[00:04] sarnold , do you think I can find someone to help me with teamviewer to fix that ? [00:06] IRC-Source_60: probably not; you'll have a lot better success if you pastebin both your configs and both log files to a site like paste.ubuntu.com and [00:07] sarnold and ? [00:11] IRC-Source_60: you've got to share the pastebin links once you've made them :) [00:11] sarnold , I am looking for someone to solve that in 10 minutes . I know it is just a little thing [01:02] hi everyone [01:02] so I seem to got Ubuntu 20 working! [01:02] yay [01:02] however, I am now have Ubuntu running on my SSD and I have it running on my regular harddrive [01:02] I have refind installed [01:03] I am trying to go onto the SSD to see how much faster my experience will be but I cannot seem to boot into it [01:03] I don't think I see it on the reFind boot menu [01:03] boot loader [01:03] * [01:03] anything I should be looking for in particular? [01:04] you'll probably need to change grub things instead of efi things [01:04] is there a guide on any of this? I do not work with grub related things that often [01:04] so any guidance would be great [01:09] adam0000345: grub config snippets are stored in /etc/grub.d, there's a higher-level config file in /etc/defaults/grub, and the generated grub config lives in /boot/grub/grub.cfg -- and is overwritten periodically by the contributions of the previous sources [01:11] adam0000345: you could directly edit that /boot/grub/grub.cfg file using the entries that are already there, and add new ones for the different installation, but I think you'd need to add more /etc/grub.d/ configuration snippets to *keep* it long-term, if you desired [01:13] o/ [01:13] hi [01:13] ._. [01:13] wrong channel [01:13] :3 [01:16] sarnold there is not a script I could run that could do this for me? [01:16] I would have to manually edit the file? [01:22] is there a way for my SSD to just be the default disk to boot ubuntu from? [01:22] that might be the easiest option for me [01:23] just a thought [01:25] or what is the easiest way for me to check what partition and what disk I am on [01:25] in terminal [01:26] nevermind i figured it out [01:30] adam0000345: :D - Maybe ' sudo grub-probe -t device /boot/grub ' and ' sudo lshw -C Disk ' will enlighten even more. [01:30] Bashing-om HI bashing [01:30] let me try [01:31] Bashing-om I got ubuntu working on my regular hard drive but now I am trying to boot ubuntu off my installed SSD [01:31] any ideas? [01:31] adam0000345: Glad ya got that Mac intstall figured out :D [01:31] I did for my regular hard drive [01:31] but not for my SSD [01:32] adam0000345: not on a Mac - sorry - no experience with ReFind. [01:32] should I just remove the standard harddrive and force my mac to use my SSD? [01:34] adam0000345: No, do not remove any drives. I too multiboot and is my experience that having ONE grub control all booting instances works best in the long term. But how to do that with an EFI system I have not done :( [01:35] ah maybe I could install grub? [01:35] erm... during `apt autoremove --purge`, I got this: https://bpa.st/OIFQ [01:35] The file descriptors leaking isn't very good, is it? [01:38] adam0000345: Grub "should" be installed on both drivers, yes. I would recommend that the primary booting system be in control of what gets booted. - But again I do not know ReFind nor EFI. [01:39] could I not just replace refind with grub? [01:40] adam0000345: Mac is a cruel master - so far as I know - and no can not use grub directly. - Vendor lock in attempts :( === zbenjamin is now known as Guest6265 === zbenjamin_ is now known as zbenjamin [01:44] hello all [01:51] im having mental problems largely stemming from wifi configuration of some rtl8812bu adapters, virtualbox, Linux zotac 5.8.0-25-generic #26-Ubuntu SMP Thu Oct 15 10:30:38 UTC 2020 x86_64 x86_64 x86_64 GNU/Linux [01:53] im trying to get my system to run as bridge between two wifi networks [01:54] network manager is confusing the hell out of me and im starting to not trust the GUI. I'm considering removing it and netplan and just using systemd. [02:18] I'm seeing that one of the wifi adapters dies when the second one is active. Sometimes (and only when its being a little bitch) one of the virtualbox hosts cant seem to send... I'm not sure where it is dropped yet. Whats the right way to have 10 virtual hosts on one broadcast network with a wifi access point, and another wifi adapter entirely pulling in the internet from another adapter? === benfrancis9 is now known as benfrancis === gabkdlly_ is now known as gabkdlly [02:57] hi, I'm having sleep/suspend/wake issues with Ubuntu 20.04. is there a good resource for troubleshooting this? most of the time, it suspends and won't wake up, so pulling logs from systemd might be impossible. [03:20] signofzeta: the usual approach is to start unplugging hardware or not loading drivers for the hardware you can't unplug [03:28] thanks, sarnold. i've got a lot of usb crud i can unplug. would usb be the problem? [03:28] sometimes, if i wait ten minutes or so, the computer eventually wakes up. no, i didn't think to grab the logs then. [03:32] oh wow [03:32] I wouldn't expect it to ever wake up :) [03:33] signofzeta: yeah, usb crud's a very good first thing to yank, especially wifi devices [03:33] usb wifi would be first thing I yank, anyway :) [03:33] time for me to bail, have fun, good luck :) [03:36] so i closed by laptop lid, and the screen went blank, with a orange light blinking. when i press any key it shows me the login manager [03:36] is this sleep or hibernation? [03:45] sorry i missed you, sarnold, but i paid the extra few bucks for a PCIe Wi-FI 6 device (Intel AX200). now that you mention it, my Wi-Fi has been dying at about the same time this sleep issue started. i have to reboot to get my Wi-Fi to reconnect. [03:46] does `apt update` log what it does? [03:55] signofzeta: sarnold is afk atm, feel free to re-ask your original question into the channel please [03:56] paraphrased: i'm having sleep/suspend/wake issues with Ubuntu 20.04. it worked great at first, now it's problematic. it actually broke at the same time my Intel AX200 (Wi-Fi 6) PCIe adapter started randomly dying. [03:57] normally, I'd journalctl me some logs, but the whole computer freezes up. [04:01] signofzeta: system up to date to kernel -52 ? [04:02] no, only 5.4.0-51. let me be more apt [04:03] !uptodate | signofzeta [04:03] signofzeta: To ensure you have all the latest known patches and security updates for your ubuntu installation, please update with the following command: `sudo apt update && sudo apt upgrade`. See also !upgrades and !security; you may also need to run `apt full-upgrade`. [04:04] thanks, bot. i thought I updated && upgraded earlier today, but I guess not. [04:05] okay, I've got -52 pending a reboot. [04:05] reboot and try again suspend [04:05] unrelated to this problem, will Ubuntu switch to the 5.5 kernel? I've got an AMD fTPM? I recently abandoned Windows 10 and returned to the Linux world. [04:06] signofzeta: ubuntu 20.10 will have kernel 5.8 [04:06] @lotuspsychje: cool, thanks! I'm going to reboot into -52. be back soon. [04:45] signofzeta, FYI at 20.04.2, focal/20.04 will also be using 5.8 if HWE is enabled (the default) === Scotty_Trees2 is now known as Scotty_Trees === auosun8 is now known as auosun === ravustaj1 is now known as ravustaja [06:30] i can't believe this: several years later, and networking is still broken in recovery mode?! It flashes a too-fast-to-read error message just saying something like it's not working, but not what to do. I mean, that's worse than amateur (but it's not like there's an attempt to be like a real Unix) [06:50] hi i am trying to create a symlink from /var/html/www/wodpress folder to /home/user/ftp/wordpress so I run ls -s /var/www/html/wordpress /home/user/ftp but /home/user/ftp is empty [06:56] maret, the first thing I might check is permissions. It should fail with a permission denied error. Where exactly is /var/www/html/wordpress ? [06:58] Also crazy question but have you verified something is in the directory? Might be worth it to do a $sudo touch /var/html/www/wodpress/testfile [06:59] check your logs. Sometimes commands fail silently. That might tell you why [07:11] Hi there [08:05] I have a 8TB external seagate HDD. Formatted as EFI ext4. I connected that to Ubuntu14.04 laptop for backup but there were some data copy I/O errors. So i used gparted and checked the disk. It corrected few nodes. After that I tried to connect the HDD to ubuntu 20/.04 desktop it gives me these errors https://paste.debian.net/1167943/ . What happened? [08:15] "Device offlined - not ready after error recovery" [08:15] it's dead, jim [08:44] mgedmin: ? [08:44] that looks like hardware failure [08:44] ok [08:44] any ideas for recovery [08:44] HDDs are notoriously unreliable; I've had four external HDDs die on me around the time warranty expired [08:45] AFAIU there are recovery services, but they're expensive [08:45] mgedmin: ok, thank you [08:45] gulzar_: I/O errors in the logs is usually bad news [08:47] lotuspsychje: so whole story is- Desktop machine with ubuntu 14.04 was being used by a student. This ext HDD was connected to it. His main HDD and this both went to I/O errors [08:47] lotuspsychje: the int HDD is not showing up (it boots up sometimes after like 10 tries but can' [08:47] hotsoup what should be the permissions of /var/www/html/wordpress? and what logs? [08:47] t can't login as I/O errro on home [08:48] and both int HDD and ext HDD went to this I/O error lotuspsychje [08:48] and its not crazy hh , i did verified contents [08:48] gulzar_: mgedmin just explained you, why are you telling me the story again? [08:48] Hello was curious why my Ubuntu isnt showing up when i do a wireless scan on a seperate device ?? It shows every other device but my Ubuntu :/ . Im not mad but am curious why this is happening and in the past that was not the case. Could it be someone is jacking my connection ? [08:49] lotuspsychje: sorry, my point here is, both int and ext HDD failed with I/O at same time [08:49] that's odd [08:50] Im pretty lost on this. Want to make sure nothing malicious is going on === Tabstar is now known as Tabmow [08:53] Has anyone ran across this before ? [08:54] random1: wireless scan? you mean hotspot? [08:54] a LAN scan [08:56] definitely not hotspot. I have a wireless network here at the house [09:00] Hi. I'd like to know why ubuntu 20.04 does not ship openssh 8.4? because of https://vulners.com/cve/CVE-2020-14145 [09:09] ubuntu backports security fixes instead of updating to new versions with new fixes but also new bugs and incompatibilities [09:09] markuman: ^ and see the notes at https://people.canonical.com/~ubuntu-security/cve/2020/CVE-2020-14145.html [09:09] there are some exceptions to this policy; e.g. browsers get new versions because backporting fixes is too hard and browser developers are trusted [09:11] I want to disable notifications about new system releases. This page: https://askubuntu.com/questions/218755/how-to-disable-the-update-manager-popup contains tons of suggestions, but which is the relevant one today?? Firm knowledge requested. [09:16] mgedmin: tomreyn ok thx === random151 is now known as random2 [09:21] Cymew: by "new system releases", you are probably referring to release upgrades, such as ubuntu 20.04 LTS becoming available while you're on 18.04 LTS? or are you referring any updates becoming available for your current release (security and bug fixes for e.g. 18.04 LTS)? === random2 is now known as random1 [09:29] tomreyn: Release upgrades to a new LTS kind of notifications. [09:29] Cymew: Software & Updates -> Updates -> Notify me of a new Ubuntu version [09:30] select Never from the menu [09:30] Is there a way to script that? I kind of need to do this on a farm of machines... [09:30] or edit /etc/update-manager/release-upgrades [09:30] Ah! That's more like it. [09:30] Come to think of it, I think that was one of the options mentioned in the thread pasted above. [09:31] Right, but if that's the latest way to do it, I will look into that promptly. Thanks a lot tomreyn and Maik! [09:32] yw [09:32] i think it's both the latest and the same method as for the past 8 or so LTS releases [09:42] Good to know. === auosun0 is now known as auosun === PowerTower_121 is now known as PowerTower_120 === bildramer1 is now known as bildramer [11:54] Hey folks, my desktop (20.04) seems to have updated itself last night and broken the nvidia and bcmwl drivers. I managed to get the nvidia drivers back, but the bcmwl is still broken. I don't really know my way around uefi to know how to fix. Any help gratefully recieved! [11:55] modprobe -v wl gives: ERROR: could not insert 'wl': Unknown symbol in module, or unknown parameter (see dmesg). [11:56] Same thing happens when I try and load v4l2loopback which was also working yesterday (and which I haven't touched) [11:58] I've been on nixos for a few years and I'm not really that familiar with developments in ubuntu land. [12:02] boothead: have you built 3rd-party kernel modules from source or something? [12:03] mgedmin - no. I installed bcmwl-kernel-source from a .deb (which I had to do when I set the machine up). AFAIR all I did last time was dpkg -i that same .deb and everthing was good! [12:05] that sounds like a yes to me [12:05] and it seems like the custom broadcom driver isn't using dkms, which would automate recompiling on kernel upgrades [12:06] Ehhh, I suppose it does :-)... What I meant was that I hadn't doing any makeing from souce :-) [12:06] mgedmin any idea what I should do intead? [12:07] avoid hardware that uses broadcom wireless chipsets if you can help it [12:07] on a more practical note, if you dpkg -i the same .deb _again_, will that help again? (until the next kernel upgrade at least) [12:08] I suppose if you modprobe and get symbol resolution errors, that means you've already tried it and it didn't work [12:08] I tried that (several times) and, yeah as you say. I also tried getting the newest version of the .deb [12:09] FWIW I just noticed in history that I installed v4l2loopback-dkms which is also having the same problem [12:09] huh, that one uses dkms and should just work [12:10] unless, of course, the new kernel security update broke api in some incompatible way [12:10] which is not unheard of, but should be unlikely in a security update [12:12] Yeah - I'm a bit lost to be honest. Beyond downloading a .deb and installing it, I'm not sure what to do. And I guess rolling back is also out of the question [12:12] you could always use the older kernel from grub if you really had to boot and have it work [12:15] doesn't ubuntu's driver thing suggest broadcom drivers for you? [12:15] by which I mean the "Additional drivers" tab in software-properties-gtk [12:16] v4l2loopback-dkms is a package in universe, so if it doesn't work, you may want to file a bug [12:16] I see none in https://bugs.launchpad.net/ubuntu/+source/v4l2loopback-dkms [12:17] ah, bad url [12:17] https://bugs.launchpad.net/ubuntu/+source/v4l2loopback has plenty [12:18] the newest one was reported in june [12:22] Thanks mgedmin. Software properties says that it's using the driver [12:26] mgedmin my google foo is failing me - any suggestions on how to get my driver back? The built in wired driver on the motherboard also doesn't work, and it's going to take until thurs to get a usb -> lan adaptor delivered. :-( [12:27] you could boot into an older kernel from the grub boot menu for now [12:32] I tried that too and it wouldn't boot (I only had one older version). Lemme try again [12:34] oh wow [12:42] just a quick quistion i run ubuntu mate, does that belong in this channel or does it have it's own? [12:48] I want to write do something like "sh export MY_VAR=7; echo $MY_VAR" but that syntax isn't correct. can someone help me? [12:51] where do you want to write it? what to you want to achieve? [13:04] yottabyte: the syntax you showed is for bash. sh is dash [13:11] Hi all [13:20] oh [13:27] it seems I just needed to do "sh -c export ...; echo ..." [13:27] Kvik79: ubuntu mate is an official !flavor, as you such can ask questions about it here, too. ideally ask mate specific questions (those about the desktop environment / graphical user interface mostly) in #ubuntu-mate but everything else should be fine here. [13:29] @tomreyn, thx mate :) [13:30] you're welcome [13:40] does anyone know about wsl? I've been on "installing, this may take a few minutes..." for about an hour (ubuntu 20.04 for wsl2) [13:41] !wsl | Baikonur [13:41] Baikonur: 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 [13:44] I remembered ther was a channel, but it's not on the ubuntu wiki irc channel list so I didn't remmeber what it was called [13:53] hi folks, trying to run a vagrant ubuntu machine and getting error https://paste.ubuntu.com/p/RRZYRGvswK/ ... i am initially concerned with the Hash Sum mismatch errors as i cannot fix them even if i clean or remove the repos list [13:53] and then update [13:54] any idea? happens on 18 as well as 20.04 machines [13:54] VMs [14:01] Question: when an update happens, at what point do the new versions of the packages get into the installation media? I have a borked wifi card and I can't install any of the divers from the installation media because libc has bumped to 9.1 and the stuff in the is is built against 9 [14:09] boothead: when a point release is prepared, it includes the (then) current versions previously available as updates. [14:10] tomreyn is there any way I can easily undo an apt upgrade? I'm stuck without a working network at the moment :-( [14:11] i don't think libc would upgrade from 9.0 to 9.1 within a release (or due to point releases). also there is no libc++ 9.1 in any supported ubuntu release [14:12] boothead: which ubuntu release are you using, which third party repositories, ho do you determine the libc version you have? [14:14] maybe you're referring to libc-bin 2.31 patch versions? those increased within focal. [14:14] (obviously patch versions do increase) [14:14] I'm on 20.04. Last night some kind of update happened and broke my nvidia driver. This morning I rebooted and reinstalled the nvidia driver, and did apt updates... After that when I rebooted my wireless card was broken and I haven't been able to fix it [14:15] are you using a self-compiled wireless driver? [14:15] tomreyn yes I think that's it. i've tried downloading the latest version of bmcwl-kernel-source and dpkg -i but I get a message about missing symbols [14:16] and you installed the nvidia driver how? [14:17] sudo apt install nividia-driver-450 [14:18] do you have broadcom-sta-dkms installed? [14:18] No, bcmwl-kernel-source. Should I try that instead? [14:20] may depend on your hardware. which numbers are there given for your wireless chipset in the first line in brackets? lspci -knn | grep -EA3 '(Network|Ethernet) controller' [14:21] make it: lspci -knn | grep -EA3 'Network controller' [14:21] did reinstalling the nvidia driver make those work again? [14:22] BMC4360 [14:22] I have [0280] in brackets [14:22] and [14e4:43a0] (rev 03) [14:22] yes, reinstalling nvidia worked [14:25] good. regarding the wireless, if what https://help.ubuntu.com/community/WifiDocs/Driver/bcm43xx says is still current, you should prefer bcmwl-kernel-source over broadcom-sta [14:25] I just dpkg --remove 'd bcmwl and dpkg -i 'd the one you suggested. modprobe is still telling me about Unknown symbol in module though tomreyn [14:26] did you reboot after installing all upüdates? [14:26] trying that now :-) [14:28] Nope - still complaining about the symbols [14:28] Anyone else having an issue running calibre in 20.04? [14:29] tomreyn not sure how enlightening this is but I get errors exactly like this in dmesg: https://askubuntu.com/questions/893726/modprobe-fails-to-load-module-wl-on-kernel-4-8 [14:30] this was the same with bcml and the one you suggested [14:31] Any idea why watchdog would execute /etc/watchdog.d/foo.sh test, but then never execute foo.sh repair? [14:31] it gets an error and just keeps running test [14:36] boothead: for now, purge broadcom-sta-dkms, reinstall bcmwl-kernel-source. share the commands you're running, the output they produce, and the output of apt policy bcmwl-kernel-source broadcom-sta-dkms broadcom-sta-source broadcom-sta-common as well as of cat /proc/version /proc/cmdline [14:36] and use a pastebin [14:38] I can't actually get the machine in question online at all (the lan port doesn't work either :-/ ) will have a look at saving that to a txt file and uploading from this machine though [14:38] boothead: do you have a phone and a wire to connect it to the computer? [14:38] * smartphone [14:39] tomreyn that didn't work either unfortunately. I didn't see the interface pop up in network manager when I connected usb tethering [14:39] strange. maybe the phone was waiting for you to confirm that you trust this computer [14:40] or yuo hadn't enabled tethering via usb on it [14:41] I did enable it in settings on the phone, just didn't see anything on the computer. I've got a usb->lan adapter coming from amz, but that won't be here till thurs [14:44] This reminds me of the 2000s when I used to get linux from a CD :-D [14:45] yet another option could be connection sharing from another computer. but maybe you should just wait until thiursday then. [14:46] i would very much recommend to always have an alternative means of network access if on-board network hardware only provides badly supported chipsets. [14:52] TIL - I didn't know broadcom was such a shit show otherwise I would have specified a different wireless card when I bought the machine! What would you suggest? [14:54] a chipset which the manufacturer provides liberally licensed, open source, linux drivers for. [14:54] atheros is good, iirc [14:55] lordcirth got any suggestions for a good pci card to buy? [14:58] it's not easy to map chipsets back to specific products you can buy. [14:59] now bcmwl-kernel-source is complaining about libc6-dev... Not quite sure what Ive screwed up now. :-D [15:00] Anyone know how to approve an activity with the "landscape-api" cli tool? My query doesn't get accepted for some reson. I thought I just added an id number... [15:01] I've got 2.31-ubuntu9.1 for libc6-dev. I assume I just have to wait for a bcmwl that's build against that version? Or is there a way I can use the version to filter what I find online tomreyn? [15:05] boothead: normally, just rebuilding the the module via dkms should be sufficient. === bugle is now known as bugLe === gabkdlly_ is now known as gabkdlly [15:24] Nevermind, I figured it out. [15:26] Cymew: for server questions, you'll best /join #ubuntu-server [15:52] tomreyn: Good to know. Thanks! === ianychoi_ is now known as ianychoi [16:27] I'd like to get something like this https://www.amazon.co.uk/HAL-Wireless-Receiver-Streaming-Projector-Black/dp/B08L1Z5LLB/ref=sr_1_4?crid=715CM7B0KLN3&dchild=1&keywords=hdmi%2Bwireless%2Breceiver&qid=1603193714&sprefix=hdmi%2Bwireless%2Bre%2Caps%2C158&sr=8-4&th=1 for receiving my screen wirelessly at a TV/projector. But does Ubuntu support this kind of thing? I'm running 20.04 on a Dell XPS13. [16:30] multifractal: that's just a generic chromecast [16:30] multifractal: you can stream from lots of video streaming services from ubuntu to a chromecast just fine [16:32] leftyfb: Does that mean it'll only accept the contents of my browser window, and not my entire screen? [16:32] It's not the google product though, it claims to support a bunch of input types (e.g. direct from a youtube app on android, airplay - whatever that is). [16:32] looking for a wiki (newer than 14.04) or instructions on how to get reiser4 fs mounted. I can create the filesystem (mkfs.reiser4) but it wont mount, gives an unknown filesystem error. I suspect it's because reiser4 is not in the mainline kernel. [16:33] !reiser [16:33] reiserfs is a journalling file system, which outperforms many others on I/O operations, but has drawbacks (such as increasing likelihood of data loss, and introducing latency unsuitable for gaming or real-time audio). Using !ext4, the default on Ubuntu, is *highly* recommended. A read-only Windows driver is available at http://p-nand-q.com/download/rfstool.html [16:34] multifractal: it's just a generic chromecast. It doesn't look like it does anything more than allow streaming of online video services to it just like the chromecast. With a chromecast you can cast your entire desktop if you like, yes [16:36] thanks for the note on reiser, I am very well aware of what it is, have been using it on gentoo for years. However my desktop is ubuntu and I need to be able to mount some data drives coming out of some of my gentoo servers. [16:43] howudodat: "sudo modprobe reiserfs" and "sudo apt install reiser4progs reiserfsprogs" should help [16:48] TJ- did the modprobe, no error, lsmod shows: reiserfs 249856 0, lsblk shows: sdd1 reiser4, mount /dev/sdd1 /mnt produces: mount: /mnt: unknown filesystem type 'reiser4'. [16:50] howudodat: check the kernel log for any further clues as to why. "journalctl -k -n 20" [16:55] leftyfb: So to be clear, if I got an official google chromecast I'd be able to go to the menu of a chrome browser window, select "cast entire screen", and I'd be able to see my entire desktop on the projector? And would I be able to do that with the generic chromecast, like the one i linked? [16:57] multifractal: yes and I don't know. I have no idea what that generic one is or isn't capable of regardless of what the marketing says. Personally I wouldn't take the chance. Regardless, I know for a fact casting a video, tab or entire desktop does in fact work on a regular Chromecast. [17:00] TJ- the problem appears to be that it's reiser4. if I format the partition as reiser, it mounts just fine. I'm suspecting the module reiserfs doesn't cover reiser4 [17:01] howudodat: it's unclear; the module's Kconfig file talks as if it is reiser4 and gives a URL to the reaiser4 wiki, but it is possible the file system you're trying to mount has features the kernel module doesn't contain [17:02] unfortunately the kernel log doesn't give any clues either [17:02] howudodat: e.g. config REISERFS_FS says "...Read to learn more about reiserfs ..." [17:04] do you store ssh connections in ~/.ssh/config ? [17:04] ash_worksi: context please. Yes, ~/.ssh/config is a place to store profiles and settings for different ssh connections [17:05] tomreyn: heya. remember how I told you yesterday that wonky media center I'm banging together had good RAM by memtest86's verdict? Today pulled one module for S&G and lo and behold: stable operation on benchmark stress test for 3 hours now [17:05] leftyfb: I mean if you frequently connect to hosts foo.example.com something.else.com and another.place.com [17:06] ash_worksi: yes, you can create profiles for those if you like [17:06] woot [17:06] howudodat: seems strange as reiser4progs specifically handles '4' [17:06] thanks leftyfb [17:06] ash_worksi: questions like that can easily be tested. Quicker than asking the question actually. [17:07] is reiserfs still a thing? [17:07] TJ- and I can create and format the partition, just cant mount it :( [17:08] deadrom, not really, why? [17:09] Oh, I see [17:09] deadrom: maybe not reiser but reiser4 is (we use it on all our high speed signal collection servers) [17:10] howudodat, is it really still faster than modern ext4 or xfs? [17:10] howudodat: this explains it https://en.wikipedia.org/wiki/Reiser4#Integration_with_Linux [17:10] "high speed singal collection"? like hunting the next wow signal? [17:11] deadrom: something like that :) [17:11] TJ-: yeah that's what I saw earlier. so I'll go to compiling the kernel module (that's why I dont use gentoo on my desktop) [17:11] definition of a high-speed signal - a traffic light getting caught on a speed camera! [17:11] TJ- LOL [17:16] Ive got to dualkboot Windows10 by necessity. Is Windows still prone to killing grub? I havent been using it for quite a few years [17:16] Reiser's wikipedia page scores in deadbeat humour (snark): "known for: ReiserFS, murder" [17:17] MWM: win10 installer will walk all over it [17:18] thought so. Fix is to chroot and sudo update-grub? [17:18] MWM: if using UEFI mode, no [17:18] MWM: only affects BIOS boots where sector 0 of the boot device MUST contain the boot-strap code [17:19] deadrom: yeah the jokes about reiser are endless, putting aside whatever I might think about the man, his work on filesystems was remarkable. I use ext2/4 depending on my needs but we have probably close to a petabyte of data on reiser4 [17:21] howudodat: now I'm curious what you actually do. [17:22] !ot | deadrom [17:22] deadrom: #ubuntu is the Ubuntu support channel, for all Ubuntu-related support questions. Please register with NickServ (see /msg ubottu !register) and use #ubuntu-offtopic for other topics (though our !guidelines apply there too). Thanks! [17:22] and yes, it is a shame, I know it has strengths - maybe the people workign on it should change the name. [17:22] howudodat: a petabyte on *reiser*? wow [17:22] deadrom: unfortunately I cant give a lot of details, but we do wideband RF signal capture and analysis [17:23] howudodat: reiser3 ate my tree once while my machine was mostly idle over the weekend.. I showed up to work monday morning and it was toast. I reinstalled with extsomething :) heh [17:25] is it possible to configure ssh connections based on what network I am (not) connected ot? [17:25] to* [17:27] like, if I am at work, and on the office network, then I should be connecting to 10..., but if I am at home I want to try dev.example.com [17:27] with just `ssh dev` [17:30] how do you guys manage your firefox package? best ppa i found is the mozillateam one with firefox-next which only releases cutting edge versions. the default apt repo is a few major releases behind. any "latest stable" best practice i can use? [17:30] leftyfb: when was channel administration replaced with channel policing? second time in 2 days I get jumped at for ot. [17:30] ever since the fire nation attacked? [17:30] the who? [17:31] ash_worksi: yep [17:31] deadrom: nevermind. just a silly joke [17:31] deadrom: if it's the 2nd time, then maybe keep on topic. Feel free to chat about non-support issues in #ubuntu-offtopic [17:32] edd_lc: jokes are not allowed in here anymore it seems [17:34] a few major releases behind? i'm on 81.0.2 which is 1 version behind, and that's not bad when 82 was only released today [17:34] and , btw, 82 isa laready available as snap [17:34] ash_worksi: how? [17:35] i don't like the snap as it doesn't have the widevine DRM module [17:46] hmm, i must be doing something wrong. ok, thanks [17:57] Here's a fun one - we have very large (in file count) directories on a CephFS. Since metadata write operations are expensive, recursively deleting them causes significant slowdowns for other IO for hours. Is there a decent way to have a script recursively delete files *slowly*? [17:57] I am thinking something along the lines of find -exec sleep $X; rm [17:58] wow that's an interesting situation [17:58] IKR? [17:59] We have been moving deleted directories to a holding location outside the user-accessible space; this is O(1) and atomic. I'm hoping to automatically drain those away over time [17:59] lordcirth: execdir, never exec! [17:59] mason, right [18:00] lordcirth: Aside from that, I can't think of a better solution. [18:01] smacking together a python or perl script to do the deletes slowly wouldn't be horrible, but getting it *right* might take some effort.. [18:02] you could recompile find to add a delay after each delete, and use -delete [18:02] (don't change the system one, just do a one-off build) [18:02] I don't want to recompile anything, and the easy ways to write a python script would probably use a ton of memory holding the path list... I suspect I can get normal 'find' to do it [18:03] that'd keep a pretty good tool that has already had all the complex corner cases handled, the stuff that'd be easy to get wrng in your own tooling.. [18:04] Yup, there we go: find . -print -exec sleep 1 \; -delete [18:04] deletes one file a second. If I set that to loop over the temp directory, it should drain away files slow and steady [18:06] lordcirth: omg find already does this? nice :) [18:07] Yeah, turns out you can specify -execdir and -delete at the same time, no problem... [18:07] lordcirth: cool cool :) thanks [18:26] I'm trying to figure out where systemd-resolved is pulling its nameserver information and putting it in /run/systemd/resolve/resolv.conf [18:27] kur1j: it's best to read systemd-resolved(8) -- it's very complicated :( [18:28] its annoying as hell lol [18:28] we have a VPN that gets passed a DNS server [18:29] it works 90% of the time, but somehow systemd-resolved will swap the DNS server from the one the VPN passes to the one the LTE device gets from its DHCP server [18:29] breaking things [18:48] Is there still no upgrade path from 18.04 to 20.4.1? [18:49] Hello, someone told me about port knocking. I am looking for something simmilar: a programm that bans any ip (via ip tables) that tryies to connect to a certain port. any hints what the correct word is for that, or does anyone know a programm that does that or help me to achive this? [18:51] qwebirc95665: are you sure what you want actually makes sense? what problem are you trying to solve? [18:52] the problem i want to solve is that i plan on running an nginx server and currently all ports are closed) [18:52] but i see already a lot of traffic on my firewall [18:53] so my plan since portscanner are usualy the first tool someone would take, before triing anything else i want to make a few ports trap ports that closes someone out [18:53] qwebirc95665: if the ports are closed, then there's no traffic and nothing to worry about [18:54] yea but an nginx server only on the inside is not that fun [18:55] qwebirc95665: look at fail2ban. You'll have to come up with some way to log every single attempt on every single port on your system. (don't) [18:55] and since a normal browser would only go to port 80 a mallicions attemt would most likely beginn with 1 and go up [18:56] qwebirc95665: it's not worth your time [18:56] what would you do instead? [18:56] to secure everything up? [18:57] qwebirc95665: only open a port to what you plan on serving out to the public, leave all the rest closed and nothing running and forget about them [18:57] i have an openwrt router and 2 ubuntu server, and 2 laptops [18:57] and on one i want a website, and on the other one ssh and vpn [18:58] what can i do to secure everything up? [18:58] qwebirc95665: I just told you [19:01] qwebirc95665: use ssh keys, disable password authentication in the ssh server [19:01] ok [19:01] qwebirc95665: install ufw and make sure only ports you want open are open [19:01] qwebirc95665: and consider using apparmor to confine any internet-facing services [19:02] thx [19:43] has anyone tried running ubuntu on the acer swift 3 (this one: https://www.acer.com/ac/en/US/content/model/NX.HULAA.003)? [19:44] marbar, why, are you having trouble? [19:44] lordcirth, nah, i'm shopping around for a new lappy (this one is falling apart in too many ways) [19:45] i just want a system the works as well as possible right off the bat (to avoid wasting time) [19:46] marbar: this one works with ubuntu https://www.dell.com/en-us/work/shop/dell-laptops-and-notebooks/new-xps-13-developer-edition/spd/xps-13-9310-laptop/ctox139w10p2c3000u [19:46] ooo yes leftyfb i was considering that one [19:46] i actually have an xps 13 - it's pretty broken [19:47] but i was gonna get the new one anyway [19:47] until i realized it's super expensive compared to some of the swifts [19:47] https://unix.stackexchange.com/questions/609715/how-to-install-debian-on-acer-swift-3-sf314-42 [19:47] marbar: looks like only a $100 difference. [19:48] leftyfb, but also that dell comes with an intel chip, which (supposedly i guess) is way worse than the beefy octacore in the swift [19:49] tomreyn, that makes sense, i was reading somewhere it doesn't work well without a newish kernel [19:49] and stock debian is 4.19 right? [19:50] oh that's what the guy says in the comment [19:50] marbar: i think so, they'd know in #debian, but i think this post means things should work out of the box on ubuntu 20.04.1 LTS [19:51] tomreyn, yeah i seem to be getting that impression [19:51] (maybe not everything) === jess is now known as jess-o-lantern [20:12] is there a way to get the current terminal tab profile name (using command line) [20:18] sruli, the terminal would have to export that somehow [20:59] Hi1 [20:59] I have a script that outputs a LOONG generated key. If I copy and paste it with my mouse it works and I can use this token to log in to my app [20:59] if I use `print_key | xclip` and paste then it won't work :S [20:59] what up :D [21:01] This doesn't make sense at all, I'll keep investigating. [21:01] nvm [21:36] Hello i have problems installing savage 2 game, where can i find help? [21:39] hearit: if they don't have much support community this or askubuntu are a good start [21:39] hearit: it's possible other people here have installed it too, or it's possible you've got a generic error [21:39] libstdc++.so.6: version `GLIBCXX_3.4.11' not found (required by /lib/x86_64-linux-gnu/libicuuc.so.67) [21:39] allright [21:39] i will try here first [21:40] it ive me that error, when execute the game [21:42] i have a brand new laptop with ubuntu 20 [21:42] just my system crashed [21:42] i wanna find out what happened [21:42] check logs rubato [21:42] gather info [21:42] what steps did u made before the system crash? [21:43] i was only visiting a webpage [21:43] nothing much [21:43] checking syslog [21:44] nothing special there [21:44] what other logs could i check [21:45]