[00:23] ohai [00:24] whats the best--> most/simple VNC to use on ubuntu 20.04 [00:24] not focues on encryption here [00:24] just ease [00:28] probably vino / vinagre [00:30] ty tomreyn !! let me look that up [00:31] ok V|A !! look that up [00:43] uuhhmm errghm === halvors1 is now known as halvors [01:06] I'm using Ubuntu 20.04 and an Intel X540-AT2 NIC. The "ksoftirqd/30" process is freqiently at 100% and I saw poor network throughput so I checked the interrupts and "enp2s0f0-TxRx" is the most frequent interrupt. Is there anything I can check to diagnose why? Throughput is less than 300Mb/s inbound/output and 20Kpps. I did Google about this NIC but found very little apart from a freebsd thread talking about disabling IXGBE_FDIR but I can't find any docs [01:06] on how to do this. [01:06] Note that this is the throughput I'm currently using, attainable throughput is higher but still much below what it should be capable of. [01:14] cuddylier: ethtool is probably the tool to fiddle with the card's settings [01:16] sarnold: I did somewhat assume that might be the case. This is the output I get for 'ethtool -k': https://paste.ubuntu.com/p/K6hkt5n9QF/ [01:19] Looks like IXGBE_FDIR is "ntuple-filters: off" so that is already off [01:19] nice nice [01:19] cuddylier: how about the output of: ip -h -iec -s l [01:19] (I copy-pasted that around ages ago, there may be better arguements, but these are ones I found once .. :) [01:21] cuddylier: the 'nstat' program can dump a ton of statistics about the networking stack; the first run shows cumulative numbers, additional runs will show you what changed [01:21] sarnold: I was running "dstat -t --top-int" which is how I knew the network interface was at fault to begin with [01:22] But that gives no detailed info of course [01:23] cuddylier: nstat's output is vastly different from dstat: https://paste.ubuntu.com/p/Z9YqV7gTmk/ [01:23] Here's the output of this command: https://paste.ubuntu.com/p/fBNcqTnWYT/ [01:24] cuddylier: my suggestion is to try a: while true ; do sleep 5; dstat ; done .. run in one terminal, transfer some stuff, and see what numbers change [01:24] nice nice, no errors, no dropped, no ovverrun. maybe this own't be useful. heh. [01:36] sarnold: https://paste.ubuntu.com/p/2wC9hpv9Gp/ this was during a speedtest [01:36] Download was hovering around 100Mb/s for a while before it eventually went up to 2 - 3Gb/s [01:36] Outbound is no where near as affected as much [01:37] Issue is primarily with inbound [01:38] cuddylier: have you checked power savings settings, (eee) ? [01:41] TJ: Performance mode is enabled: https://paste.ubuntu.com/p/3shsVM4vHR/ and I'm using the tuned-adm virtual-host profile (I tried with it off/balanced but made no difference). [01:45] cuddylier: omg I'm sorry. I screwed up. I meant to suggest running while true ; do sleep 5; nstat ; done -- that way you can see the kernel's networking counters and see if there's anything obviously fishy there [01:47] cuddylier: I've often heard the advice to turn off the offloading and checksumming features in cards, on the theory that they inject enough errors that they add latency to the overall connection. I don't know if those errors would show up in the stats you've already collected or not [01:54] sarnold: here's the nstat outputs: https://paste.ubuntu.com/p/NP6N8D8skz/ The top one is during 'ok' inbound throughput and the bottom two is when it dropped to very poor numbers. [01:55] cuddylier: woot [01:58] do you see anything interesting? [02:01] cuddylier: no :( those IcmpOutDestUnreachs in the middle one stands out a bit, but that could be entirely unrelated.. the TcpExtDelayedACKLost looks remarkably similar among them all, given that there's ten times as much traffic in the first one as the last two.. [02:01] cuddylier: bummer. I was hoping for something to stand out here. === sherlock2 is now known as sherlock [02:03] cuddylier: the bpfcc-tools package has a bunch of tcp tools that might be worth trying. it's just kind of a wild stab, too, but takes more effort than the nstat test: https://github.com/iovisor/bcc#tools [02:07] sarnold: I appreciate your help, thank you. [02:08] cuddylier: my pleasure, I just wish we were having more success :) [02:09] Me too, I guess it could literally be something I'd never track down unless I was a kernel debugging expert. [02:09] X550 doesn't seem to have the same issue [02:30] cuddylier: did you try enabling debug= on the ixgbe module? [02:31] TJ: I haven't, wasn't aware such a thing existed [02:32] cuddylier: "modinfo ixgbe" [02:33] TJ: https://pastebin.com/raw/EuULEcR4 [02:36] cuddylier: yes, I have that here. I was showing you how to see the 'debug' option :) [02:38] TJ: Would I issue the command "modprobe ixgbe debug=16"? [02:42] cuddylier: assuming you've previously successfully unloaded the module, yes. Else you'd need to set the option via an /etc/modprobe.d/ file [02:45] TJ: Thanks. Does the debug output just output to the screen straight away upon enabling debug? [02:47] kernel log === jgrim91055 is now known as jgrim9105 [03:52] Hi. I have connected to a machine with ssh and started apt upgrade. When certain network package was update, the connection was interrupted, however apt is still running. How do I reconnect to apt to get its progress? === MalkbabY_ is now known as Malkbaby === Malkbaby is now known as MalkbabY [05:35] Firefox & MPV had a window header and frame in Ubuntu 21.04 Live Desktop, but after installing and updating they have no window frames. [05:35] So for MPV can't resize etc. [06:47] https://linuxappsummit.org/ t [06:47] https://linuxappsummit.org/ [06:47] * [06:55] https://linuxappsummit.org/ [06:56] bittin: did you have a question? [06:58] nope [07:04] then please don't post random links, this is a support channel [07:16] !offtopic [07:16] #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! [08:28] Anyone using gnote note app? [08:28] !ask | euler2718 [08:28] euler2718: 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 [08:29] is there any way to add gnote app to top bar of ubuntu? like tomboy. [08:29] euler2718: not all apps have indicators [08:30] so its app dependent. we can't do anything from Gnome settings? [08:31] euler2718: gnome shell doesnt provide indicators for individual apps [08:31] maybe through a extension? [08:31] any suggestion? I have searched but did not get anything useful. [08:31] euler2718: you need to find an app that uses an indicator or as Maik says a gnome extension [08:32] euler2718: first hit on Gnome Extensions website: https://extensions.gnome.org/extension/769/gnotetomboy-integration/ [08:33] Hey guys, are there any resources to learn about how does Ubuntu work internally? E.g. what services are used for which part of the system, how do they all fit together, and how to tinker without breaking anything? :) [08:33] argh... actually that's a old one euler2718, not compatible with newer Gnome Shells. My bad [08:33] Maik: Thanks. Let me check that extension [08:34] !systemd | larkfisherman a part of the services [08:34] larkfisherman a part of the services: systemd is the default init system for Ubuntu 15.04 onwards. For information on transitioning from upstart to systemd, see https://wiki.ubuntu.com/SystemdForUpstartUsers For a guide to basic service management with systemd, see https://www.digitalocean.com/community/tutorials/how-to-use-systemctl-to-manage-systemd-services-and-units === bildramer1 is now known as bildramer [08:41] lotuspsychje: Systemd is only a service *manager*. The way I see it, learning about systemd doesn't help me with e.g. figuring out which services I can safely turn off and which have indirect consequences that I need to expect. Apparmor, for example. I see lots of error logs in journalctl related to it, but I have no idea if pruning it from my system will cause some unrelated part to break... [08:42] lotuspsychje: I appreciate the help, though. [09:25] hi, i'm using libreoffice as snap and accessing csv file on remote host (fish::// from krusader) but it libre office complains about permisions [09:25] "access to /home/..../.cache/koexec/krun/......csv was denied" === Skaface82 is now known as Skaface [09:36] morning === Skaface is now known as Cowface [09:52] HI guys my system running xubuntu 20.04 completely locked up this morning and now every time I try to boot I get stuck at stuck at /dev/sda5: clean, 414951/61022208 files, 103882571/244058880 blocks. I booted to the installer usb and ran fsck -f /dev/sda5 but everythings fine. [09:52] Can someone please help me boot back into my desktop without having to do a complete reinstall.... [09:54] Reviziis: is this a classic 'spinning rust' HDD, an SSD, something else? what were you doing when it locked up? [09:54] it's an HDD & I was just trying to open a new tab in Firefox.. [09:55] I done a fresh xubuntu 20.04 install on Monday there. [09:55] install smartmontools [09:55] then use it to diagnose the hdd [09:55] !smart [09:55] smart is Self-Monitoring, Analysis and Reporting Technology, a monitoring system for hard drives. See https://help.ubuntu.com/community/Smartmontools [09:56] you can install it on the live system [09:56] can I install & run that from the live usb ?? [09:56] thanks I'll try that now. [09:57] booting from the usb now. Will report back in a few moments.... [09:57] if the disk is *now* /dev/sdb (i'm guessing here!), you can run sudo smartctl -x /dev/sdb &>1 | nc termbin.com 9999 to share the output here [09:59] if the disk is *now* /dev/sdb (i'm guessing here!), you can run sudo smartctl -x /dev/sdb 2>&1 | nc termbin.com 9999 to share the output here [09:59] ^ i got this wrong the first time around [10:01] Anyone hwo knows how to install woeUSB not ng version been trying all ppa's and they dont work i'm running ubuntu 21.04 [10:02] missing dependecies that cant be downloaded [10:03] oehman: can you not install with snap? [10:10] oehman: not supported here. but there is https://github.com/WoeUSB/WoeUSB/releases/download/v5.1.0/woeusb-5.1.0.bash [10:15] Don't know if my last two messages sent or not, but here's the output >> http://termbin.com/1oo1 [10:17] Reviziis: they didn't. indeed you need to manually post the url here, as you just did. [10:17] Accessing here from my phone so this is all new to me.... [10:19] Reviziis: this western digital disk should be fine. check all the file systems on it using fsck. [10:20] I've already run fsck from my live usb and there was no errors.... [10:22] On the live usb I installed boot repair and here's it's output >> http://paste.ubuntu.com/p/MB23pdcDz7/ [10:23] try mounting the root file system that's on this hdd, then use journalctl with the -D option to point to /var/log/journal on this mount [10:24] this should get you access to the on-disk system log, so you can exmaine what was logged before things broke [10:24] if this sounds too complicated, another option is to try booting a different kernel images from the grub menu [10:24] !grub [10:24] GRUB2 is the default Ubuntu boot manager. Lost GRUB after installing Windows? See https://help.ubuntu.com/community/RestoreGrub - For more information and troubleshooting for GRUB2 please refer to https://help.ubuntu.com/community/Grub2 [10:25] maybe you can also boot to recovery still, and work from the root shell provided there [10:25] !recovery [10:25] If your system fails to boot normally, it may be useful to boot it into recovery mode. For instructions, see https://wiki.ubuntu.com/RecoveryMode [10:26] i don't think bootrepair will help here, since the sysstem did boot fine, fsck was run from the initrd [10:26] Here is the output from journalctl >> https://termbin.com/q553 [10:27] Ok so before I came on here I tried booting from the boot option in recover mode, but all i got was a black screen that never loaded anything...m [10:28] Reviziis: the log you posted is cur off toawards the end, because termbin.com limits how much can be stored. and the log covers three days. you can pipe the journalctl output into tail -n 1000 before you pipe it into nc termbin.com 9999 [10:29] oh sorry. I'll repaste a link now. one sex [10:29] sec×× [10:29] the log you just posted shows that two HDDs were connected. maybe the other one is faulty? did you run smartctl against it, too? [10:30] the other one is external and I have it unplugged ATM.... [10:31] so you'Re saying the system would not boot whether or not the external disk was connected, you tried both? [10:31] Here's what you wanted >> https://termbin.com/kl15b [10:31] Yea I've tried it connected and not connected. [10:32] It only hold pictures and videos. there's no os stuff on it or anything.... [10:32] you have nvidia hardware. try booting in failsafe graphics mode, too, if you haven'T tried this, yet [10:32] how do I do that :? [10:33] it should be an option on the grub menu, unless you installed in legacy bios mode [10:33] !nomodeset [10:33] Systems with certain graphics chipsets may not boot properly out of the box. "Temporarily Add a Kernel Boot Parameter for Testing" as discussed at https://wiki.ubuntu.com/Kernel/KernelBootParameters to add the "nomodeset" parameter there. [10:33] ^ this also works [10:33] yea I'm pretty sure I installed in bios though I'm not 100% [10:34] searh your latest log for "INFO: task kworker". this is a problem. i'm still reading up, though [10:35] this seems to be related to wireless power saving [10:36] ok so would that have caused my lockup earlier & failure to reboot correctly :? [10:37] what we'Re looking at was logged at May 13 09:41:14 in your local time zone [10:37] so this morning. i do not know how this correlated to the events you mentioned [10:37] *correlateS [10:38] those events did certainly delay booting by several minutes, and may have made it look like a complete lockup [10:38] I've never actually ever experienced anything like this before and I've never really had any problems so I'm completely lost.... [10:38] Failed to register: GDBus.Error:org.freedesktop.DBus.Error.AccessDenied: [10:39] snap problem [10:39] Reviziis: did you apply any power saving tweaks before this happened? [10:39] ok so should I reboot and let my system just sit for a long time :? [10:39] I change some brightness setting yesterday morning but that's about it...m [10:40] hey ubuntu snapper, your software is broken [10:41] metbsd: do you have an ubuntu support question? [10:41] snap is not ubuntu? [10:41] Reviziis: if you have a physical wireless kill switch, try switching it, see if you can boot fine then [10:42] metbsd: snapd and snaps are used in ubuntu. you have yet to ask a support question. [10:42] ubuntu is promoting snap store and not helping fix? [10:42] i can't start remmina with that error [10:42] its on one of the function keys. I'll try it now. should I also remove the ethernet cable. ? [10:42] can i have a support fix [10:43] pleeease [10:43] metbsd: see, you can ask a normal question, it's not too difficult. [10:43] metbsd: now let's see whether anyone able to help with this is around. [10:43] but i assume all snaps use gdbus will face the same bug [10:44] Reviziis: no, ethernet works fine [10:44] ok trying now... [10:46] Reviziis: is this a dual boot system? [10:46] it's at /dev/sda5: clean. How long should I have to wait here :? [10:46] i.e. do you have windows installed, and booted into that recently? [10:47] if you workaround had worked the way i imagined, you would not have to wait there. [10:47] if my workaround had worked the way i imagined, you would not have to wait there. [10:47] sorry, I'm still waking up here apparently [10:48] ok so the line cursor below the text is solid. [10:48] now it's just started blinking. [10:48] you can wait some minutes, and answer my other questions in the meantime [10:49] I haven't used windows since 2002. Only run xubuntu on this machine. [10:53] hmm, ok. so i'm assuming that your wireless chipset is in bad state. this can happen when you dualboot, because windows would not shut down cleanly. but it can also be a side effect of another problem, which occurred originally, and made you hard power off the system. [10:54] resetting the bios to vendor defaults (and redoing any changes you require) may help there. [10:54] I can't remember the last time I was in my bios, but I'll go have a look at it now. I'm still stuck on a blinking cursor.... [10:55] !sysrq [10:55] In an emergency, you may be able to shutdown cleanly and reboot by holding down Alt+PrintScreen and typing slowly, in succession, S, U and B. For an explanation, see http://en.wikipedia.org/wiki/Magic_SysRq_key [11:01] you could also try a bios update, this may help the system recovering from this (supposedly, i'm guessing there) bad firmware state. you have a lenovo thinkpad E470 type 20H1CTO1WW, BIOS R0DET86W (1.86 ) 08/31/2017. latest bios for this model is version 2.03 from 2020-07-23 [11:01] https://pcsupport.lenovo.com/de/en/products/laptops-and-netbooks/thinkpad-edge-laptops/thinkpad-e470/20h1/20h1cto1ww/downloads/driver-list/component?name=BIOS%2FUEFI [11:04] Thanks for the link. I'll boot my live usb and try that now.... [11:19] Reviziis: i forgot to ask something: do you have a docking station connected? if so, remove it [11:19] hmm no doesn't look like it [11:20] no I don't have a dock, but I do have a cooling tray plugged into one of the usb ports.... [11:21] because the laptop has overheating issues? [11:22] no overheating issues. the temps shown in conky reach 70 odd when playing some games jist [11:22] i see [11:27] heya, I try to install libtool, but it seems to be not compatible with my version of ubuntu? do I need to use a different package? [11:27] libc6-dev : Depends: libc6 (= 2.31-0ubuntu9.2) but 2.31-0ubuntu9.3 is to be installed [11:29] hello [11:29] cranberry: which version of ubuntu are you running, how did oyu try to install libtool, whats the commmand you ran that created the above output, can we ssee the full output of this command, on a pastebin? [11:29] hi houlei [11:30] cranberry: and di you run apt update before trying to install? [11:30] + sudo [11:32] hi tomreyn, absolutely: https://pasta.lysergic.dev/?6e62058c36cc88ff#5KZZLWtD33Kc8X2AXfPfGG4WokmGUp3MH2EWZxrUV59S [11:32] I did attempt to update too [11:36] cranberry: can you show the output of sudo grep -hEv '^([ ]*#.*)?$' /etc/apt/sources.list{,.d/*.list} 2>&1 [11:37] whatever you ask from me :] [11:37] https://pasta.lysergic.dev/?6e5dc43be728efb1#CDvnjhnNa98Jt5bh5bBNQ46L8ozF3FdWRgHR2kABVLwq [11:39] cranberry: now sudo apt-get -yf install 2>&1 [11:39] please post that output, too [11:39] https://pasta.lysergic.dev/?aa9675bcf4fa4082#AP9r9QBLfkE6rbRvkYbHqutCH7gRyqSit5VU7u1u9gR3 [11:41] tomreyn: I've just opted to do a complete reinstall.... [11:42] Thank you very much for all of the help you have provided :) [11:44] cranberry: did you install any of those packages manually? did you mark them to be held? [11:46] if not, i *think* this can be a bug introduced by bad package management on the mirrors, where 2.31-0ubuntu9.3 was pushed to mirrors, then removed (after people had already installed it). there was a situation like this about a week ago, but i can't find the references to this now. [11:48] tomreyn: of which packages? as for the mirrors.. our current policy is to use the default digitalocean ones in addition with the ones for nginx/docker/teleport but if you suggest switching to alternative mirrors I can ask about doing so [11:48] or can I clean it up by removing and reinstalling some packages? === ghoti_ is now known as ghoti [11:53] cranberry: hmm what i remembered was just about the aRM64 architecture, not aMD64 (which you have), so i hope it's not related [11:54] cranberry: you can maybe try downgrading the libc6 packages, but this can be unsafe, i don't want to recommend this. === ghoti_ is now known as ghoti [11:55] oh ok. hmm should I expect a new libc package to be released? [11:56] cranberry: the problem ther eis that the libc6 package you have has a sliughtly higher version than the libc6-dev package you have available to install [11:58] libc6 is installed in version 2.31-0ubuntu9.3, but libc6-dev, which you are trying to install, is only available in version 2.31-0ubuntu9.2, but depends on this very same version of libc6 [11:58] cranberry: apt policy libc6 libc6-dev may help verify this [11:59] got it.. I tried looking up libc6-dev for ubuntu but the highest I found was for 9.2 [12:00] ye that confirms libc6 is candidate for 9.3 and -dev only for 9.2 [12:00] so I guess waiting for a newer -dev package it is? [12:00] yes, i think the 9.3 was pushed to mirrors, then pulled, because there were problems with it. for usre in arm64 land, not sure about amd64 [12:01] there probably isnt a newer version than 9.2 availöable for both libc6 and libc6-dev now [12:03] ok thanks a lot for the input! [12:05] cranberry: i asked "did you install any of those packages manually? did you mark them to be held?", you asked "of which packages?". i meant packages libc6 and lic6-dev. knowing this, what is your answer to the original question? [12:05] and what's the output of apt list --installed | grep ',local\]$' [12:08] tomreyn: ah, no, I did not. I attempted to install libc6-dev simply because it was a dependency for libtool, but I have not touched anything libc related before that attempt [12:09] https://pasta.lysergic.dev/?7aa6f4fa320a9d45#HZRE8Qv3GNqMn9rxiXPaqpbYtT64eDYZ9WPDWzCiBAK8 [12:09] cranberry: can you show the output of the apt policy command i posted aboive? [12:09] sure [12:09] https://pasta.lysergic.dev/?608bd1afbd923692#F4VnCS8EHdesF6AxJx3YSaUjWyUi7e1RM95Ro6sbp3h1 [12:11] cranberry: you could try this, but it could break this installation if things go wrong, so YMMV: sudo apt install libc-bin=2.31-0ubuntu9.2 libc6=2.31-0ubuntu9.2 locales=2.31-0ubuntu9.2 [12:12] it's a staging system so I'll give it a try [12:14] Howdy all [12:14] tomreyn: hm it looks fine so far! should I attempt a reboot? [12:17] cranberry: let's see whether there are more: apt list --installed | grep -F '2.31-0ubuntu9.3' [12:18] popos. sorry. cant get external hdmi monitor to work. anyone here know if thats a common issue? [12:18] cranberry: does this output anything but the WARNING? [12:18] x260. intel. [12:18] Foxfir3: this channel supports Ubuntu only [12:18] tomreyn: no output at all :) [12:18] cranberry: no WARNING? [12:18] tomreyn: the punishment for being honest [12:18] tomreyn: ah the warning it printed to the console, yes [12:19] my bad, was only looking at the paste output [12:19] Foxfir3: it's good that you'Re honest, it's bad that you assume you could get support with a derivative here [12:19] cranberry: but nothing but the warning? [12:20] exactlya [12:20] tomreyn: well.. its ubuntu, and its debian, and then its not :D More than one issue tuning up with PopOS, so maybe I'll stick with Ubuntu [12:21] cranberry: run sudo update-initramfs -k `uname -r` -u and, if it runs without errors, reboot [12:22] Foxfir3: it's not ubuntu by our definition. you can try to get support in ##linux [12:22] ok! [12:25] tomreyn: run fine! seems everything is working now. thank you very much! [12:25] anything to look out for upon upgrading things? [12:26] cranberry: no, you're now running the versions that are on the mirrors, when a new version is published there, you'll get it as soo n as you install updates [12:26] alright, thanks a lot [12:27] you're welcome. and i'm sorry that you had to run into this situation, which could very well have been prevented by the release team IMO [12:27] you and many, many other people, i guess [12:28] see this post by jochensp on #ubuntu-devel on april 29: https://irclogs.ubuntu.com/2021/04/29/%23ubuntu-devel.html [12:29] but this is now going beyond support [12:33] tomreyn: cheers, I guess it can happen :) [12:36] Hello. After a recent Ubuntu Vanilla update my GNOME dock stopped showing some windows that I open, and it sometimes "remembers" the ones it does show. THey do not get added to Favorites, just hang there and a relogin clears them. They also vanish after some time. [12:36] Have you experienced this problem and/or know how to solve it [12:36] ? [12:37] I've upgraded to 20.11 (Groovy) right before seeing this issue [12:37] agvantibo: do you mean ubuntu, with vanilla gnome? or is there a linux distribution called "ubuntu vanilla"? [12:38] tomreyn: Ubuntu with vanilla gnome, sorry if unclear [12:38] thanks for clarifying, i wasn't sure what you meant. [12:39] !groovy [12:39] Ubuntu 20.10 (Groovy Gorilla) is the 33rd release of Ubuntu. Download at https://ubuntu.com/download - Release notes at https://wiki.ubuntu.com/GroovyGorilla/ReleaseNotes [12:39] you probably mean 20.10 there [12:40] yep, sorry [12:41] just checking... i could not comment on the specific issue, though. if you can take a screen shot or video, this might help attributing the issue to vanilla gnome or a graphics driver issue or similar. [12:41] maybe also try whether this happens with both Xorg and XWayland [12:42] https://i.imgur.com/gAOksrr.png [12:42] Looks like this [12:43] hmm, okay this is likely a bug in gnome then. [12:43] can anyone help me find a broken package?. I'm on 20.4 and get a 'you held broken packages' error but I can't find what package is causing this. I checked synaptic but don't see anything untoward there and 'dpkg --get-selections | grep hold' is also not returning anything [12:44] agvantibo: i assume you can upgrade to to 21.04 now, maybe it's solved there. === ghoti_ is now known as ghoti [12:45] DocMors: run this in a terminal: sudo apt-get update then post the command you ran and the output it produced to https://paste.ubuntu.com and post the final url here [12:46] the upgrade path to 21.04 officially opened up yesterday [12:46] good to know, thanks Maik [12:46] you're welcome [12:48] tomreyn, there are no errors else I wouldn't ask [12:49] where do you find this offending package, what part of Ubuntu is responsible for that flag? [12:49] DocMors: what are you doing that triggers the error message? [12:49] apt install gimp for example [12:50] ok, can you show the full output of this? [12:50] it's not telling me what package is blocking [12:51] *also* show apt policy gimp and apt list --installed | grep ',local\]$' [12:51] tomreyn, https://paste.ubuntu.com/p/vfYdYvhdGC/ [12:51] I switched to sudo -i to ensure no user permission is blocking [12:52] those latest two commands would not need ot be run as root [12:53] that is helpful [13:05] DocMors: so did you solve it then? what was it? [13:08] DocMors: have you tried 'apt install -f'? === ghostcube__ is now known as ghostcube [13:27] I'm looking for a pkg that provides /usr/lib/x86_64-linux-gnu/cmake/Qt53DExtras/Qt53DExtrasConfig.cmake and packages.ubuntu.com tells me it's qt3d5-dev, but I have already installed that pkg, but I don't have that file. how come? [13:28] I have similar related files, e.g. /usr/lib/x86_64-linux-gnu/cmake/Qt53DCore/Qt53DCoreConfig.cmake [13:28] only Qt53DExtrasConfig.cmake is missing [13:29] Igor will talk Ubuntu and Canonical on Linux App Summit now in 1 minute [13:29] ah... wrong suite (I'm in focal) [13:31] and hellsworth [13:31] bittin: do you really need to post it in every channel you're in? [13:32] it's considered cross posting and isn't allowed [13:33] bittin: why don't you post stuff like that in #ubuntu-discuss and there only, it would be a more proper place for it [13:34] sigh... kids [13:35] did not know there was a discuss channel [13:37] pretty sure you've been in the discuss channel before [13:37] Maik: must be so long ago i forgot about it then [13:37] just a few months [13:38] you were in there with your other nick luna [13:38] anyway, that's OT [13:48] I'm trying to mount an xfs partition via fstab with `/dev/sdb4 /mnt/hdd2 xfs rw,noexec,nodev,noatime,nodiratime,largeio,inode64,barrier=0 0 2` which is failing with `mount: /mnt/hdd2: wrong fs type, bad option, bad superblock on /dev/sdb4, missing codepage or helper program, or other [13:48] error.` but if I do a normal `mount /dev/sdb4 /mnt/hdd2` it works fine and xfs_check shows no errors -- anyone know what's up there? [13:59] Hey, I am on Ubuntu 20.10 and for some reason my SSH logins do not result in /etc/profile.d being read. What might be the issue? [13:59] (I am trying to do a multi-user Nix install on the machine) [14:00] I did not touch the original files like /etc/profile (etc profile does iterate over /etc/profile.d/*.sh and my script falls under that category [14:00] ) === ghoti_ is now known as ghoti [14:10] drozdziak1: does your user(s) have access to read and run/source the file(s) in /etc/profile.d/ ? Please test. Also, is your script in there executable? [14:10] tsujp: so it'll be one of those options you're passing, i guess. [14:10] I have another entry with another partition with the same options that is fine though tomreyn [14:10] tsujp: you could try passing them with -o to mount [14:10] that other partition may be in a different format [14:11] run file -s against both, and compare [14:12] identical for both file -s on the partitions [14:12] gives `/dev/sda5: SGI XFS filesystem data (blksz 4096, inosz 512, v2 dirs)` [14:12] and /dev/sdb4 gives the same [14:13] tsujp: try mounting the filesystem with -o and the mount options you're trying to pass with fstab [14:13] it appears `barrier=0` is the culprit [14:13] idk why though [14:13] xfs_info should probably tell [14:14] tsujp: it should be barrier=none [14:14] leftyfb: I have SSH session reuse on so that's a culprit too, but also I'm not sure if I understand profile vs. /etc/bash.bashrc correctly [14:16] drozdziak1: profile gets called first, then /etc/bash.bashrc [14:17] that's unless there are ~/.bash_profile or ~/.bash_login [14:17] oh wait, you were discussing system level, not user level [14:18] leftyfb: tomreyn Yeah, I'm desperate right now and put a snippet sourcing it in /root/.profile and /root/.bash_profile [14:18] might just do bashrc too [14:19] profil is only read for login shells. is it a login shell? [14:19] drozdziak1: why are you logging in as root via ssh????? [14:19] I'm over ssh, but my user somehow doesn't source /etc/profile.d/nix.sh [14:20] leftyfb: I never said anything like that. Also, chill out, it's a dev box for an open-source project [14:20] drozdziak1: please read my original post to you above === Scotty_Trees is now known as Scotty_Mines [14:20] drozdziak1: does your user have access to run/source that file after logging in? [14:22] leftyfb Good question, it doesn't, for my user I gave up and did it in .zprofile [14:23] but then .zprofile might not have interest in being like bash [14:23] drozdziak1: .zprofile is only for zsh, not bash(default) [14:24] Yes, my goal is to go from just me to everyone and /etc/profile appeared perfect for that [14:25] drozdziak1: you know /etc/profile is only called for bash users right? It does not apply to zsh [14:25] drozdziak1: zsh uses /etc/zprofile [14:25] leftyfb: That I didn't know :( [14:26] Seems there probably isn't a one-size-fits-all for this but I'd love to avoid having people write custom snippets to pull in that file [14:27] drozdziak1: copy the script to something like /usr/local/bin and call it from both /etc/zprofile and /etc/profile or .d/somescript === Scotty_Mines is now known as Scotty_TrZzz [14:36] leftyfb Yeah I realize now that this is a sudo issue primarily, and not a shell one [14:36] drozdziak1: what does sudo have to do with anything in /etc/profile* ? [14:38] leftyfb The high-level goal is to include PATH set by /etc/profile.d/nix.sh (created by Nix installer) in all shells so that the commands installed from nix are visible. Any user that wishes to opt into using it can figure it out and I probably should not impose using Nix on them to begin with. [14:39] drozdziak1: I highly agree with your last comment [14:39] The only exception being the ability to do system-wide nix stuff using sudo (like managing available package repos) [14:40] I'm having an half-failing dns situation. "ping domain.com" fails, but "resolvectl query domain.com" returns success [14:40] drozdziak1: any particular reason/package you need to implement nix over just using apt/snap? [14:41] leftyfh Turning our project into a reproducible build is important to me and should enable some nice code reuse [14:42] all in all, sudo `which nix` is not painful to write tbh [14:42] giaco: systemd-resolve --status|grep -i "dns servers" | nc termbin.com 9999 [14:42] drozdziak1: why not make your project into a snap? [14:43] leftyfb: I prefer nix' rigidity and flexibility [14:44] drozdziak1: ok. I think going forward you would be better off seeking support for this topic from the NIX project. I'm sure they have a solution. [14:44] leftyfb Yeah, been there done that [14:44] leftyfb: that commands returns my gateway correctly (10.10.5.1) and it works (tested with dig) [14:44] giaco: it returns only your gateway? [14:45] leftyfb: yes [14:45] giaco: and which public nameserver is your gateway configured to use? [14:45] google 8.8.8.8 [14:46] "dig @10.10.5.1 wwwl.google.com" works correctly from same machine [14:46] giaco: what is the exact error you are getting with ping? Is this failing for all public web addresses or just some super secret domain you wish not to share? [14:47] leftyfb: "ping: wwww.google.com: Name or service not known" [14:47] leftyfb: let me explore a bit more [14:47] giaco: "host google.com" [14:48] WAIT! [14:48] I typed 4 w [14:48] :D [14:48] read the error above [14:48] DUMB [14:53] i'm having a steam issue where steam play titles seem to be crashing instantly when launched. native titles work fine and perform great. [14:57] geosmin: you can also try #ubuntu-steam [14:59] i think it's probably because i'm using the flatpak version. will try valve's .deb === thinkofher0 is now known as thinkofher [15:12] confirmed [15:13] whoho i did it :) [15:13] https://www.omgubuntu.co.uk/2017/06/create-bootable-windows-10-usb-ubuntu [15:16] I try to use firehol with ipsets with a config like this -> https://pastebin.com/YEAAZQn7 But I get this error: "Cannot accept an empty 'inface'. " [15:16] What is meant with empty inface? [15:17] the same firehol.config works on an gentoo machine with no problem at all [15:19] tehuty: I'm going to guess ${wan} isn't resolving for some reason [15:20] leftyfb: I have set to WAN="interfacename" [15:20] Hi all! I've got a little bit of a weird question this morning. On rare occasion, my group has gotten itself into some sort of package versioning issue, where apt reinstalls the same package every time we run upgrade. This particular time, it's happening because we're trying to backport cloud-guest-utils 0.32-1 from hirsute into bionic. If we introduce the package to systems via a private nexus [15:20] repository, things work. if we try and introduce into a repostory that is built and managed by inhouse tooling, the upgrades with every apt upgrade. The only difference we can detect is that our private repository adds a Source: line and a SHA512 that is not present in the nexus repo. [15:20] tehuty: ok, then that's why. You set "WAN", not "wan" [15:22] I guess the simpler question is "what should I be looking at to identify why apt-upgrade successfully reinstalls the same package every time it runs" [15:22] I have two accounts with hosting provider Heroku. I need to run bash commands on one account or the other and I'm logged in via some kind of CLI token thing. Can I create some kind of separate session that will allow me to switch between logins without having to login and logout via oauth/browser, which is a bit of a pain? [15:22] Or should I just use a different ubuntu user account? :) [15:23] mustmodify: sudo su [15:25] That's totally plan B. I was hoping that I could switch back and forth between two heroku users from within my single ubuntu user. [15:26]