| pcfreak30 | gah... still hanging | 00:00 |
|---|---|---|
| shinobi_ | Anyone have an answer? https://askubuntu.com/questions/1289496/is-possible-to-integrate-zfs-to-the-desktop-managers-recycle-bin | 00:00 |
| pcfreak30 | sarnold: my nic is on the mobo, so no changing that :D | 00:01 |
| pcfreak30 | WTF | 00:01 |
| pcfreak30 | sarnold: it JUST changed to enp4s0 | 00:02 |
| sarnold | pcfreak30: are there any messages in dmesg that explain this? | 00:02 |
| pcfreak30 | nothing that stands out | 00:04 |
| ruedii | pcfreek30 are you using dhcp? You might need to reduce your renewal time. | 00:04 |
| pcfreak30 | im using static dhcp assignment router side | 00:04 |
| pcfreak30 | it works live cd | 00:04 |
| pcfreak30 | so i can keep ip space central | 00:05 |
| ruedii | Yeah, increase your renewal speed on your client side. | 00:06 |
| pcfreak30 | networkctl lists enp4so as unmanaged, and br0 as no-carrier, confifuring | 00:06 |
| pcfreak30 | enp4s0 | 00:06 |
| pcfreak30 | configuring* | 00:07 |
| pcfreak30 | router has unlimited dhcp leases, and static dhcp is no lease time set | 00:08 |
| pcfreak30 | just manually deleted leases from router | 00:09 |
| pcfreak30 | hmm got it | 00:12 |
| pcfreak30 | deleted the mac on a dynmic lease and used netplan + dhclient | 00:12 |
| pcfreak30 | got a pong | 00:12 |
| pcfreak30 | ping | 00:12 |
| pcfreak30 | unsure why it did a diff ip in dhcp, but :shrug: | 00:12 |
| pcfreak30 | sarnold: im either being forgetful or crazy as somehow the mac address changed vs the static assignment | 00:17 |
| sarnold | pcfreak30: woot | 00:19 |
| pcfreak30 | sarnold: this is getting stupid | 00:29 |
| sarnold | pcfreak30: oh, IU thought you got it fixed | 00:29 |
| pcfreak30 | on manual yes | 00:30 |
| pcfreak30 | every time i change cloud init, it changes the nic to the opppsite | 00:30 |
| pcfreak30 | ens2 vs ens4 | 00:30 |
| pcfreak30 | rofl | 00:30 |
| pcfreak30 | i purged dhcp assignments and rebooted that | 00:31 |
| sarnold | pcfreak30: that's so strange -- can you file a bug report on cloud-init? | 00:31 |
| pcfreak30 | if i knew the cause... | 00:31 |
| pcfreak30 | im going to add them both in. if i get ens6 or end8.... | 00:31 |
| pcfreak30 | going to pill my hair out | 00:32 |
| pcfreak30 | btw recovery modes networking still seems to be broke, doesnt use netplan. | 00:35 |
| pcfreak30 | sarnold: its STILL hanging... is there a way to interrupt the boot or get to a tty? | 00:36 |
| sarnold | pcfreak30: I've not tried in the last ~decade or so but ^C used to be able to do that | 00:37 |
| pcfreak30 | yea systemd as init mucks with that. and i cleared dhcp again | 00:38 |
| pcfreak30 | im very close to just nuking cloud init | 00:38 |
| pcfreak30 | i dont need it | 00:38 |
| pcfreak30 | yup foobaring it. | 00:39 |
| pcfreak30 | sarnold: this is crazy | 00:46 |
| pcfreak30 | if i boot normally it hangs | 00:46 |
| pcfreak30 | if i go toot, and use netplan apply then dhclient, , exit and boot normal, it works? | 00:47 |
| pcfreak30 | root* | 00:47 |
| sarnold | pcfreak30: I wonder why the dhclient dhcp client works so differently than the built-in systemd dhcp client | 00:50 |
| sarnold | pcfreak30: is there anything in the journal from those failed attempts? | 00:50 |
| pcfreak30 | sarnold: it seems like its fighting dhcp between br0 and the main nic. | 00:56 |
| sarnold | pcfreak30: how many nics do you have? | 00:57 |
| pcfreak30 | 1 hardware, 1 bridge | 00:57 |
| sarnold | ah! | 00:57 |
| sarnold | wait | 00:57 |
| sarnold | is that NIC in the bridge? | 00:57 |
| === PowerTower_121 is now known as PowerTower_120 | ||
| pcfreak30 | yes. | 00:58 |
| pcfreak30 | its a kvm server | 00:58 |
| sarnold | aha | 00:58 |
| sarnold | remove the dhcp from the nic | 00:58 |
| pcfreak30 | it is | 00:58 |
| pcfreak30 | thats whats wird. dhcp4: false | 00:58 |
| pcfreak30 | i just fixed a config error, testing it | 00:59 |
| sarnold | when you put a nic in a bridge, the nic kinda loses itself -- no IP on the nic, no dhcp on the nic, etc. IP / dhcp etc happens on the bridge for some reason | 00:59 |
| pcfreak30 | it makes sense, the bridge becomes the public face, the nic is the physical lie | 00:59 |
| pcfreak30 | line | 00:59 |
| pcfreak30 | i said br0 a few time lol | 01:00 |
| sarnold | sorry, I missed it an hour ago | 01:01 |
| pcfreak30 | config error fixed it. now i just need to get static dhcp back | 01:01 |
| pcfreak30 | and hope it doesnt break again | 01:01 |
| pcfreak30 | but yea it seems that cloud-init does funky things that causes the nic to play musical chairs | 01:02 |
| pcfreak30 | sarnold: https://pastebin.com/1EDGa7a2. thats the config. thanks so far for your time. the pointers are what solved this :) | 01:04 |
| sarnold | pcfreak30: I'm a pretty decent rubber duck :) | 01:04 |
| pcfreak30 | i got a lot more to do and more config stuff tomorrow. i was afriad i would be up all night with this | 01:05 |
| === pcfreak301 is now known as pcfreak30 | ||
| pcfreak30 | ok looks like things are "normal" again. thanks sarnold | 01:11 |
| pcfreak30 | kudos also Bashing-om :) | 01:11 |
| Bashing-om | pcfreak30: Just looking over yall' | 01:14 |
| Bashing-om | s shoulders :D | 01:14 |
| sarnold | pcfreak30: excellent :D thakns for reporting back, good night, and good luck tomorrow :D | 01:14 |
| pcfreak30 | yea no kidding. i have a hardware shipment coming in to match what came today | 01:15 |
| * pcfreak30 gets to work | 01:16 | |
| === Grldfrdom_ is now known as Grldfrdom | ||
| === Grldfrdom is now known as Grldfrdom_ | ||
| === de-facto_ is now known as de-facto | ||
| dylanchase | Hi, I'm trying to install postgresql and I get the message "re-enter pam_mount password". How do I know what my pam_mount password is? | 02:09 |
| tomreyn | how do we know which ubuntu and postgresql you are running? | 02:11 |
| dylanchase | ah sorry ubuntu 20.4 and whatever postgres is in the repos | 02:12 |
| sarnold | wow I've never seen that message before; how'd you get there? | 02:13 |
| tomreyn | do you have a custom pam configuration? | 02:13 |
| tomreyn | https://unix.stackexchange.com/questions/120762/why-does-pam-mount-ask-for-password discusses this issue, but it's not recent and not ubuntu specific | 02:13 |
| dylanchase | sarnold: just did sudo apt install postgresql and postgresql-contrib | 02:14 |
| dylanchase | tomreyn: I'm not sure, maybe. I'm not familiar with pam_mount but perhaps I configured it at some point. | 02:15 |
| sarnold | what does grep -r pam_mount /etc/pam.d/ report? | 02:17 |
| dylanchase | grep -r pam_mount /etc/pam.d/ | 02:21 |
| dylanchase | oops | 02:21 |
| dylanchase | /etc/pam.d/common-auth:auth optional pam_mount.so | 02:22 |
| dylanchase | /etc/pam.d/common-session:session optional pam_mount.so | 02:22 |
| dylanchase | /etc/pam.d/common-password:password optional pam_mount.so disable_interactive | 02:22 |
| sarnold | dylanchase: one of the stack overflow answers on tomreyn's link suggests adding disable_interactive on the auth and session management tasks | 02:27 |
| tomreyn | FWIW, "sudo grep -r pam_mount /etc/pam.d/" returns nothing on my 18.04 LTS system (which doesn't have postgresql installed). | 02:30 |
| tomreyn | i assume this wont have changed for 20.04 LTS | 02:30 |
| tomreyn | so your pam_mount configuration is probably custom | 02:31 |
| Bashing-om | tomreyn: confirmed: "sysop@2004x-c:~$ grep -r pam_mount /etc/pam.d/" returns to prompt. | 02:32 |
| dylanchase | tomreyn: Thanks. Not sure what I did to set it up! I did configure disk encryption when I set up 20.4. I'm not sure if pam_mount is related to that... | 02:33 |
| tomreyn | "pam_mount - A PAM module that can mount volumes for a user session", maybe dylanchase needed to mount /var/lib/postgres or something else this way? | 02:33 |
| sarnold | perhaps assuming /root or postgres's home dir may require ecryptfs unlocking? | 02:34 |
| tomreyn | see the use cases discussed here: http://manpages.ubuntu.com/manpages/bionic/man8/pam_mount.8.html | 02:34 |
| sarnold | or nfs? | 02:34 |
| dylanchase | I do connect to an nfs, maybe that is when I configured it. | 02:35 |
| PipeItToDevNull | How can I edit the grub menu when using grub-efi? | 02:51 |
| Bashing-om | !details | PipeItToDevNull | 02:52 |
| ubottu | PipeItToDevNull: Please elaborate; your question or issue may not seem clear or detailed enough for people to help you. Please give more detailed information; for example, we might need errors, steps, relevant configuration files, Ubuntu version, and hardware information. Use a !pastebin to avoid flooding the channel. | 02:52 |
| === zbenjamin is now known as Guest48110 | ||
| === zbenjamin_ is now known as zbenjamin | ||
| === de-facto_ is now known as de-facto | ||
| newlinux^ | hi, how can I use chainproxy for a specific application? | 03:08 |
| newlinux^ | and not all other the system | 03:08 |
| sarnold | most applications that support http proxies let you set the proxy via an environment variable 'http_proxy' or something similar | 03:09 |
| newlinux^ | spcks5? | 03:11 |
| newlinux^ | socks5? | 03:11 |
| newlinux^ | btw chinproxy is a app that force others to use proxy | 03:11 |
| az | hi, how do I configure LUKS password period, currently it waits for few seconds before continue booting | 03:38 |
| sarnold | strange, thought mine would wait forever.. | 03:40 |
| az | it's encrypted home. | 03:40 |
| az | so it can continue booting but wouldn't login | 03:41 |
| sarnold | az: I *think* cryptsetup refresh -t ... can do it | 03:44 |
| az | any where to read about that? | 03:45 |
| sarnold | az: you may or may not need the --persistent flag -- I found this command, and a note about --persistent, in crypttab(5) | 03:45 |
| az | thank you | 03:46 |
| az | will check | 03:46 |
| sarnold | there's gobs of docs in cryptsetup(8) but it's .. not fantastic. I couldn't tell what commands, if any, --timeout applies to. | 03:47 |
| dylanchase | Hey all, thanks for your help. I got my pam_module issue resolved using the disable_interactive command. | 03:55 |
| sarnold | dylanchase: nice :) | 03:59 |
| Vatum | what does /run and /var/run do? ejabberd from apt installs something there, but using the deb file from upstream does not. | 05:06 |
| vysty | I used an external HD to create an ubuntu boot drive for a new computer. I want to reformat the external drive so I can use it as a backup. My new computer isn't recognizing it when I plug it in. How can I a.) get my new computer to recognize it and b.) format it? Running 20.04. | 05:08 |
| esro | hey, am looking for elegant way of disabling local acccounts ssh login | 05:10 |
| esro | i want only PAM users to be accepted | 05:10 |
| bindi | someone should probably take note that panorainer is sending discord links in private to join some tech community or whatever. | 05:55 |
| marcoagpinto | panorainer is a spammmer | 07:26 |
| marcoagpinto | he is sending links with viruses in private | 07:27 |
| === denningsrogue6 is now known as denningsrogue | ||
| leumashm | Hi i have an ubuntu server, and we are trying to replicate the same on another ubuntu server. So we are trying to find out what are installed on the previous server, so we can install on the new server. anyone know what command should i use? | 07:31 |
| Seveas | leumashm: dpkg -l | grep ^ii | 07:32 |
| Seveas | or dpkg --get-selections | 07:32 |
| TJ- | leumashm: are you installing to the same Ubuntu release version? | 07:33 |
| leumashm | TJ- apparently no. older is 18.04, newer is 20 | 07:34 |
| leumashm | cause some web development was done on the old server. but we want to move to the cloud but not sure if we do a migration to the cloud with the existing hdd 1TB, would it provision the entire 1TB or used space 30GB. so we trying the alternate route to find out what is installed in the old server, and install on the new server instead. | 07:36 |
| TJ- | leumashm: you'll be better off using 'debfoster' then, since it identifies the top-level packages but doesn't list their dependencies (often versioned library packages) | 07:36 |
| TJ- | !info debfoster | leumashm | 07:36 |
| ubottu | leumashm: debfoster (source: debfoster): Install only wanted Debian packages. In component universe, is optional. Version 2.7-2.1 (focal), package size 35 kB, installed size 158 kB | 07:36 |
| TJ- | leumashm: using debfoster is a 2-stage process. 1) scan and build the database "sudo debfoster -q" 2) list the top-level packages "debfoster --show-keepers" (if I recall correctly. Check "man debfoster") | 07:38 |
| leumashm | TJ- im a bit worried to run it on my production server ubuntu. i ran on my local ubuntu it does give me some outputs. | 07:43 |
| TJ- | leumashm: why are you worried? | 07:45 |
| leumashm | TJ- cause the lead developer passed away. im just an infra guy. | 07:46 |
| summonner | programming is hazardous to your health | 07:47 |
| TJ- | leumashm: you're not aiming to install or remove packages on the source system, simply create a database listing all the top-level 'keepers' and the report that list which you can pass to "apt install" on the new 20.04 host (I'd test it locally in 20.04 virtual machine or container) | 07:49 |
| leumashm | TJ- thanks that was helpful. im just scared. but managed to get the list of items installed on the server~ Thanks! | 07:56 |
| strk | I've /etc/resolv.conf pointing to 127.0.0.1 and netstat tells me it's dnsmasq listening on port 53. I can't resolv names | 08:09 |
| strk | how do I tell what server is dnsmasq using ? | 08:09 |
| strk | nmcli tells me DNS servers: 192.168.1.1, and that server *does* resolv names | 08:10 |
| strk | logs contain: dnsmasq[4288]: using nameserver 127.0.0.53#53 | 08:11 |
| strk | is dnsmasq querying itself ? | 08:11 |
| strk | does that after: dnsmasq[4288]: no servers found in /run/dnsmasq/resolv.conf, will retry | 08:11 |
| strk | who's in charge to write servers into /run/dnsmasq/resolv.conf ? | 08:12 |
| TJ- | strk: 127.0.0.53 tells you systemd-resolved is being used | 08:20 |
| TJ- | strk: dnsmasq shouldn't be operating and wouldn't be expected to be referenced in /etc/resolv.conf. We'd expect /etc/resolv.conf to be a symlink: /etc/resolv.conf -> ../run/systemd/resolve/stub-resolv.conf | 08:21 |
| TJ- | strk: what release of Ubuntu is that? | 08:22 |
| strk | TJ-: Ubuntu 18.04.5 LTS | 08:23 |
| strk | symlink (from 2013!) is 2013 /etc/resolv.conf -> ../run/resolvconf/resolv.conf | 08:23 |
| strk | good hint about .53 being systemd, I didn't know that | 08:23 |
| TJ- | strk: hmmm, so possibly gone through some release-upgrades from 12.04 ? | 08:24 |
| strk | interesting, I didn't notice | 08:24 |
| strk | tcp 0 0 127.0.0.1:53 0.0.0.0:* LISTEN 4288/dnsmasq | 08:24 |
| strk | tcp 0 0 127.0.0.53:53 0.0.0.0:* LISTEN 3352/systemd-resolv | 08:24 |
| strk | TJ-: yes, many upgrades | 08:24 |
| strk | a frankstein | 08:24 |
| TJ- | strk: that'd be why - release-upgrades don't replace such things | 08:24 |
| strk | and I'm an old linux user, who missed lot of recent changes | 08:24 |
| TJ- | strk: is it a desktop or server ? | 08:24 |
| strk | it was so simple in slackware times, to just manually edit /etc/resolv.conf :P | 08:25 |
| strk | desktop | 08:25 |
| TJ- | strk: and could break so easily too :) | 08:25 |
| strk | I recently slowley learnt some nmcli | 08:25 |
| strk | easy break and easy repair | 08:25 |
| strk | like all simple things | 08:25 |
| strk | anyway, what's this systemd resolv thing ? | 08:25 |
| strk | is dnsmasq using systemd upstream or vice-versa ? | 08:26 |
| TJ- | strk: OK, with Desktop it'll be using NetworkManager which, for 18.04, does have a Recommends: dnsmasq-base | 08:26 |
| strk | I have both dnsmasq and dnsmasq-base installed | 08:26 |
| strk | file /run/dnsmasq/resolv.conf points to nameserver 127.0.0.53 | 08:27 |
| strk | which now makes more sense, now that I noticed .54 ~= -.1 :) | 08:28 |
| strk | so dnsmasq uses systemd upstream | 08:28 |
| strk | then the next question is: how do I know what upstream is systemd using ? | 08:28 |
| strk | because my main problem was with being unable to resolv | 08:28 |
| strk | I learnt that adding a proper upstream in /etc/resolvconf/resolv.conf.d fixes things | 08:29 |
| strk | BUT, if my /etc/resolvconf/resolv.conf.d/head does NOT contain a nameserver, then the resulting /etc/resolv.conf (upon sudo service resolvconf restart) points to nameserver 127.0.0.1 | 08:29 |
| strk | which is dnsmasq, not systemd-resolver | 08:29 |
| strk | is systemd-resolved using /etc/resolv.conf ? | 08:30 |
| strk | as input, I mean | 08:30 |
| strk | my /etc/rc1.d/K02dnsmasq: echo "nameserver 127.0.0.1" | /sbin/resolvconf -a lo.$NAME | 08:31 |
| TJ- | so it looks like you've got a mix which is causing issues. I'm unsure why/how systemd-resolve was enabled when NetworkManager is installed and expects dnsmasq-base :s | 08:31 |
| strk | can I get rid of it ? | 08:32 |
| strk | I'm always happy to collect garbage :) | 08:32 |
| TJ- | strk: *usually* what happens is, NM manages an instance of dnsmasq(-base) and when a connection comes up it 'tells' that instance of dnsmaq the upstream DNS server IP addresses. (In later Ubuntu release NM uses systemd-resolve and not dnsmasq) ... I'm not sure quite how you've ended up with NM > dnsmasq > systemd-resolve :S | 08:33 |
| TJ- | strk: It's so long since I used 18.04 Desktop I'm not sure what the correct solution is either | 08:34 |
| TJ- | strk: in your scenario I assume what is happening is NM's instance of dnsmasq is 'told' the DNS IP addresses but instead of using them it is asking the local systemd-resolve on 127.0.0.53 | 08:35 |
| TJ- | strk: does this give clues as to what dns resolver NM is controlling? "ps -efly | egrep 'dns|resolv|NetworkManager' " | 08:37 |
| strk | .about NM it only tells me: /usr/sbin/NetworkManager --no-daemon | 08:48 |
| strk | and then both systemd-resolved and dnsmasq processes exist | 08:49 |
| strk | but it's likely systemd being controlled | 08:49 |
| strk | uhm, actually I don't really know :) | 08:49 |
| strk | but I see a dnsmasq.d/ directory under /etc/NetworkManager | 08:50 |
| strk | (empty) | 08:50 |
| strk | this may be interesting: | 08:50 |
| strk | NetworkManager.conf.dpkg-old:dns=dnsmasq | 08:50 |
| strk | Sep 26 2014 NetworkManager.conf.dpkg-old | 08:50 |
| strk | so, in 2014, NetworkManager managed dns via dnsmasq | 08:50 |
| strk | Dec 4 2019 NetworkManager.conf # as of 2019 there's no "dns" key in that file | 08:51 |
| strk | current NetworkManager.conf is short: http://strk.kbt.io/tmp/NetworkManager.conf | 08:51 |
| * strk autoremoving --purge | 08:58 | |
| TJ- | strk: I'm not sure what to reccommend. I've been looking in an 18.04 container, and that shows NM has the Recommends: on dnsmasq-base but no references to systemd-resolve. *if* the default in 18.04 is to enable systemd-resolved.service then that would explain what you're seeing | 08:58 |
| strk | TJ-: I'll start by dropping dnsmasq | 08:59 |
| strk | ok, dnsmasq is now gone | 09:00 |
| strk | now /etc/resolve.conf only has nameserver 127.0.0.53 | 09:00 |
| strk | and that server seem to resolve DNS just fine | 09:00 |
| TJ- | strk: easiest way would be to start a VM and see what the 18.04 desktop installer in 'Try Ubuntu' mode has/does | 09:00 |
| TJ- | strk: ah... and is that a symlink or an actual file? "ls -l /etc/resolv.conf" | 09:01 |
| strk | /etc/resolv.conf -> ../run/resolvconf/resolv.conf | 09:01 |
| strk | (same symlink as before I removed dnsmasq) | 09:01 |
| strk | I might want dnsmasq back one day, but for now I'll do w/out | 09:03 |
| strk | (or does systemd-resolved also cache?) | 09:03 |
| TJ- | they both do basically the same thing in this context | 09:04 |
| strk | so both do cache ? | 09:08 |
| Onepamopa | Why would this not work: //192.168.254.83/Downloads /mnt/Windows_Downloads cifs uid=myuser,credentials=/home/myuser/.smbcredentials,vers=3.1.1,iocharset=utf8,sec=ntlm 0 0 | 09:56 |
| Onepamopa | running mount results in mount error(2): No such file or directory | 09:56 |
| summonner | no such file or directory. it doesn't exist to the remote machine. are you sure the source and destination are available? | 09:58 |
| Onepamopa | and CIFS: VFS: Error connecting to socket. Aborting operation. in dmesg | 09:58 |
| Onepamopa | Yes they are, I opened the same location from another windows 10 pc w/o any issues | 09:58 |
| Onepamopa | with the same user/pass | 09:58 |
| summonner | check the event log, it might say something more about the cause? has that machine connected to another resource previously? or first time | 09:59 |
| Hamidreza | Hi, what is cloudInit in ubnutu 20 ? | 10:02 |
| Hamidreza | because I can not configure it as template (and the error says disable cloudinit) | 10:02 |
| Hamidreza | If I disable it , what would happen? | 10:02 |
| summonner | event log->/var/log/messages or something (sorry, thought I was on a windows channel for a second) | 10:03 |
| === khfeng_ is now known as khfeng | ||
| locsmif_work | Hi all. I can't find the dpi settings for my multi-monitor setup in my system settings.. where can I find it? | 10:37 |
| oerheks | there is no DPI setting, but Scaling. | 10:39 |
| locsmif_work | oerheks: I think I found it in font settings, I switched no some kind of force setting. I moved my workplace due to corona so now I have to reconfigure.. it's a bit of a mess. | 10:41 |
| locsmif_work | Trying something, brb | 10:41 |
| EdLin | what version of openzfs does Ubuntu 20.10 include? | 11:16 |
| oerheks | !info zfsutils-linux | 11:24 |
| ubottu | zfsutils-linux (source: zfs-linux): command-line tools to manage OpenZFS filesystems. In component main, is extra. Version 0.8.3-1ubuntu12.5 (focal), package size 345 kB, installed size 1266 kB | 11:24 |
| oerheks | https://wiki.ubuntu.com/ZFS | 11:24 |
| oerheks | !info zfsutils-linux groovy | 11:25 |
| ubottu | zfsutils-linux (source: zfs-linux): command-line tools to manage OpenZFS filesystems. In component main, is extra. Version 0.8.4-1ubuntu11.1 (groovy), package size 351 kB, installed size 1279 kB | 11:25 |
| ubusr | hi, I have an ubuntu 18.04 system that when booting, it seems stuck in the init phase | 11:25 |
| ubusr | and I can't get a login prompt even in text | 11:25 |
| ubusr | ideas? | 11:25 |
| ubusr | (i can get to root prompt via recovery shell) | 11:26 |
| EdLin | oerheks thanks. | 11:26 |
| ubusr | any1 ? | 11:28 |
| ubusr | join #linux | 11:31 |
| Maik | !patience | ubusr | 11:32 |
| ubottu | ubusr: Don't feel ignored and repeat your question quickly; if nobody knows your answer, nobody will answer you. While you wait, try searching https://help.ubuntu.com or https://ubuntuforums.org or https://askubuntu.com/ | 11:32 |
| Maik | ubusr: also stop spamming to join other channels | 11:37 |
| ubusr | i forgot a / relax | 11:39 |
| daysun | is running the command unattended-upgrade in a script hourly enough to have it function as good as it would otherwise if you scheduled it like it's done usually (i.e. activating the right unit files and so forth)? | 11:43 |
| daysun | (It's going to check if a service needs a restart). | 11:45 |
| bart- | can anybody tell me whats the big diff. when i install a package via root or via sudo with apt-get install? | 11:54 |
| bart- | hello btw. | 11:55 |
| Kon | Does the default journald config save the log of previous boots? Had an overnight crash but only seem to be able to read this session's log | 11:56 |
| Kon | Nevermind, it was a bug with the GUI frontend. Logs are accessible through CLI | 11:59 |
| rastan | is there a program that will allow you to set custom nvidia gpu fan curves? | 12:01 |
| ravanan | I am having trouble installing Ubuntu 20.10 in Legacy mode. I clearly selected Legacy Boot when I booted my Dell Latitude E5450 with my bootable USB stick. Whenever I try crossing the partitioning step in the installer, I get a warning about not having created an EFI system partition. As far as I know, I don't have to create one for Legacy boot. I | 12:02 |
| ravanan | checked /sys/firmware and it doesn't have anything related to EFI. If I go ahead with the installation, ignoring the warning, my grub-install fails. What should I do? | 12:02 |
| ubusr | I have this Xorg segfault described here: https://askubuntu.com/questions/1259689/ubuntu-18-04-stuck-in-boot-segmentation-fault-xserver in "/var/log/Xorg.0.log.old" what can I do ? | 12:04 |
| BluesKaj | Hi folks | 12:15 |
| SteelRose | Hi all | 12:20 |
| === mIk3_09 is now known as mIk3_08 | ||
| daysun | hi 50% of everyone | 12:32 |
| pizzaiolo | hello 50% of daysun | 12:36 |
| daysun | touche! | 12:40 |
| daysun | the other 50% of daysun doesn't deserve it. daysun has been trolling today :( | 12:41 |
| === mateen1 is now known as mateen | ||
| clarkk | I have enabled noise cancellation for the default mic in my ubuntu 18.04 system. Is there any way to make it work for an external, bluetooth microphone too? Here's the instructions I followed https://www.informaticar.net/enable-noise-cancellation-in-ubuntu/ | 13:51 |
| pepee | so, I wanted to install chromium-browser from apt-get. it got stuck at "INFO Waiting for restart", so I had to kill it. now snapd is downloading... something | 13:54 |
| pepee | how do I check what snap is downloading? | 13:54 |
| sgaur | /quit | 13:55 |
| Macroman | My pc keeps freezing a couple of times a day. I can see segfaults in my syslog at crash time (https://gist.github.com/MacroMan/daf60f4edc50f0f3a032547711d41a68). Trying to figure out what's causing this | 13:56 |
| Macroman | The main culprit seems to be VizCompositorTh which looks to be part of Chromium, but would that be crashing my whole pc? | 13:57 |
| pepee | ... and if I stop the snapd service, it restarts automatically and keeps downloading... something, lol. nice concept of control, security and usability, canonical | 13:57 |
| Foxfir3 | rpi4 ubuntu 20.10 server. Cant log in :( any tips? | 14:27 |
| Foxfir3 | waited for 10 minutes, like some suggested. it is just really strange. ubuntu/ubuntu is the root user? | 14:27 |
| Maik | Foxfir3: on first boot it's ubuntu/ubuntu then you are asked to change the password | 14:28 |
| Maik | re-flash the card and start over | 14:29 |
| Foxfir3 | Maik: thanks. tried all sorts of combinations. will re-flash. it does seem a little shaky though | 14:32 |
| waveform | Foxfir3, cloud-init takes a while to run on the pi; that's what sets up the initial user (along with whatever else it's been told to do - the configuration for it is the "user-data" file on the boot partition where you can customize the initial user, add ssh-import-id stuff, etc.) | 14:33 |
| waveform | as for why the login prompt is permitted to appear before cloud-init has run: that's a complex issue but the tl;dr version is: it's possible to come up with a cloud-init config that creates a user then gets stuck doing some long network operation - if the login prompt is delayed until c-i is done then the admin can be left unable to login even though the user exists at that point | 14:35 |
| waveform | I'd love to have slightly more fine-grained control over this so we could launch getty (the login prompt) once c-i had definitely created users but was still working on "other stuff", but sadly we don't have that option at the mo | 14:36 |
| === zbenjamin is now known as Guest66139 | ||
| === zbenjamin_ is now known as zbenjamin | ||
| deki_01 | hi there | 14:42 |
| Foxfir3 | waveform: cloud-init failed. fallback to.. didnt remember to what. so I will try to reflash. | 14:43 |
| Foxfir3 | ran manjaro and fedora on this pi, so i find it a little strange that such a bug exist | 14:44 |
| deki_01 | is it possible to install remotely a ubuntu OS,could not find something newer | 14:44 |
| waveform | Foxfir3, once you've reflashed, before you boot the card take a look at "user-data" on the boot partition; there you can customize the name & password (and whether the password is expired) for the initial user. However, note that applies to first boot only - it won't change things if modified after the first boot | 14:45 |
| Foxfir3 | waveform: thn | 14:46 |
| Foxfir3 | waveform: thanks. i actually just made that mistake. | 14:46 |
| Foxfir3 | i run ubuntu server on vultr. Zeo | 14:47 |
| Foxfir3 | zero isues with that | 14:47 |
| waveform | I must admit one of the annoying things about cloud-init is that if you muck up the config you potentially wind up with a broken system (e.g. no default user), but then the idea is you experiment until you get the config you want, then you can easily stamp out as many instances/machines as you want with that config by just copy/pasting it | 14:48 |
| waveform | (it's just that experimenting is a bit more painful on the pi given the reflashing involved, than with instances in the cloud) | 14:49 |
| Foxfir3 | waveform: manjaro arm (of all things) is the best i have run on the rpi4 so far. I dont like the cloud-init feature. it should be optionable and not default | 14:53 |
| waveform | while cloud-init has its rough edges (the getty issue being the most visible), it's effectively a standard for initializing cloud images (and the ubuntu pi server image is derived from the ubuntu cloud images). To make it optional would effectively double the server images to those with and those without c-i | 14:56 |
| sgen | So thunderbird wont stop opening. I dont even use it but it keeps opening over and over. Any idea why or how I can stop this? Its making my computer hang. | 14:58 |
| pavlos | sgen: if you dont use it, why not remove it? | 15:00 |
| === zbenjamin_ is now known as zbenjamin | ||
| sgen | Thats a good point. Im more worried that it wont stop opening. Its just strange. Might be a key bind but Im not sure how or which ... | 15:03 |
| === zbenjamin_ is now known as zbenjamin | ||
| === zbenjamin is now known as Guest94509 | ||
| === zbenjamin_ is now known as zbenjamin | ||
| === zbenjamin is now known as Guest9862 | ||
| === zbenjamin_ is now known as zbenjamin | ||
| Vatum | can someone explain to me how pid files in /var/run are actually used? a program i use in its config file has the option to use a pid file in /var/run or disable it altogether. and in different distros its either active or not | 15:17 |
| === zbenjamin_ is now known as zbenjamin | ||
| === Sven_vB_ is now known as Sven_vB | ||
| Foxfir3 | waveform: thanks. i actually just made that mistake. | 15:31 |
| Foxfir3 | waveform: everything working now. login issue is a bug. its possible to enter user and pass before the python script has been run | 15:32 |
| waveform | Foxfir3, indeed - and a long standing one (https://bugs.launchpad.net/ubuntu/+source/cloud-init/+bug/1861128 <- the work-around mentioned in there is present in the images but doesn't appear to work) | 15:36 |
| ubottu | Launchpad bug 1861128 in cloud-init (Ubuntu) "getty.target starts before cloud-config is done" [High,Triaged] | 15:36 |
| Foxfir3 | ubottu: you are a robot. you can fix it | 15:37 |
| ubottu | Foxfir3: I am only a bot, please don't think I'm intelligent :) | 15:37 |
| Foxfir3 | ubottu: you are girlfriend material | 15:37 |
| ubottu | Foxfir3: I am only a bot, please don't think I'm intelligent :) | 15:37 |
| Maik | Foxfir3: stop it please | 15:37 |
| Foxfir3 | waveform: worked just to wait for it. So they could fix it with a simple warning in the shell | 15:37 |
| lumpybump | Hey guys, I've got ubuntu 20.04 live on a usb. Wondering if it's possible to use the standard installation wizard to put it on another usb with a persistent volume. Or should I just use mkusb? | 15:38 |
| oerheks | use mkusb | 15:38 |
| lumpybump | alright | 15:39 |
| lumpybump | so the installation wizard only lets you install on the hard drive? | 15:39 |
| oerheks | no, one can select the 2nd usb to install too. | 15:40 |
| waveform | Foxfir3, that was something I attempted (way back when this was first reported) but there's something that clears the screen at some point and before I could figure that out other more pressing things got in the way. If I get some time this cycle I'll see if I can re-visit it though | 15:40 |
| === zbenjamin is now known as Guest25922 | ||
| === zbenjamin_ is now known as zbenjamin | ||
| lumpybump | oerheks then why use mkusb instead of install wizard? | 15:41 |
| lumpybump | does it not let you put persistence? | 15:42 |
| Foxfir3 | waveform: change the userscript so the password isnt temporary? | 15:42 |
| Foxfir3 | waveform: no. you metioned that wont work | 15:43 |
| waveform | you can make the password non-temporary in user-data but doing so is a security risk, so I can't make it the default (has to be "opt in") | 15:43 |
| oerheks | lumpybump, mkusb for the install iso + persistence. usb to usb for 'normal ' install, which will be horribly slow | 15:44 |
| waveform | Foxfir3, basically there's no "universally good" solution at the moment, but I should have another look at a warning message as it's a decent enough mitigation that *should* be simple (although it wasn't entirely when I first looked at it) | 15:44 |
| lumpybump | aha I get it | 15:44 |
| Foxfir3 | waveform: dammit. i see. well.. it works. Ubuntu desktop, cloud server and rpi4 server. all ubuntu'ed up. you can imagine the remarks i get from old Arch users :D | 15:45 |
| lumpybump | but what's the difference between 'normal' install and iso + persistence? Maybe it's a stupid question, but I've only ever really used ubuntu in live sessions. | 15:46 |
| lumpybump | oerheks ^ | 15:46 |
| oerheks | this live/install iso does not have the latest kernel fixes and other updates. | 15:46 |
| Foxfir3 | waveform: anyone at Cannonical to fix it? They should have programmers working from Isle of man | 15:47 |
| oerheks | use it with care. | 15:47 |
| Foxfir3 | waveform: it has to be low level. Asking 'do you want to use cloud-init?' | 15:48 |
| waveform | Foxfir3, there's pretty much no way to do something like that - you can't guarantee any particular interface on boot (c-i is designed for completely headless remote systems), so it'd come down to having with-c-i and without-c-i images and the release team are highly unlikely to go for that :) | 15:50 |
| === zbenjamin_ is now known as zbenjamin | ||
| lumpybump | oerheks so a normal installation has kernel fixes/updates and the live/iso aren't always up to date? | 15:50 |
| oerheks | jups | 15:50 |
| Foxfir3 | waveform: i see. so basically it would have to be two separate images | 15:50 |
| Foxfir3 | waveform: or the fix you attempted | 15:51 |
| waveform | Foxfir3, as to who's bug it is to fix: almost certainly mine - I'm the guy responsible for the pi images at canonical. Unfortunately, while it's on my list it's pretty low down (below things like "get bluetooth working on the CM4" :) | 15:51 |
| Foxfir3 | waveform: i see :D well.. good to know there is a list. didnt test dual monitors or sound yet. will be nice with bt. very sweet. | 15:52 |
| Foxfir3 | waveform: feels like Ubuntu is 'coming back' | 15:53 |
| waveform | Foxfir3, dual monitor does work, but there's a whole pile of edge cases with at the moment (most of which we can't currently do much about as they're "upstream firmware" issues, e.g. the way full-screen audio doesn't work unless it's on HDMI0 instead of HDMI1). Sound should work fine, but the cert team did alert me to an issue with the head-phone jack which I've failed to replicate so far | 15:55 |
| oerheks | Foxfir3, install gnome-tweak-tool for ore settings :-) | 15:55 |
| oerheks | c/more | 15:55 |
| tpw_rules | hello all. i have a bluetooth joystick on ubuntu 20.04 (xbox 1 s emulation type using the kernel drivers) and i've got it paired and everything works fine. however, the first time i start up my computer and connect it, nothing works. it appears in /dev/input but nothing can read from it. e.g. xf86-input-joystick, any SDL apps, etc. jstest can however. if i disconnect and reconnect it everything is good again. | 16:01 |
| ol500nix | #archlinux | 16:08 |
| Foxfir3 | waveform: i see. so basically minor issues. where to report issues? github? | 16:10 |
| Maik | ubuntu issues are reported on launchpad Foxfir3 | 16:12 |
| ogra | Foxfir3, on a sidenote ... there is only one ubuntu developer on the isle of man .. thats mark shuttleworth 😉 | 16:35 |
| Foxfir3 | ogra: https://businesstech.co.za/news/government/90990/why-shuttleworth-lost-his-r250-million-tax-battle/ | 17:21 |
| lumpybump | Quick question guys. On this post (https://askubuntu.com/questions/80455/no-root-file-system-defined-error-while-installing-ubuntu) user Saidi Awad advises to use ext4, ext3, or ext2 but NOT FAT32/NTFS when installing Ubuntu on a partition. | 17:21 |
| lumpybump | Why are ext filesystems preferred? | 17:21 |
| lumpybump | Thanks! | 17:21 |
| TJ- | lumpybump: not FAT32 since it doesn't support attributes for ownership or access (modes) | 17:22 |
| Foxfir3 | lumpybump: exfat is commercial | 17:22 |
| Foxfir3 | lumpybump: ntfs should be free though | 17:23 |
| TJ- | lumpybump: and although NTFS does with ACLs its support on Linux and access patterns don't lead to much confidence | 17:23 |
| TJ- | lumpybump: unless there are over-riding reasons, choose an in-kernel file-system | 17:24 |
| lumpybump | alrighty, thanks! | 17:25 |
| === ijohnson is now known as ijohnson|lunch | ||
| pcfreak30 | sarnold: I just had to put enp1s0 in the netplan b/c the nic switched again. So i have enp1s0 and enp4s0 | 18:17 |
| sarnold | pcfreak30: oh man :( I had really hoped this was all sorted :( | 18:18 |
| pcfreak30 | its up, but i had to go manual again to fix it | 18:18 |
| pcfreak30 | idk why its deciding to keep renaming | 18:18 |
| pcfreak30 | at this rate ill be bridging 1-10 | 18:18 |
| === ijohnson|lunch is now known as ijohnson | ||
| === BrianG61UK__ is now known as BrianG61UK_ | ||
| Rojola | hi | 19:10 |
| Rojola | I know, that I installed a package in "snap" but I fail to uninstall it | 19:11 |
| Rojola | how can I please find out the real name of the package? | 19:11 |
| Rojola | "snap remove gnome" won't work | 19:11 |
| Rojola | I ran "df -h" and saw this line: | 19:11 |
| Rojola | /dev/loop5 163M 163M 0 100% /snap/gnome-3-28-1804/145 | 19:11 |
| === BrianG61UK__ is now known as BrianG61UK_ | ||
| sarnold | Rojola: 'snap changes' may report something | 19:14 |
| Rojola | thank you, sarnold, but that only shows recent events | 19:15 |
| sarnold | dang | 19:15 |
| Rojola | $ sudo snap remove gnome | 19:15 |
| Rojola | snap "gnome" is not installed | 19:15 |
| Rojola | ^ that is my problem | 19:15 |
| hofer | snap list | 19:15 |
| Rojola | I found it! | 19:16 |
| Rojola | snap remove gnome-3-28-1804 | 19:16 |
| Rojola | sorry for bothering you guys! | 19:16 |
| Rojola | thank you sarnold | 19:16 |
| Rojola | thank you hofer | 19:16 |
| sarnold | Rojola: aha! I hadn';t thought of that :D | 19:19 |
| sarnold | thanks | 19:19 |
| vicatcu | hey all, I recently updated to 20.04 and I've run into some trouble with an older software that depends on libssl.so.1.0.0 and libcrypto.so.1.0.0; can anyone tell me if it's possible and how to add these shared libraries to ubuntu 20.04 for the benefit of installing my legacy software? | 19:20 |
| sarnold | the easiest thing might be to use lxd to run an older version of ubuntu in a container | 19:22 |
| vicatcu | interesting, what's lxd? | 19:24 |
| vicatcu | found it https://linuxcontainers.org/lxd/introduction/ | 19:24 |
| KonstaapeliNalle | Hi, I do have funny problem while installing Ubuntu. I'm trying to install multiboot system with Win10. This is about 5 yr old ASUS laptop, and UEFI is used, no fastboot etc problem found. There was some problems I got fixed, but now Ubuntu Live USB boots Win 10 when I try boot it :D What the heck? I don't know what other things to tell, so please | 19:25 |
| KonstaapeliNalle | ask :) Thank you! | 19:25 |
| blb4393 | vicatcu: I'd create symlinks first to 1.1 versions of both libs Maybe it'll do if api is backward compatible | 19:26 |
| vicatcu | sarnold, so is lxd appropriate for running GUI programs? The software I need to install is an older version of EagleCAD | 19:26 |
| vicatcu | blb4393, oh that's an interesting idea | 19:27 |
| sarnold | vicatcu: good question :) it'd certainly be more complicated than most folks use of lxd | 19:29 |
| blb4393 | there's systemd-nspawn nowadays | 19:29 |
| blb4393 | if symlinking won't work | 19:30 |
| KonstaapeliNalle | Ye, will go and try again :D | 19:39 |
| vicatcu | blb4393, where should I put the symlink? | 19:42 |
| vicatcu | /usr/lib64? | 19:42 |
| sarnold | /usr/lib/x86_64-linux-gnu -- if that matches your architecture. check dpkg -L libssl1.1 for the location of your current library | 19:43 |
| vicatcu | k | 19:43 |
| sarnold | openssl 1.0.1 was famous for breaking a lot of things, so I'm skeptical, but maybe if they just used it for something stupid like local license checking... | 19:44 |
| vicatcu | that seems to have worked | 19:46 |
| vicatcu | yes, exactly local license checking | 19:46 |
| sarnold | blb4393: sweeeeeeet :D | 19:46 |
| vicatcu | thanks much | 19:46 |
| sarnold | masn that's way easier than lxd, hehe | 19:46 |
| vicatcu | oh no! ./eagle: symbol lookup error: ./eagle: undefined symbol: CRYPTO_num_locks | 19:48 |
| vicatcu | feels like i should just be able to go get the .actual libcrypto.so.1.0.0 file and put it on this machine | 19:49 |
| vicatcu | instead of aliasing 1.1 -> 1.0.0 | 19:49 |
| sarnold | the risk there is not getting security updates for it | 19:50 |
| sarnold | but you can grab an older one by following links from https://launchpad.net/ubuntu/+source/openssl | 19:50 |
| sarnold | and dpkg -i the thing | 19:50 |
| blb4393 | vicatcu: the old libs could be compiled for different libc version but worth to try | 19:50 |
| sarnold | look for the libssl* packages, not the openssl package, that's just the binary front end | 19:50 |
| vicatcu | hah, that worked | 19:58 |
| vicatcu | i just copied the so files from an 18.04 computer | 19:58 |
| maciek | hi there, regarding the https://ubuntu.com/security/CVE-2021-3156, it says that a patch for 20.04 was already released, but I can't find it in the repo… | 20:00 |
| sarnold | maciek: http://security.ubuntu.com/ubuntu/pool/main/s/sudo/ | 20:02 |
| sarnold | maciek: the security updates are copied to the archive mirrors in batches, if it's not on them yet, then the batch hasn't happened | 20:02 |
| blb4393 | maciek: hm, if i do apt update && apt upgrade it offers me to upgrade sudo | 20:02 |
| maciek | blb4393: probably the mirror thing as sarnold said | 20:03 |
| maciek | I'm using the polish one (pl.archive.ubuntu.com) | 20:03 |
| sarnold | you should also have the security.ubuntu.com listed in your sources | 20:04 |
| blb4393 | this is what's in the update log | 20:04 |
| blb4393 | * SECURITY UPDATE: heap-based buffer overflow | 20:05 |
| blb4393 | - debian/patches/CVE-2021-3156-pre1.patch: sanity check size when | 20:05 |
| sarnold | https://launchpad.net/ubuntu/+mirror/ubuntu.task.gda.pl-archive -- "This mirror was last verified 16 hours ago. " -- so, sixteen hours ago, this mirror looked up to date | 20:05 |
| sarnold | that's encouraging, some of the mirrors can be a few days out of date :( | 20:05 |
| maciek | huh | 20:05 |
| sarnold | not everyone who runs a mirror keeps on top of it | 20:06 |
| maciek | I've added: deb http://security.ubuntu.com/ubuntu focal main restricted universe multiverse | 20:07 |
| blb4393 | i have http://archive.ubuntu.com/ubuntu in my sourses.list | 20:07 |
| maciek | and… it's still not there | 20:07 |
| gst568923 | Hi, I have created a xfce launcher (run in terminal) and I have write this command: `sudo renice -n -20 -p $(pidof twinkle)` but not works, why? | 20:08 |
| sarnold | gst568923: in what way does it fail? | 20:09 |
| sarnold | maciek: hmm, did apt report which IP address it contacted? I checked the four ipv4 addresses I can reach and all four have the sudo update | 20:11 |
| gst568923 | sarnold I prepended the following command to prevent the terminal from exiting and not showing me the error | 20:14 |
| gst568923 | sarnold `xfce4-terminal --maximize --hold -x sudo renice -n -20 -p $(pidof twinkle)` | 20:15 |
| maciek | sarnold: no, but I dont really know why I'm using the polish mirror anyways, so not an issue anymore after changing to the nova.clouds.archive.ubuntu.com | 20:15 |
| gst568923 | sarnold renice: bad process ID value: $(pidof renice: bad process ID value: twinkle) | 20:15 |
| sarnold | gst568923: aha! :) $() is a shell thing, and apparently this command isn't running through a shell. try this... xfce4-terminal --maximize --hold -x sh -c "sudo renice -n -20 -p $(pidof twinkle)" | 20:16 |
| gst568923 | sarnold great, it works; to understand what sense does it make to put "sh" if xfce4-terminal is already present? | 20:19 |
| sarnold | gst568923: xfce4-terminal is probably doing it's own command-line parsing to avoid going through the shell. So, if you want to use shell features in your command, you'll need to call the shell yourself | 20:27 |
| sarnold | gst568923: this is getting more common; eg systemd unit files don't use the shell by default. If you want shell features in systemd unit files, you need to call the shell explicitly there, too | 20:27 |
| gst568923 | sarnold if I run for example the command "top" inside xfce4-terminal, I see in the system monitor this tree: xfce4terminal --> bash --> top | 20:32 |
| sarnold | gst568923: huhn. that's curious. | 20:32 |
| sarnold | gst568923: wait, do you mean, you typed 'top' into a shell that was running inside an xfce4-terminal? | 20:32 |
| sarnold | gst568923: or did you construct a .desktop file for it, like this renice command? | 20:33 |
| gst568923 | sarnlod no .desktop file I opened xfce4-termianl and I writed for example "top" command, then I run `gnome-system-monitor` and I have see the tree xfce4terminal --> bash --> top | 20:35 |
| RoseBus | encryptfs is opensource, but if I install it using apt am I garunteed it wasn't altered? | 20:36 |
| RoseBus | does apt install exactly as seen from github? | 20:36 |
| sarnold | gst568923: aha! then in that case the terminal has run a shell for you to use :) | 20:37 |
| sarnold | RoseBus: apt will install whatever is located in the package on launchpad | 20:38 |
| gst568923 | sarnold going back to the previous twinkle command, can I replace "sh" with "bash" is it the same? | 20:38 |
| sarnold | RoseBus: packages in ubuntu usually come in through debian. debian may have modified the package. ubuntu may further modify the package. | 20:38 |
| sarnold | gst568923: yes, you can; that might run a little slower, since dash is faster to launch than bash, but then you'll get access to bash features in the command line, too | 20:40 |
| gst568923 | sarnold so `....-x bash -c ....` OK? | 20:42 |
| sarnold | gst568923: exactly :D | 20:42 |
| gst568923 | sarnold one last question I wanted to give "-20" as the renice value, is this the maximum I can give to a process to run it in real time?, in my case I need real time because twinkle takes care of calls with the SIP protocol | 20:45 |
| sarnold | gst568923: there are also realtime scheduling classes available | 20:46 |
| gst568923 | sarnold that is? I am poorly documented about it | 20:47 |
| maco | does anyone remember the command for querying "what are all the packages that depend on $foo?" when i run `apt-get dist-upgrade` it wants to add apache to my system. | 20:48 |
| sarnold | maco: apt-rdepends -r, I think is what you're looking for | 20:49 |
| sarnold | with apache it's a long output.. | 20:49 |
| maco | thanks, i'll pipe it to less | 20:50 |
| sarnold | gst568923: there's a handy rtkit-daemon that's supposed to make this easy but I don't know how to use it :( | 20:50 |
| sarnold | gst568923: realtime scheduling classes get to monopolize the CPU until they explicitly drop the cpu; the default kernel config gives other processes a tiny amount of time each second to run.. so the application may not actually work great in realtime scheduling classes | 20:51 |
| gst568923 | sarnold I chose the renice command because it happened to me that during multitasking that is starting other processes in addition to twinkle, the voice during a SIP session arrived with a lot of delay | 20:54 |
| sarnold | gst568923: alright, forget rtkit :( check out schedtool package -- the schedtool program can do this | 20:54 |
| frad | I use sleep (command) as a timer, but I cannot see minutes and seconds counting down. How do I do that? | 20:57 |
| tomreyn | sarnold: doesn't apt-rdepends with '-r' give you reverse depends, so all packages that depends on "apache", rather than all packages "apache" depends on? | 20:58 |
| tomreyn | maco: ^ | 20:59 |
| sarnold | tomreyn: I interpreted the question as "how do I find out what depends upon apache now?" | 20:59 |
| tomreyn | sarnold: hmm, yes, you're right, i think i read it wrong. | 20:59 |
| tomreyn | sorry! | 21:00 |
| sarnold | frad: afaik there's no easy tool that does that already in ubuntu | 21:00 |
| sarnold | I mean, it might be there, but apt-cache search countdown timer didn't give me a lot of good results | 21:00 |
| frad | sarnold, an app with a gui maybe? | 21:01 |
| sarnold | I'd hope gnome or kde would have *some* kind of simple countdown timer, it feels like an obvious thing, doesn't it? :) | 21:02 |
| sarnold | they might not really advertise it as a feature in a package description though :/ | 21:02 |
| tomreyn | i bet there is some gnome extension which does this (or pretends to, and works for exactly one gnome-shell version, while for all others, it breaks your desktop) | 21:03 |
| maco | @tomreyn yep reverse depends were what i wanted. i'm trying to figure out what i have installed that wants to pull in apache2 on upgrade. though... i'm not sure apt-rdepends has the uhh foresight to tell what's *going* to depend on it after upgrading... | 21:05 |
| frad | geez | 21:05 |
| tomreyn | frad: maybe https://extensions.gnome.org/extension/604/teatime/ | 21:07 |
| tomreyn | maco: it just lists *everything* that could trigger the given package to be installed. it's not specific to what is (not) installed on your system (or what would be installed after a release upgrade). | 21:09 |
| tomreyn | ktimer/bionic 4:17.12.3-0ubuntu1 amd64 | 21:09 |
| tomreyn | countdown timer | 21:09 |
| tomreyn | also kteatime/bionic 4:17.12.3-0ubuntu1 amd64 | 21:10 |
| frad | ill stay with sleep, I only need to see seconds counting down on a cli, not a new app with huge digits | 21:11 |
| TJ- | maco: try "apt-get -o Debug::pkgDepCache::Marker=true dist-upgrade" | 21:12 |
| maco | @tomreyn yeah my plan was to cross reference `dpkg -l | grep ^i` | 21:13 |
| maco | @TJ *magic* and magic i didn't know even when i was spending my weekends on ubuntu | 21:13 |
| TJ- | maco: for other Debug:: options see "man 5 apt.conf" and the DEBUG OPTIONS section | 21:14 |
| maco | thanks! | 21:14 |
| leftyfb | frad: while true; do date +%S ; sleep 1; clear;done | 21:15 |
| maco | i'll give a vote for kteatime | 21:15 |
| frad | what exaclt will that do leftyfb ? | 21:16 |
| leftyfb | frad: that will show you seconds from your clock. If you want a countdown, then you can use this as a function: for i in `seq 0 $1`; do echo $i;sleep 1 clear;done | 21:16 |
| tomreyn | https://linuxconfig.org/time-countdown-bash-script-example looks useful (though you might want to reformat that output somewhat) | 21:16 |
| tomreyn | TJ-: i stumbled on this before, but then didn'T know what the output means, are you aware of any documentation? | 21:17 |
| tomreyn | https://termbin.com/1oeu is returned by this on my 18.04 system: sudo apt-get -o Debug::pkgDepCache::Marker=true dist-upgrade --assume-no 2>&1 | nc termbin.com 9999 | 21:19 |
| tomreyn | and i wouldn't know what "@ii mK NPb IPb > FU=0" means exactly | 21:19 |
| leftyfb | frad: https://pastebin.ubuntu.com/p/KNBRDJfRhh/ | 21:19 |
| leftyfb | frad: called with "countdown 5" to count down from 5 to 1 | 21:20 |
| frad | leftyfb, seconds? | 21:21 |
| leftyfb | frad: yes | 21:21 |
| maco | I've been gone almost a decade. Nowadays, does "dist-upgrade" actually move you to the next version, or is it still the same as "full-upgrade" was? And do "recommends" get automatically installed like "depends" or are they left out like "suggests"? | 21:21 |
| leftyfb | give or take some milliseconds :) | 21:21 |
| frad | so if I need to countdown 10 minutes that would be 600? | 21:21 |
| leftyfb | frad: yep | 21:21 |
| leftyfb | frad: it will count from 600 down to 1, with a "sleep 1" between each one | 21:22 |
| frad | ok, im an absolute noob. How do I name the script, where do I save it and how do I execute it? | 21:22 |
| TJ- | tomreyn: docs are pretty much internal - it's debug info - but the indentation shows the dependency graph for each package, so if e.g. apache2 is indented you move to left column to find ultimate 'keeper' package that pulls it in | 21:23 |
| tomreyn | maco: to my knowledge, apt(-get) "dist-upgrade" and "full-upgrade" do exactly the same. and, on a default installation, recommended packages are installed by default (treated as Depends:), but you can change that. | 21:23 |
| leftyfb | frad: call it whatever you want, personally I would use this as a function, not a script. I have my functions broken out in ~/.bash_aliases.d/<purpose> but you could just copy it to the end of your ~/.bashrc | 21:23 |
| tomreyn | TJ-: thanks. there's only a single level of indentation on the output ii posted, so maybe that'S a bad example. | 21:24 |
| leftyfb | frad: add in an app called toilet and you can write the numbers fairly large on your terminal using ASCII. I use it for doing things like resetting hardware using the 30/30/30 routine | 21:24 |
| maco | leftyfb what lovely name for an app :P | 21:25 |
| leftyfb | maco: I didn't name it. But it's handy | 21:25 |
| frad | leftyfb, I aved it to my home directory under 'countdown.bashrc' am I doing this right? | 21:26 |
| leftyfb | frad: negative. Add it to the end of your already existing ~/.bashrc file | 21:27 |
| leftyfb | ooo, TIL osd_cat | 21:28 |
| frad | done.ok leftyfb , done | 21:29 |
| leftyfb | frad: now type: source ~/.bashrc | 21:29 |
| leftyfb | frad: now type: countdown 5 | 21:29 |
| frad | leftyfb, thanks a lot!! | 21:30 |
| frad | im using this now instead of sleep | 21:30 |
| maco | tomreyn: thanks. i remember when recommends started being treated as depends, but that's the sort of thing that could've changed again | 21:31 |
| frad | but wait, I still need an alarm. Can I edit your script to play a tune after it's counted down? I tried this, but it doesn't work: https://pastebin.ubuntu.com/p/PyjrwBjHQh/ | 21:34 |
| leftyfb | frad: put it on a new line under "done" | 21:37 |
| maco | frad: you could make it beep https://unix.stackexchange.com/questions/1974/how-do-i-make-my-pc-speaker-beep | 21:37 |
| TJ- | tomreyn: maco: explanation of the symbols in the MarkInstall report https://git.launchpad.net/ubuntu/+source/apt/tree/apt-pkg/prettyprinters.cc#n19 | 21:37 |
| leftyfb | maco: that doesn't work on a lot of machines these days | 21:37 |
| tomreyn | TJ-: thanks for looking this up for us. | 21:37 |
| frad | leftyfb, logic, I get it, plain chronological order. Thanks leftyfb !! | 21:38 |
| frad | I combined it and it started playing that song each second | 21:39 |
| maco | TJ: thanks. i figured out which package it is. it's coming from ondrej sury's php-releases ppa. now i'm looking at the src package trying to figure out if libapache-php is marked as a dep or a rec on php8. | 21:39 |
| TJ- | maco: fred: use paplay (Pulseaudio Play) e.g. " /usr/bin/paplay /usr/share/sounds/freedesktop/stereo/message.oga " | 21:40 |
| maco | frad can you paste what you have into https://paste.ubuntu.com/ ? | 21:40 |
| frad | leftyfb, any idea how much cpu this consumes? | 21:40 |
| leftyfb | frad: nope, but probably very little | 21:40 |
| maco | playing every second sounds like it's inside the loop instea d of outside | 21:40 |
| frad | lol right, I created a loop! | 21:40 |
| maco | also this channel is WAY lower traffic than i remember, holy moly | 21:40 |
| frad | https://pastebin.ubuntu.com/p/XyCjBTCJR8/ | 21:41 |
| sarnold | maco: hah, yeah, I remember around 2010 or so it was *insane*; these days there might even be an hour or two when it's just joins and parts | 21:41 |
| maco | yeah i used to spend hours in here answering questions 2008-2011, and i'd be helping like 4 people at a time | 21:42 |
| maco | and so would a couple dozen others | 21:44 |
| maco | couldn't really function in those days without nick highlighting cuz traffic was so high | 21:44 |
| EriC^^ | they set the chaanel to +r, killed it, lazy ops | 21:44 |
| maco | interesting | 21:45 |
| EriC^^ | yeah back in 2016 they set +r, we even plotted a graph of time vs lines spoken in channel to clearly show the +r effect, they removed it for a month or something then put it back when spammers would join *shrug* better noone talk than having to actually be a channel operator | 21:47 |
| maco | so where do newcomers get help these days? | 21:47 |
| zebrah | maco, tiktok | 21:47 |
| maco | not that i've used ubuntu with a GUI since 2013, so for all i know it's to the point where point & click users don't need much help | 21:48 |
| maco | zebrah: i can't tell if you're joking | 21:48 |
| maco | if jdong said that, i'd take it as a joke | 21:48 |
| maco | ...pricechild too | 21:48 |
| EriC^^ | im guessing most join irc, get thrown into #ubuntu-unregged then leave, the more tech savy who know irc etc will probably register and then join which are just a few | 21:48 |
| PeGaSuS | hello guys. any idea why my external drive is mounted as read-only? I can copy files from the external drive to my internal drive but not the reverse | 21:49 |
| sarnold | PeGaSuS: check dmesg, there may be a message in there | 21:49 |
| maxFlexGuest | What 64-bit processor architectures do you support? Normally, I wouldn't be talking about AERobot on this channel, but in this case, I need information to find the right processor architecture for the robot. | 21:49 |
| EriC^^ | PeGaSuS: "dmesg" might show what's going on | 21:49 |
| tomreyn | maco: https://ubuntu.com/support/community-support | 21:49 |
| sarnold | maxFlexGuest: x86-64, ppc64el, aarch64, risc-v64 | 21:50 |
| snake | does anyone know if the ubuntu lts install medium comes with memtest option? | 21:52 |
| maco | tomreyn: a discourse site? neat. good to see current timestamps on the old ubuntu forums too. my local loco died out around the time dtchen's work sent him overseas and i quit motu. are locos still a thing in general? | 21:53 |
| sarnold | snake: if you boot into legacy / csm mode, it should be available; iirc the free one doesn't work with uefi, so when it boots uefi, no mem test | 21:53 |
| tomreyn | snake: i think it does for (legacy) bios booting | 21:53 |
| snake | so how do i run memtest on a uefi system | 21:53 |
| snake | imma just try it and see what happens | 21:54 |
| maco | PeGaSuS: this might also help https://askubuntu.com/questions/175739/how-do-i-remount-a-filesystem-as-read-write | 21:55 |
| tomreyn | maco: i'm not sure, my impression is that much of the sponsored community got dried out, and only the non-sponsored parts have remained. but i may be wrong, i'm not really into it. and this is not the best place to discuss this - let's move to #ubuntu-discuss ? | 21:55 |
| maco | sarnold: i think i might kinda know why i'm not seeing the sudo update. the security repos are in my sources.list, but `apt-get update` doesn't show them being queried | 21:57 |
| sarnold | maco: that's *strange* | 21:58 |
| maco | sarnold: that's the pg-13 way to say it | 21:59 |
| maco | sarnold: wait no im wrong. they are being queried. they got hit *first*. i expected them at the end like they are in the sources.list | 21:59 |
| frad | a x265 encoded file doesn't consume more cpu than a x264 encoded one, right? | 22:00 |
| maco | sarnold: (oh and there we go. got the update now. yay) | 22:01 |
| PeGaSuS | I should have mentioned that I have two OSes installed in that same drive. Ubuntu and Linux Mint. this is all I got when I attach the drive: https://paste.ubuntu.com/p/RXvJ9Rd5hx/ | 22:01 |
| PeGaSuS | complete dmesg output after booting and attaching the drive: https://paste.ubuntu.com/p/jBC3pQRbGz/ | 22:12 |
| sarnold | PeGaSuS: rebooting clears the dmesg buffer, so any errors in the filesystem or block storage that might have been saved are wiped.. | 22:13 |
| sarnold | PeGaSuS: try journalctl -b-1 and see if you have any logs from the previous boot or not | 22:14 |
| PeGaSuS | output: https://paste.ubuntu.com/p/VNQ3RNkhDf/ | 22:15 |
| sarnold | PeGaSuS: sweet! line 2466 and onwards | 22:18 |
| PeGaSuS | that's weird. where's how I have things: Windows + Ubuntu in the internal drive (eMMC, 64GB). then on the Toshiba external drive I have: 642GiB free on NTFS + Ubuntu on /dev/sda2 + Linux Mint on /dev/sda3 | 22:21 |
| PeGaSuS | brb | 22:23 |
| PeGaSuS | sarnold: booting into Windows and shutdown with the shift key pressed, solved the issue. | 22:42 |
| PeGaSuS | but what causes that? | 22:42 |
| sarnold | PeGaSuS: windows 'shutdown' is more like a suspend these days, which can leave the filesystem in an inconsistent space | 22:43 |
| PeGaSuS | so, to avoid future issues, I should use the "shift key" trick when shutting down windows | 22:44 |
| pytorch | is there a way to share screens between 10 people? not webcams but actual computer screens? Any software that allows to do that? | 22:45 |
| sarnold | PeGaSuS: right | 22:45 |
| sarnold | pytorch: can you describe a bit more what you're trying to do? | 22:45 |
| PeGaSuS | tbh, I just keep Windows around for gaming. would ejecting the external drive and then shutting down windows the normal way spare me from this issue in a future? | 22:47 |
| sarnold | probably | 22:47 |
| leftyfb | PeGaSuS: https://www.computerhope.com/issues/ch001762.htm | 22:48 |
| leftyfb | pytorch: zoom will allow you to share your screen with multiple participants | 22:48 |
| leftyfb | pytorch: as will google meet and microsoft teams | 22:48 |
| leftyfb | pytorch: if you don't care about who you're sharing to, you can use twitch.tv or youtube | 22:49 |
| PeGaSuS | also Jitsi meet allow such. and thanks for the link, leftyfb | 22:50 |
| QuentinM | Dear Ubuntu Community. I am trying to understand whether Ubuntu 14.04 ESM will be patched for CVE-2021-3156. I do not see anything related to that online at the moment. Cheers. | 22:55 |
| tomreyn | maxFlexGuest: Are you referring to https://affordableeducationrobot.github.io/v1.0/hardware.html ? This hardware won't be running Ubuntu, or Linux, or any general purpose OS at all, I'm afraid, it's too low spec. | 22:56 |
| tomreyn | you have a microcontroller there, no proper cpu. | 22:56 |
| sarnold | QuentinM: I expect it will be patched, yes | 22:57 |
| QuentinM | @sarnold Thank you! | 22:57 |
| PeGaSuS | leftyfb: you've saved my life of a torment with that link. it works flawlessly. thanks a lot! | 22:57 |
| sarnold | nice :D | 22:58 |
| tomreyn | QuentinM: in case you haven't found this page, yet, it would track its status: https://ubuntu.com/security/cve-2021-3156 | 22:59 |
| PeGaSuS | I just wish that Linux (and Ubuntu in this case) would have a better gaming environment and that game producers actually did games for Linux | 22:59 |
| QuentinM | @tomreyn Thank you very much, we've been monitoring the page. We simply were not sure whether that was actually being considered or it "Needs triage" was just an automated placeholder. Given we did not see any note regarding 14.04 / ESM in the Mailing list or USN, it made us doubt slightly. We'll keep monitoring the page. | 23:01 |
| tomreyn | QuentinM: alright. :) You can also ask such questions in #ubuntu-hardened, or, if you need a definitive answer, via commercial support (IRC is mostly community support). | 23:08 |
| jayjo | I'm having trouble with an autoinstall and cloud-init using the new autoinstallation config format | 23:12 |
| tomreyn | try #ubuntu-server | 23:12 |
| jayjo | ok, thanks | 23:12 |
| octav1a | I suddenly can't seem to paste any files using nemo or nautilus. Permissions are fine, I can click and drag files to a new folder, or use the terminal to mv or cp, but if I copy some items (either with ctrl+c or right click+copy) I can't paste (paste greyed out in right click menu) in any folder. | 23:21 |
| maxFlexGuest63 | What processor architectures do you support on Ubuntu for robots? | 23:45 |
| sarnold | maxFlexGuest63: are you talking about ROS specifically or just in general or what? | 23:46 |
| maxFlexGuest63 | Are ROS robots? | 23:47 |
| sarnold | yes https://www.ros.org/about-ros/ | 23:48 |
| sarnold | https://ubuntu.com/robotics/what-is-ros | 23:49 |
| maxFlexGuest63 | I am talking about all processors supporting Ubuntu 22.04 LTS in general. | 23:50 |
| sarnold | that'll be amd64, ppc64el, riscv64, s390x, aarch64, and maybe armhf. (I've been surprised how long we've managed to keep supporting armhf.) | 23:51 |
| frad | what i | 23:54 |
| frad | what is the future of intel? | 23:54 |
| frad | i always bought them, thinking i was getting quality | 23:54 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!