklap95: Have you installed [14:01] nklap95: Have you installed: Package bluez-tools does not exist in eoan [14:01] jeromelanteri: ideally you would find a store that lets you test an ubuntu liveusb on it to test things [14:01] so... no screenpad usable... no luck. [14:02] i read on redit something around that, but it is "old" (last year). === Trieste_ is now known as Trieste [14:02] is there a certain release one can recommend for a old machine "core2duo" old [14:02] friendlyguy: lubuntu/xubuntu and !flavours [14:03] well... and for gigabyte aero ? someone use it there ? [14:03] and there is no way to scroll up to see a reason for a kernal panic? [14:03] (i scare about wifi compatibility also...) I read problem on AX-xxx wifi drivers. [14:03] jeromelanteri: Typically bleeding edge devices and technology take time for Linux to catch up to enable such features by default. It depends on the communities interest. In this case it doesn't matter the brand, the underlaying support will be the same [14:03] yes, sure. [14:04] jeromelanteri: Ubuntu specifically works for stability. Meaning it will take time before such support for "screenpad" devices is likely to show up. [14:05] it seems to be the same for wifi. [14:05] jeromelanteri: that wifi AX bug has a fix [14:05] ho good... [14:06] ok, thank you for share your device driver's knowledges. Have a good day/night. [14:18] got an odd issue here. I have php-curl installed and enabled. But when I do a php -i from the CLI, php doesn't think it is loaded. [14:18] This is 16.04 [14:19] pragmaticenigma: I installed it but I still can't see the org.bluez.mesh interface. Maybe there is something messed up with my configuration? [14:20] nklap95: I'm unsure, I did a quick google search and thought tool kit might help [14:21] Nevermind.. PHPBrew issue. [14:22] nklap95: I can't find any documentation related to bluetook mesh networking and ubuntu. not sure if it has been added or is supported by the bluetooth stack [14:26] nklap95: The closest that thing I can find to BT mesh, is possibly to look into zephyr for it's support and implementation of BT mesh [14:31] texlive-full installed now... without the PPAs... PPAs are scary... [14:32] pragmaticenigma: I'll try and build it from source with custom configuration [14:33] nklap95: The recommenadtion of zephyr came from an support article noting that zephyr has a full stack support, while most the other libraries are for clients only [14:33] lubuntu kernel panics as well [14:34] i cant see whats causing the kernel panic [14:34] any ideas? [14:34] friendlyguy: F1 to switch to textboot [14:34] pragmaticenigma: Thank you for your help, I'll take a look [14:34] friendlyguy: then write down the kernel panic, or take a pic [14:34] its booting with text, but the stack trace moves the source outside of the window [14:35] erm, screen [14:37] nklap95: I had closed the browser window to soon... here is the source of said srticle: https://stackoverflow.com/a/54872897 [14:47] hi [14:47] welcome kek_ [14:51] enow trying bodhi linux [14:52] so I just had a server that crashed (it likes to crash randomly). When I told my friend who is physically present to check the screen it just showed a "terminal cursor". Couldn't write anything. Anyone has an idea how the server got in that state? [14:52] just wanna boot up "some" linux and transfer a bunch of files before i burn that machine [14:52] ide 2.5"... [14:54] crap... this one creates a kernel panic as well [14:54] the kernel output is just SO fast its impossible to read it [14:55] Hello. I'm trying to figure out why my cron jobs do not appear to do anything. As a test I have " * * * * * export DISPLAY=:0 && /usr/bin/notify-send -t 10000 "This is a test." " in my user's crontab, and I see that it is running in /var/log/syslog but nothing shows up. [14:56] ctrl+s to pause the output, ctrl+q to resume [14:56] its below 2 seconds [14:57] ill try with 960 fps per second recording speed [14:57] mindofmateo: probably not enough to guess the right DISPLAY variable. You probably need to set some DBUS variable as well [14:57] maybe this will work [14:57] mindofmateo: better to run it from inside the session instead of from cron [14:58] I do not know what that means. [14:58] mindofmateo: run it from an autostart application (xdg-autostart), so it gets run when you log in to your session, with all the necessary environment variables set [14:59] I am just using notify-send as a test to see what I'm doing wrong. I'm trying to run a script in /home/myuser/scripts/scriptname.sh [15:00] same, run it as an autostart application [15:02] mindofmateo: setting DISPLAY is not enough, you also need XAUTHORITY and possibly DBUS_SESSION_BUS_ADDRESS [15:04] mindofmateo: try to put the output somewhere to see whats going on, DISPLAY=:0 notify-send 2> /tmp/log [15:04] i recall it being enough to just put DISPLAY [15:04] in unity [15:04] notify-send just silently fails though. You can "emulate" the cron-behavior with env -i "DISPLAY=$DISPLAY" notify-send "this is a test" [15:05] env -i is a very good way of testing [15:05] I'd just avoid running it from outside the X session in the first place [15:07] I'm not actually trying to use notify-send, I have a script I want to run periodically, I was just using notify-send as an "easy" (or so I thought) way to test cron. [15:09] if the actual script does not require an X session, then sure, use cron. Redirect all the output to a file; * * * * * /your/script >/tmp/testing-cronjob.log 2>&1 [15:10] OK. Thanks for the help. I thought I tried that before, but the outfile was empty. I will try again though. [15:10] if it's a shell script, you can add set -x at the start of the script to get some more data to debug by [15:10] I'm just confused, I even modified the script to only use full paths, too. [15:11] Mh. Don't do that. Just makes it harder to maintain. [15:11] Instead, just set PATH at the top of the script to include all the directories you need commands from [15:12] well, i managed to get glimpse on the kernel panic, but... i think even the display is to slow to show the error msges [15:12] looks like a memory error to me [15:13] @geirha That's what I thought too, I was just trying to narrow down points of failure. === oerheks1 is now known as oerheks [15:24] can anybody help me with, what the default 'sh' shell is? [15:25] in WSL (Windows Subsystem Linux) it point /bin/sh to /bin/dash [15:26] mcm_: your default shell is readlink -f $SHELL [15:26] is this also the case on a default ubuntu instl? [15:26] mcm_: for WSL specific help please join the dedicated channel: [15:26] !wsl [15:26] Windows 10 has a feature called Windows Subsystem for Linux, which allows it to run Ubuntu (and other Linux distro) userspace programs without porting/recompliation. For discussion and support, see #ubuntu-on-windows or ##windows. For installation instructions, see https://msdn.microsoft.com/en-us/commandline/wsl/install_guide [15:27] tomreyn, good idea, but my question was really regarding how a default ubuntu is setup... [15:27] mcm_: i believe the default shell is dash on most ubuntu versions. you did not specify a version. [15:27] "default shell" is perhaps not the right question; on ubuntu /bin/sh is dash but the default login shell is bash [15:27] tomreyn, ok thank you very much! [15:28] mgedmin, yes, thanks. [15:28] when I hear "default shell" I tend to think about login shells [15:29] i had issue with '[[' recognition on WSL ubuntu. and wanted to understand if it would also happen under default Ubuntu. [15:29] mgedmin, yeah me too usually ;-) [15:29] on 18.04 desktop it seems to be bash by default [15:30] thank you guys, you helped me a lot! :-) [15:30] if you use [[ in your shell scripts, use #!/bin/bash (or #!/usr/bin/env bash) as the shebang line [15:30] ... welp I forgot if bash is in /bin or /usr/bin on older ubuntus (newer ubuntus have /bin a symlink to /usr/bin so it doesn't matter, except if you guess wrong the script won't run on older ubuntus or other distros) [15:32] env is best, as it is portable to other distros [15:32] There are distros where bash is in neither /usr/bin or /bin. [15:33] mgedmin, yeah, it's an installer that extracts it beforehand... so i might need to change the /bin/sh to bash for a while.. [15:33] * mgedmin checks a bionic container in lxc and finds a /bin/bash and no usrmerge [15:33] on 18.04, /bin is not a symlink, and bash is at /bin/bash [15:34] Hi, I made 2 AWS instances, with the same CPU and RAM, and installed Ubuntu in one, and Windows 2019 in the other. On loading firefox in both the instances and checking the same exact websites, the Ubuntu instance seems to use a lot more CPU and gets stuck for sometimes 1-2minutes before displaying the page. Is it because Ubuntu uses more [15:34] CPU/resources than a Windows OS? [15:35] I have repeated this with a couple more instances [15:35] But the same issue [15:36] * mgedmin would check for background jobs -- e.g. apt update -- and also video drivers -- is ubuntu using software rendering via llvmpipe? [15:38] did ubuntu discontinue network install cds? [15:38] Yeah, graphics drivers could be weird on AWS [15:39] hardwarenoob: firefox will be a tiny bit slower on some specific websites (because GPU accel is not as optimized in linux), not by some full minutes though. [15:39] dkdn, https://ubuntu.com/download/alternative-downloads [15:40] legreffier I see, this could be the reason why it stays at 100% CPU? [15:40] hi how can I extend a disk size in kvm? I tried with growpart but I got NOCHANGE: partition 1 could only be grown by 2046 [fudge=20480]. I presume the problem is because of the extended partition is close to the primary. [15:40] legreffier sometimes the instance needs to be rebooted as everything stops working (including ssh client) [15:42] hardwarenoob: how much RAM did you give to the VM? running out of RAM tends to result in nonresponsive systems that spend all the time paging executables in and out 100% of the time [15:42] hardwarenoob: what instance type do you pick ? [15:42] t2 and cpu-credit oriented box might be too cpu hungry for this. [15:43] mgedmin legreffier the t2.micro free tier [15:43] and yeah, once you used all the credit , the box is broken beyond repair. it's just not made for this. [15:43] The same free tier on windows seems to work faster [15:43] how much is that in gigabytes? [15:44] mgedmin I think 1GB RAM [15:44] 1vcpu, 1G RAM [15:44] ouch, that's not a lot [15:44] it was memory related... removed dimms one by one. culprit found, laptop boots [15:44] give a try to bigger boxes [15:44] firefox is a bit infamous in being more optimized for windows than for linux (which, granted, is where the majority of their users are) [15:44] don't forget to delete unused instances when you're done trying [15:46] My experience, tells me firefox on windows is more "hoggy" than it is on linux [15:47] it's certainly, subjectively, less pleasing to use on windows [15:47] legreffier yeah i deleted them and made new t2.micro instances of ubuntu 18.xxx [15:47] the issue persists [15:48] firefox loading any big site makes the whole box stuck. apache/sshd all stop working and there doesnt seem to be a way to kill firefox [15:48] and the instance needs to be manually rebooted [15:49] !crosspost [15:49] Please don't ask the same question in multiple Ubuntu channels at the same time. Many helpers are in more than one channel and it's not fair to them or the other people seeking support. [15:49] hardwarenoob: you have a graph for cpu credit. [15:49] if there's not enough RAM to run firefox + apache + ssd, recreating an instance with the same amount of RAM is not likely to help [15:50] it's unrelated to ubuntu though. [15:52] Hi, something weird. on an amd64 I can easily install `g++-s390x-linux-gnu` on ubuntu 18.04. but in my CI (which is s390x arch) it says that `Package g++-s390x-linux-gnu is not available, but is referred to by another package`. why? https://travis-ci.org/elichai/bitcoin/jobs/616745699#L98 [15:52] eugenio_, what format is the virtual disk in? [15:52] elichai2, what is "s390x arch"? [15:53] any idea why I can connect to my OpenVPN host but not to any other hosts on it's subnets from my client-pc? cant figure out what's missing... firewall is off and ip/packet forwarding is enabled [15:53] legreffier mgedmin this is how top looks like [15:53] toptop - 15:52:25 up 5 days, 4:36, 0 users, load average: 0.04, 0.01, 0.00Tasks: 210 total, 1 running, 132 sleeping, 0 stopped, 41 zombie%Cpu(s): 1.3 us, 0.3 sy, 0.0 ni, 98.3 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 stKiB Mem : 1007612 total, 75564 free, 526728 used, 405320 buff/cacheKiB Swap: 0 total, 0 free, 0 [15:53] used. 319116 avail Mem [15:53] A CPU arch? [15:53] yes [15:53] https://wiki.ubuntu.com/S390X [15:54] fritzroy15, you probably do not have the right routes configured on the client [15:54] elichai2: https://packages.ubuntu.com/bionic/g++-s390x-linux-gnu confirms this -- the package is not available for architectures other than amd64 or i386; I wonder why (failed to build probably) [15:55] lordcirth, qcow2 [15:55] ha [15:55] !info gcc-5-s390x-linux-gnu [15:55] mgedmin: but then programs ask for it :\ [15:55] Package gcc-5-s390x-linux-gnu does not exist in eoan [15:55] !info gcc-6-s390x-linux-gnu [15:55] Package gcc-6-s390x-linux-gnu does not exist in eoan [15:55] !info g++-6-s390x-linux-gnu [15:55] Package g++-6-s390x-linux-gnu does not exist in eoan [15:55] good news! in eoan it also supports arm64 and ppc64el, but no s390x: https://packages.ubuntu.com/eoan/g++-s390x-linux-gnu [15:55] i think it is part of gcc-5-cross [15:56] eugenio_, power off the VM and "qemu-img resize foo.qcow2 +20G" [15:56] lordcirth i picked them up from an existing machine, copied the entire /etc/openvpn config directory; here is a comparison of the routes: https://gist.github.com/lupupaulsv/2ac476a47c50f0b95b3316c8f586f34d [15:56] does it make sense to use a cross compiler when you're compiling for the host architecture? [15:56] lordcirth, I have already dit it, my problem is in the next step, resize the partition of the VM [15:56] mgedmin: autotools sometimes look for that [15:57] lordcirth for example, I can access 172.31.15.254 through the old machine, but not through the new one [15:58] oerheks: how can I check if it's inside gcc-8-cross-base? [15:58] eugenio_: pastebin your partition table inside the VM? (output of sfdisk -l will do) [15:59] possibly you need to growpart the extended partition before you can resize partitions 5-n [16:00] elichai2, check launchpad? https://packages.ubuntu.com/bionic/devel/gcc-8-cross-base [16:00] !info g++-8-s390x-linux-gnu [16:00] g++-8-s390x-linux-gnu (source: gcc-8-cross (31ubuntu2)): GNU C++ compiler (cross compiler for s390x architecture). In component universe, is optional. Version 8.3.0-23ubuntu2cross1 (eoan), package size 6767 kB, installed size 19618 kB [16:00] oerheks: doesn't say what binaries it have [16:00] !info gcc-8-cross [16:00] Package gcc-8-cross does not exist in eoan [16:01] Hello [16:01] mgedmin, https://pastebin.com/YZL47YpV [16:02] Ubuntivity, hi [16:02] What is the user associated with Ubuntu login screen? [16:02] any other ideas? [16:03] I fiddled with xgamma one day, and the login screen got different gamma values than normal, however, once I login the gamma values return to normal as associated to my user account [16:03] eugenio_: ah! yes, you're right, the end of the root partition cannot extend past the beginning of the extended partition; you'll have to delete it and recreate it later (and the swap partition too) [16:04] mgedmin, so I have to remove both vda2 and vda5 [16:04] if you remove vda2 vda5 will disappear automatically, I'd expect [16:04] yep. boost's autotools wants the exact cross compiler :( https://travis-ci.org/elichai/bitcoin/jobs/616751013#L844 [16:13] hello. I have three VOB files audio1.VOB, audio2.VOB, audio3.VOB . How can I make a ISO DVD (audio only) from them? I'm trying k3b but I don't understand where to put these audio files [16:17] !crosspost | Ubuntivity [16:17] Ubuntivity: Please don't ask the same question in multiple Ubuntu channels at the same time. Many helpers are in more than one channel and it's not fair to them or the other people seeking support. [16:17] Ubuntivity, your question is on topic, given you want to know how to change the gamma on your login screen? [16:17] yes lordcirth, sorry for cross-posting [16:18] Ubuntivity, No problem. Perhaps you just want to change the X default gamma globally? [16:18] lordcirth: maybe that will work, but how exactly? [16:21] Ubuntivity, once you have logged in, if you use xgamma, does it work? [16:22] lordcirth: yes. Once I login, the gamma returns to default (i.e. 1,1,1), and I can change the gamma using xgamma in terminal with no problems [16:22] Ubuntivity, ok. So, what is probably easiest is to stick your xgamma command in your user's startup [16:23] So it will be run whenever you log in [16:23] lordcirth: I'm not having problem with my user, it 'spontaneously' return to normal after logging in (without any startup script) [16:23] lordcirth: however, the login screen is being shown abnormal early at startup [16:24] Ubuntivity, yes, so if you want it to be changed for your user, set it to be changed when you log in [16:24] so I still didn't figure otu why a server crashes every few weeks. Apparently it get's suspended but why? https://bpaste.net/show/MUEH2 How could I find out what triggered it being suspended? [16:25] kek_, well that's odd [16:26] kek_, its commong for drives and power supplies and ram to fail. run memtest86+, and check the SMART info for the drive to see if its failing [16:26] there also seems to be a problem of the communication of the USP and the server. So maybe somehow that "broken communication" triggered a suspendion? (but the USP logs just repeatedly show that the communication was lost, nothing else) [16:26] kek_, try disabling suspend: sudo systemctl mask sleep.target suspend.target hibernate.target hybrid-sleep.target [16:27] lordcirth, yeah but I'd still like to know why I got that in the first place :) [16:27] kek_, oh you have a UPS and an automatic shutdown configured? [16:27] lordcirth: I think I'm not explaining myself well. I have no problem regarding my user, the problem only affect the login screen. And I want to fix that. [16:27] lordcirth, I didn't set up the UPS and I don't know a lot about it. Basically it was there when I arrived. The logs get spamed by it. [16:28] Ubuntivity, so, you changed the login screen gamma, and you want to change it *back*? [16:28] lordcirth: EXACTLY [16:28] Ubuntivity, oh ok then. Well, how exactly did you change it? [16:29] lordcirth: That is what I DON'T really know :/ [16:29] lordcirth, https://bpaste.net/show/Y7PFK and lots of "lost connection" - it's just a guess. I'll for sure try to set it up correctly but I won't be physically there so I'll just tell them to unplug it for a few weeks. [16:30] kek_, it's possible that disabling suspend will cause the culprit to log an error [16:30] lordcirth, ok thanks a lot, I'm ging to do that and check the logs and see where it leads. [16:41] Anybody have advice on how to add a printer? It worked so well in 19.04 but utterly fails for both the printers here in 19.10. [16:42] Is it worth filing bugs on specific printer models that don't work? [16:43] akk, if you can reproduce that it works on 19.04 but not on 19.10, then yeah, that's a regression and should be reported. [16:44] step 1: check openprintin and the vendor website if it is supported [16:44] unclear if it is the same printer you tried on 19.04 and now on 19.10 [16:45] The vendors (Dell and Brother) do have downloads, but without installing those CUPS shows 3-4 different options for each printer [16:45] which generally all have the same name, so it's hard to tell which one I'm trying when I delete the printer and try again. === SimonNL is now known as SimonNL_Afk === mnemonic is now known as Guest28649 [16:50] akk, quite sure Brother wants its own drivers [16:51] Right now, on the Dell, I tried adding it choosing the first driver, did "Print test page" and got many pages of a couple lines each of symbols [16:52] so I did "Delete printer", and after it did that, it had added a different, "driverless", entry for the same printer [16:53] so I tried that and now a test page works, but "lp -o sides=two-sided-long-edge -o collate=true" prints 2-sided but flips around the short edge, not the long one [16:54] Anyone here have experience with OPAL drive encryption? Any issues? [16:55] and lp -o sides=two-sided-short-edge -o collate=true flips exactly the same way, there doesn't seem to be an option that will flip in a sane way. === lemsip is now known as emilsp [17:07] I just discovered every single other user of my home PC can read the contents of all the files in my home directory while being logged in as themselves [17:07] why is this the default, and how do I prevent this? [17:07] I want my ~/ be readable only by myself [17:09] oxek: You can use chmod to change files that already exist, and umask to make sure newly created files aren't readable. [17:09] oxek, chmod 0700 ~/ [17:09] lordcirth: chmod 0700 or chmod 0770? [17:09] It is the default because historical reasons [17:09] With -R if you want it to apply recursively to all files inside the homedir. [17:09] oxek, either works [17:09] I would not use -R [17:10] what can break if I do this? [17:10] -R could break things. just changing the top level directory can be easily reverted [17:10] Actually yeah, don't use -R with 700, that will make all your files executable. [17:11] There might be some services that run as non-root that read things in your homedir. But I don't think any are critical. [17:11] But be aware -R exists in case you later want to do something like chmod -R o-r ~ [17:11] careless chmod -R is , however, a common source of support requests here. [17:12] would it not be safer then to use `chmod o-rwx` instead of `chmod 0700` because the second option will modify which files are writeable my myself as well? [17:12] I understand that I am the only member of my group, so I only care about "others" right? [17:14] oxek, it can be useful to add service users to your group if you want to whitelist them to access your homedir [17:14] o-rwx is also valid. [17:15] if I do the change only to my home directory, not recursively, will that prevent other users from reading files in my home directory if they know the exact location and filename of such a file? [17:17] oxek: Yes. 711 would allow them to access a file if they know the exact location, 700 will keep them out entirely. [17:19] oxek, nope, if he had sudo right [17:19] oxek, 'r' on a directory allows 'ls'; 'x' allows access to the files underneath. so '--x' allows only going to known paths, '--- [17:19] * allows nothing [17:19] Obviously if someone has sudo they can do what they want, yeah [17:19] oxek, and i think he has [17:20] oxek, i'd go with encryption [17:28] which group a user must be in to use sudo? [17:29] dldm, by default, 'sudoers' [17:29] https://help.ubuntu.com/community/Sudoers [17:30] the group sudoers does not exist [17:30] it's sudo [17:31] 'groups' will tall the name; sudo, and read to wiki howto add an user [17:42] Indeed, the printer works great in 19.04, so I filed a bug. [17:45] akk: thank you for reporting bug #1853879 [17:45] bug 1853879 in cups (Ubuntu) "Dell E310dw: Default driver doesn't work, driverless fails on sides=two-sided-long-edge " [Undecided,New] https://launchpad.net/bugs/1853879 [18:02] <_Sean_> ---In case it helps anyone and because I couldn't find /anything/ on the googles, if anyone hears of a Dell user having NIC issues w/ any variant of the WD19 dock, it looks like the current workaround is to set thunderbolt to 'No Security' in the BIOS. (We currently have similar issues w/ different laptops in windows and it's the same fix and known to Dell) [18:02] <_Sean_> both dell stock Ubuntu 18.04 and a custom baked image [18:05] _Sean_: could you explain a bit more of the nic issues, what happens exactly [18:11] <_Sean_> OS doesn't see the NIC presented over thunderbolt in the Ubuntu use-case [18:13] <_Sean_> i don't have a whole lot of other telemetry that someone would typically want as I just needed to get a functional image over the line and my focus was not on debugging that or the why it didn't work. But for users with WD19 docks even in Win10 it seems there are a lot of known issues where that's the simple/temporary fix. On the win side Dell has reported back to my service desk that a future BIOS update is intended for the lattitude line ... [18:13] <_Sean_> ... to address WD19 issues with security enabled [18:15] <_Sean_> on the Ubuntu issue i was have it was Precision 5530 & 5540's the OS just simply never saw the hardware being presented -- never showed up in lspci and no errors or mention in syslog for the Realtek NIC. [18:19] _Sean_: so bios set to no security has fixxed your issue? [18:54] is it normal that my root directory has gone from having 186GB free to just 113.7GB without me doing any kind of installation? [18:56] Rabid_Raven: if your /home is shared with /, then your personal files may be why it has shrunk so much. [18:56] and yet my personal files haven't been changed in any way, ryuo [18:57] Rabid_Raven: well something is using up the space. that was just one possibility. [18:57] Rabid_Raven: du can help narrow it down. [18:57] like I'm using the kde widgets to monitor my drives and in the time it took to type this sentence, it went from 112.4GB to 112.1GB just like that [18:58] if you're handy with a terminal, you could try: [18:58] du -sh ~ [18:58] it'll eventually spit out a total of how much your home directory is using. [18:59] somewhere to start. [18:59] i would check growing logs [18:59] that too. [18:59] though, that fast? [18:59] A few days ago, my system literally told me that i was out of space in /var [18:59] sure, i have seen 5 gigabyte/hr [19:00] I couldn't imagine what kept causing it to lose space [19:00] but it seems to be doing it again [19:00] or that was .xsessions errors [19:00] bleachbit doesn't help and neither does removing /var/log [19:00] well [19:00] it would help to find out how your space is distributed [19:00] that's where du can help. [19:01] or even find [19:02] sorry, guess i'm not helping. i'm used to using a terminal to solve such issues. [19:03] Rabid_Raven: did you try some live tails of your logs yet? perhaps there's something really quick spamming there? [19:05] lotuspsychje_: how do i do that? [19:05] I'm comfortable in linux but definitely not out of the newb territory yet [19:06] Rabid_Raven: I always use `du -hax | sort -h | less` [19:06] Rabid_Raven: could try tu hunt a few different logs, tail -f /var/log/syslog or journalctl -f [19:06] vlt: hax! [19:06] du -hast [19:06] ;) === boshhead_ is now known as boshhead [19:07] hello guys. which files should I edit to change the SSH and SFTP ports? [19:07] journalctl -f just tells me that it can't find localhost [19:07] PeGaSuS: May I ask why you intend to do so? [19:08] is it possible that the fact that hosts no longer has a line explicitly saying localhost 127.0.0.1 is causing this error? [19:08] Rabid_Raven: we are looking for fast spamming errors, is that the case? [19:08] lotuspsychje_: yep [19:08] which are causing / to fill up insanely quickly [19:09] Rabid_Raven: can you pastebin a few please? [19:09] whislock: security reasons and because as far as I know two computers can share the same SSH port on the LAN? or I'm mistaken? [19:09] Rabid_Raven, 'ncdu /var/' might be helpful [19:09] the results of journalctl? [19:09] PeGaSuS: To answer the technical question, the SSH listener port is specified in /etc/ssh/sshd_config. That controls both SSH and SFTP. [19:09] PeGaSuS: However, in terms of security, changing the port does effectively nothing. [19:09] ncdu not installed, one sec [19:10] PeGaSuS: It can decrease the count of log entries, but given how little that consumes in terms of disk space, that's hardly a real benefit. [19:10] whislock, PeGaSuS: It keeps your auth.log much cleaner ;-) [19:10] says that /lib/ is 678.8MiB but that's about it for ncdu [19:10] vlt: That's not a security improvement. ;) [19:12] I have two laptops that will be accessible from the outside. so I need to change the port of one of them [19:12] PeGaSuS: Ah, so port forwarding is required? [19:12] ok, i think i found the error [19:13] it seems to have something to do with my edited hosts file [19:13] PeGaSuS: If that's the case, you don't need to change the port on the laptops themselves. You'd just forward two different ports from the firewall to port 22 on each laptop. [19:13] the localhost entries were incorrect and were likely causing problems as a result of some services not finding what they were looking for [19:13] Is there any built in ways to track battery usage/drainage? [19:13] whislock: hum.. something like 23 for 22 of the specific internal IP of the laptop? [19:14] PeGaSuS: Don't use another well-known port. You could use 22 for one, and, say, 22000 for another. [19:14] I see [19:14] time to restart and check [19:14] is just to separate them in a matter of fact [19:14] 22 would forward to one laptop's port 22. 22000 would forward to the other laptop's port 22. [19:14] Irssi 1.2.1-1ubuntu2 (20190628) - https://irssi.org [19:15] PeGaSuS: Within the same LAN, there's nothing wrong with all systems listening on port 22. They differ in IP address already. [19:16] whislock: I wasn't sure about that tho. is the SFTP port the same as the SSH port? [19:16] Yes. [19:17] SFTP is a file transfer mechanism that is part of SSH. They aren't separate. [19:18] also, while SFTP works, it's not the best performance for file transfers.. [19:18] oh. so, FTP listens on 21 and SSH/SFTP on 22? [19:19] PeGaSuS: Yes. [19:19] Manouchehri: If you're using SFTP, it is likely that performance is a secondary concern to security. [19:20] You are right, though. It would be great if it were faster. [19:20] whislock: SFTP is pretty annoying to lock down properly with multiuser. [19:20] Manouchehri: I actually have that problem solved! [19:20] whislock: Ty [19:21] Manouchehri: I'll write up how I accomplished it and paste a link here, if you like. [19:22] Manouchehri: That said, if your goal is to disable SFTP for certain users, that's generally not feasible. Any user can easily execute the sftp handler. [19:22] But I have gotten the config down pat for sftp-only users. [19:25] <_Sean_> if lotuspsychje ever comes back - answer to his question = yes [19:28] disable SFTP for certain users... [19:28] I wonder. [19:29] ryuo: As I said, any user that can log in interactively can just manually execute the sftp handler. [19:30] ok so if I _shutdown_ the computer and log back in, whatever space I lost is returned [19:30] whislock: i do know sshd can be configured to have special rules depending on attributes of the client. [19:31] but if I _restart_, it ends up lost [19:31] where should i look if that's the case. Which part gets emptied? [19:31] i guess not for SFTP though. [19:31] ryuo: It can be, but some of the more advanced modifications that you can make to the config won't necessarily have the desired effect due to what a user can do once logged in. [19:32] indeed. [19:32] so you'd likely have to disable the normal ssh options. [19:32] for that user. [19:33] Yeah, but as long as you're allowing interactive login, the problem exists: The user can manually run the sftp handler. [19:33] guys, thanks for your help so far but if I lose space while I am working and a restart doesn't reclaim the space but a shudown does, what is it that is likely wasting all of that space in /? [19:34] Rabid_Raven: Are you in a full-fs condition right now? [19:34] whislock: i'm sorry if I don't understand but what does that mean? [19:34] full file system condition? [19:34] Rabid_Raven, does 'ncdu -x /' show anything? [19:34] Yes. [19:35] lordcirth: using the command as we speak [19:35] lordcirth: /home is 31.2GB which is normal, /var is 6.5GIB, /usr is 5.7GB ... [19:36] I doubt /usr would grow so it looks limited to /var which is what my system told me i ran out of space in the other day anyway [19:36] but even if I delete /log during those times, it doesn't fix the problem [19:36] Rabid_Raven, yeah. so descend into var and see if you can find anything unusual === led_dark_2 is now known as led_dark_1 [20:01] go right to /var/lib and have a look [20:02] how can i install ex? [20:04] Phruis``: hmm, I'm not sure how to find all the packages that can provide an ex, but at least on my system /usr/bin/ex is provided by vim (through vim.basic) [20:04] hmm ok [20:06] yeah i already have vim installed [20:06] is it possible to get access on a ext4-usb-drive on guest session? how do i configure the guest session to have access to ext4 formatted drives? cos normally it needs permissions that i dont have in guest mode [20:08] peter22222, a specific ext4 drive, or any USB the guest inserts? [20:10] it is my usb drive that i formatted in ext4. i d like to have access to it via guest mode. i d like to set up the pcs with the providing guest sessions that even ext4-formatted drives are accessible. is that possible? [20:11] peter22222, it should be. Whatever you do, make sure that it is mounted with 'nosuid' [20:12] according to: https://askubuntu.com/questions/1126199/how-do-gain-permissions-to-access-a-usb-formatted-as-ext4-in-guest-session [20:12] it doesnt seem possible that easily [20:15] lordcirth thats the thing... in guest mode I dont have any permissions to trigger mount options.. [20:16] peter22222, non-root users can mount USB sticks with the permissions that, eg pmount allows. It should be possible to do the same for guests [20:17] I'm guessing you could either modify or add udev rules to recognize a filesystem uuid or similar and use different mount options [20:17] Hi [20:17] Stuck on "Failed to Connect to lvmetad, Falling back to Device Scanning" [20:17] you may also be able to modify polkit rules to let the guest user or all users mount things without prompting, would that help? [20:18] Can no longer drop to terminal, even with nomodeset [20:18] ericus, what Ubuntu version? [20:18] sarnold thank you... i ll have to research about tht... am kind of a noob [20:18] This happend after I uninstalled amd drivers that didnt work with new kernel, worked on prevoius two [20:18] thank you, too, lordcirth [20:18] 18.04 [20:19] peter22222, alternatively, are you sure you need to use the guest user? [20:19] ericus, ok, and was this working before? does a recovery boot work? [20:19] The old way with lightdm was to do something like: adduser $USER plugdev to /etc/guest-session/prefs.sh [20:19] lordcirth yeah.. im a teacher at a school and we use ubuntu in our pc rooms. students only have acces via guest mode. well, 99% of them have usb sticks formatted in not-ext4 though [20:20] genii: heh is that it? [20:20] ( for guest access to removable storage, CD, etc) [20:20] peter22222, https://help.ubuntu.com/community/CustomizeGuestSession [20:20] sarnold: With gdm, I dunno. But this used to work if you had LightDM [20:20] genii: does gdm have guest session? [20:20] You need to force nosuid as well, though, if you allow any filesystem that supports suid [20:20] nope [20:21] Havent tried recovery boot, doing it now [20:24] You are in emergency mode [20:25] Something like that [20:27] Wasnt able to choose anything from the rescue menu before that popped up, monitor is now blinking. Black for like a sec, text for about 0.1 sec [20:28] ericus: It appears that your installation is borked, probably easier to start from scratch at this point [20:28] i still dont have permission to the ext4 formatted usb.... how do i add nosuid? [20:29] Oh that is aids [20:32] I was able to run recovery on a older kernel [20:33] Lets see IF that one boots with nomodeset [20:34] No, same crap again.. === TreyHarr1 is now known as TreyHarris [20:36] mount is not possible in guest mode [20:37] peter22222, did you read the link i posted ? [20:37] yes i did [20:38] i did adduser $USER plugdev to /etc/guest-session/prefs.sh [20:40] ioria in the link is unfortunately nothing written about mount permissions in guest mode [20:40] You would have to change default from GDM to LightDM also [20:40] peter22222, there is no such a cmd in that page [20:40] i guess is already using lightdm [20:40] peter22222: ohhhhhh. so, now that you've got it mounting, ext4 is still enforcing the permissions, and you haven't given world read/write permissions? [20:41] sarnold yes it is mounted, but i dont have permissions to read and write [20:42] peter22222, what release of ubuntu ? 16.04 ? [20:42] 16.04 yes [20:43] ioria it s an older version, yes [20:43] peter22222: afaik there's no mount flag that you can give the kernel to ask it to *not* enforce permissions on ext2/3/4 filesystems; you'll either need to set world read/write on the files and world read/write/execute on the directories, or you'll need to add ACLs to the filesystem (and mount command) that allow accesses from the entire guest ranges [20:44] peter22222, https://askubuntu.com/questions/186279/how-to-access-mounted-file-systems-as-a-guest-user [20:44] peter22222: *maybe* you can abuse shiftfs to do something similar.. but I'm not sure how that works [20:46] ioria... thank you i ll try that! sorry i didnt find that in the first place :-) [20:46] good luck [20:50] I was just able to boot into GUI when installing amd drivers again [20:51] And now all three monitors fire up and I reach the login screen, yay! [20:51] So basically what happend was that the kernel and drivers didnt match? [20:52] not working... [20:52] thanks anyway... good night [20:54] pragmaticenigma [21:00] has anyone seen issues with disney+ on chrome? [21:04] hi, I tried to run an install script and accidentally installed to partition with not enough space. Now when I try and run ./install again, I get an error of not having enough disk-space (in the old directory) without the installer booting for me to select another destination. Is this a common problem/has a specific name? As I can't find much from googling [21:04] like is there a way to remove all files associated with the old install process? [21:05] ah I've figured it out, no worries [21:06] marxS: ....ok, what I've do is just format the partition and install there, formatting will remove all files in it [21:06] but you're golden :) [21:06] ya I just removed the related files in /tmp [21:06] and it seemed to work [21:09] although its still trying to install a bunch of stuff to /tmp, ive no idea why [21:10] "an install script" does not help us either [21:11] so /tmp could well be used to unpack archives, and when the installation is borked, those would not be removed [21:26] hello everyone ! is there any way to compile a specific kernel module with ubuntu ? I cant find a package named kernel headers or whatever [21:27] In my case I just need to patch an existing module (ufs filesystem) and change one parameter on it === SimonNL_Afk is now known as SimonNL [21:28] there is a build your own kernel wiki page.. [21:28] why i don't have a sound in ubuntu 19.10 fresh install [21:28] everything other it's ok [21:29] oerheks: I dont need the whole kernel, only download c code of that specific module and overwrite the existing one [21:32] snothing muted [21:32] everything higher [21:32] kernel headers package is kernel-headers or kernel-headers-generic [21:33] you probably want src package for linux-image though [21:41] kpoman: linux-headers-generic is what you need to build kernel modules in Ubuntu [21:43] kpoman: Since you want to patch an existing module, you will need source code for linux-image-$(uname -r) linux-modules-$(uname -r) linux-modules-extra-$(uname -r) depending on module [21:45] jeremy31: this module hasnt changed in years, can I just download its code and only the headers of the rest of the kernel ? [21:46] kpoman: It might be possible [21:47] kpoman: The linux-headers-generic should get installed with the kernel [21:47] jeremy31: yes I was looking there in /usr/src/ [21:50] kpoman: Is the driver in /drivers/scsi/ufs [21:52] jeremy31: this one: /lib/modules/4.15.0-70-generic/kernel/fs/ufs/ufs.ko [21:52] I just want to patch its max fragment size from 4096 to 8192 [21:54] kpoman: If you have source code repos enabled in terminal do> apt-get source linux-modules-extra-$(uname -r) [21:59] jeremy31: cool, I downloaded and patched what I want ! Now trying to figure out how to Make it [22:00] kpoman: in the fs/ufs directory, in terminal> make -C /lib/modules/$(uname -r)/build M=$(pwd) modules [22:01] jeremy31: could, it compiled fine ! now do I need to insmod or copy it to /lib/modules/etc... ? [22:02] kpoman: sudo cp ufs.ko /lib/modules/$(uname -r)/fs/ufs [22:04] !cookie | jeremy31 Above and beyond the call of duty [22:04] jeremy31 Above and beyond the call of duty: Wow! You're such a great helper, you deserve a cookie! [22:05] jeremy31: yep thanks ! [22:05] kpoman: cross your fingers and reboot and/or unload and load the module [22:26] how can i remove ubuntu keyring that asks me for some apps sometimes? [22:26] to disable it [22:27] openface, do you have 'seahorse' app installed or it is from something else? [23:02] Hi, why would a usb wifi dongle cause my entire wifi network to go hellwire? I just installed ubuntu and the device disconnects everything in my network [23:02] strangest thing ever [23:03] i just unhooked the usb wifi in order to bo able to connect to irc [23:06] Elodin: can you run the 2 commands at https://gist.github.com/jeremyb31/a2bee9856d8c13f42f1835bc31bf9480 in terminal and post the termbin URL after the second command is done [23:07] kek, ill find a way to usb those scripts to the machine [23:08] Elodin: see https://ubuntuforums.org/showthread.php?t=2082305&p=12350385&viewfull=1#post12350385 and then paste the contents at paste.ubuntu.com [23:10] sheesh that script [23:11] sarnold: I didn't write it [23:14] long time i see that info script [23:24] jeremy31: https://termbin.com/gz9b [23:24] alright that is some pretty output :) [23:26] Elodin: in terminal do> sudo sed -i 's/3/2/' /etc/NetworkManager/conf.d/* [23:26] Elodin: then reboot [23:27] eeeeek [23:27] match-device=driver:rtl8723bs,driver:rtl8189es,driver:r8188eu,driver:8188eu,driver:eagle_sdio,driver:wl [23:27] there's '3's in that line that you probably don't want changed [23:28] Elodin: or do> sudo sed -i 's/wifi.powersave = 3/wifi.powersave = 2/' /etc/NetworkManager/conf.d/default-wifi-powersave-on.conf [23:29] sarnold: I copied chili555's command from a ubuntuforums post. I think all the mac address changing was disabled after 17.10 [23:31] jeremy31: hmm I hadn't heard that -- not that I necessarily would have, but I feel like someone would have filed a bug report about it that I would have seen somewhere along the way :) [23:31] blacklist microcode.. [23:31] ow [23:31] jeremy31: just changed the powersave-on value [23:31] lemme try iinstalling software and see what happens [23:31] now it is intel-microcode or amd-microcode [23:32] Elodin: if nothing else, power cycle the wifi router [23:33] sarnold: bug report https://bugs.launchpad.net/ubuntu/+source/network-manager/+bug/1681513 [23:33] Launchpad bug 1681513 in network-manager (Ubuntu) "Ubuntu 17.04/17.10: New feature in NetworkManager stops several WiFi adapters from working (MAC Address Randomization issue)" [High,Fix released] [23:33] jeremy31: i already did that: power cycle the router [23:34] Elodin: you can look at chili555's other basic recommendations at https://ubuntuforums.org/showthread.php?t=2354328&p=13614520&#post13614520 [23:35] jeremy31: thanks! === Oresrian is now known as stwalkerster [23:48] sarnold: I only remember that bug affecting USB wifi like my ath9k_htc supported TP Link WN 722N