[00:07] Wifi still not fixed on ubuntu for hp laptops :( [00:12] ok this is kinda noobish, but gotta ask [00:13] when i connect an external HD to a ubuntu system, i can usually just click the disk and it gets mounted into /media/, right? I think it's usually even writable. Now i have a monitor-less machine and want to do something similar (i'm connected over ssh). [00:14] If i just connect the drive, what could be the mount command equivalent to double clicking it from the desktop? [00:14] sybariten: having just learned this half an hour ago perhaps I'm too optimistic :) but I'm pretty sure the udisksctl mount -b /dev/whatever will be very similar to that [00:15] sybariten: dmesg | tail ; sudo mount /dev/ /mnt [00:15] lol, this is a funny coincidence [00:16] sybariten, just see the scrollback above 🙂 [00:16] sarnold: so thats a command, udisksctl? [00:16] yes [00:16] I guess that'd be closer as it'll mount as the user. [00:16] note that it requires udisks2 installed ... which you typically only find on ubuntu desktop installs [00:16] ogra: i believe this is a desktop install [00:16] i.e. wont work on servers ... [00:17] the actualy thing your filemanager does is a d-bus call for the mount/umount commands, the udiskctl command just mimics that [00:20] sybariten: For completeness, sudo mount -o uid=foo,gid=foo works too. === Erisa10 is now known as Erisa [00:26] ogra, sarnold : hm that is funny, that is basically the same question... ogra, mind explaining the d-bus call thing again? [00:28] sybariten, well, there is udisksd that runs as root ... it listens on dbus for events and can mount/unmout disks ... [00:29] per policy it does that under /media as the calling user if possible [00:30] any client app ... i.e. a filemanager or some other tool can send a dbus message on your behalf to that daemon ... [00:31] you can hand over some other info to that dbus message ... if you call "udisksctl mount" without any args it will show you which [00:32] if you want to deep dive: http://storaged.org/doc/udisks2-api/latest/ [00:33] uh wait a second, i attached it now, it automounted.... my question was actually unnecessary cause i was under the impression you needed to double click the drive with the mouse, in the desktop environment, to get a mount... [00:33] hah, no 🙂 [00:34] ogra: thanks === JackFrost is now known as Unit193 [01:35] how do I repair perl on my 20.04.4 install [01:35] it depends upon how you broke it :) [01:35] dunno tried to run inxi and it just sits there till I terminate the script [01:35] inxi is a perl executable [01:36] a one megabyte perl script omg [01:36] lol it works fine normally [01:37] just on this system it doesn't and I'm guessing something wrong with perl cause when I run perl it just sits there till I exit the process [01:37] tried apt-get repair but I guess that is not an option ;( [01:38] what's apt-get repair? [01:38] I made it up [01:38] I can't find it in the manpage [01:38] oh [01:38] lolk [01:39] tried gdb inxi but it said it is a script and not binary [01:39] there's an apt-get install --reinstall option that might come in handy if you know which package needs the care and attention, but that's just blindly throwing things around hoping you eventually stumble on the right answer.. [01:40] agreed [01:40] well perl --help works so perl not broken [01:42] maybe strace inxi ? [01:44] hmm maybe strace -f instead -- and strace is kinda brutal if you're not used to it, maybe start with strace ls or something to get a feeling for what it does first [01:45] illegal seek? [01:45] it was working when I used it last on this notebook [01:46] ESPIPE fd is associated with a pipe, socket, or FIFO. [01:46] yah strace ls works great no issues [01:46] yep that is what it said ESPIPE [01:47] -1 ESPIPE (Illegal seek) [01:47] so, find the filedescriptor that it tried to seek on, read backwards to find the call that created the filedescriptor, and see if you can spot what's broken there [01:47] donofrio: inxi has a debug mode - might see what it reports. See the man page: ' inxi -F --debug 10 ' . For an instance. [01:48] oh that might be way easier [01:48] Bashing-om, I tried "inxi -Fxzwrc0" like I always do and it just sits there [01:49] but I'll try the --debug [01:49] nope it just sits there [01:49] donofrio: Might also try with only single switches - inxi -s - [01:50] nope but good try [01:50] how do I test espipe without the script? [01:55] something diffrent about running this session from powershell and not just ubuntu.exe that seems to be causing this - https://dpaste.org/Ybta [01:57] if you re-run with strace -f you can probably catch the child process executing something [01:57] it's *probably* executing /bin/xset [01:57] but that's a guess [01:58] bingo [01:58] execve("/usr/bin/xset", ["xset", "q"], 0x56434e07ab68 /* 36 vars */ === hendry1 is now known as hendry [02:01] upgraded to 21.10 and rkhunter is giving me a ton of warnings. Any fix or is this normal? [02:01] Maximalist, 21.10 is not normally talked about here #ubuntu-next if I'm correct [02:02] thank you [02:02] anytime [02:03] Maximalist: I don't think rkhunter is supposed to be useful [02:03] donofrio lives in the past :) 21.10 is supported here [02:04] Maximalist: once upon a time it showed every single ubuntu system as infected because it was just kinda rubbish. dunno if it's improved since then.. [02:04] ravage, yah busted that was routing logic from freenode I've not even checked if -next is even on libera lol [02:04] solved, just needed to run '/usr/local/bin/rkhunter --propupd' [02:05] all rkhunter ever showed me were annying false positives [02:05] I mean I was able to fix a few processes that were sucking up too much memory because of it [02:05] but idk [02:06] smem or top or htop or ps or whatever might have done the same :) [02:06] do you suggest an alternative to rkhunter? [02:06] I like topas ;) [02:07] no topas or mtop ;( [02:07] Maximalist: afaik there's no useful tool for this; having a decent set of auditd logs sent to a remote system is probably far more useful === M4he is now known as mahe [02:22] sarnold: this might be what youre talking about [02:22] Command: brave [02:22] [22:06:28] UID: 10662 PID: 4932 [02:22] [22:06:28] Pathname: 22600 [02:23] Possible Rootkit: Spam tool component [02:23] Thats what it returns [02:23] heh, did it flag a web browser as "spam tool"? [02:24] lmao literally === autiboy[m] is now known as autiboy === autiboy is now known as autiboy2005 === autiboy2005 is now known as autiboy [03:09] Hi! I'm trying to write a udev rule to rename/symlink a specific device. The device presents as both an audiocodec and a HID input... I want to target just the HID. Is there something more specific than just vendor/product I can do? dmesg says this about the device: hid-generic 0003:0D8C:013A.0004: input,hidraw3: USB HID v1.00 Device [....] on usb-0000:00:14.0-5/input3 [03:10] I'm not familiar with what the ".0004" is or the ":14.0-5" bit is - I'm hoping one of them is something a rule can target [03:13] sarnold: One of the things about rkhunter is that it has some known positives that are really easy to identify and instructions on how to tune it to avoid those, along with explanations as to why they happen. [03:14] astronut: you could try; sudo lshw -C sound, or lsusb or inxi -F to show more details of your hardware [03:20] more googling turned up udevadm info [03:21] in this case i think SUBSYSTEM=="hidraw" will be enough to seperate the two [03:22] lotuspsychje: thanks [03:28] keep getting this from seahorse when i try to import a keypair to gpg [03:28] (seahorse:487160): Gtk-WARNING **: 23:26:11.297: gtk_widget_size_allocate(): attempt to allocate widget with width -15 and height 16 === jrenken_ is now known as jrenken [05:38] Hi everyone. Is there any library/package that will allow me to convert a package name to its CPE? [05:41] Teachmehow: "CPE" would stand for what in this context? [05:46] tomreyn: Common Platform Enumeration (CPE). Each published vulnerability is associated with a list of identifiers. I can match the CPE identifier of the software installed on my PC to the CPE published in the National Vulnerability Database to know if my system is affected. What I have now is the name of the package but I want to be able to get the CPE from those package names. [05:53] Teachmehow: you'll probably get a more accurate response on that in #ubuntu-security [05:55] tomreyn: Thank you. [05:58] Teachmehow: are you aware of USN, though? https://ubuntu.com/security/notices They enable you to match information on security vulnerabilities to Ubuntu packages and versions as well as CVE IDs. [05:58] but i'm not sure whether that'd match your needs. [06:12] Hello! I'm using Ubuntu Ubuntu 21.10. I have `x-udisks-auth` in /etc/fstab for one of my partitions but whenever I click on the partition name (on the left hand side under "Devices") in file manager Thunar it simply gets mounted without asking password. Why is that? I expected a prompt asking password. [06:12] Here is /etc/fstab: http://paste.debian.net/plain/1234786 [06:16] Same with nautilus. Why doesn't `x-udisks-auth` produce a password prompt? [06:32] rootkea: you probably chose to have the password stored in gnomes' password store, editable using the seahorse application [06:35] rootkea: ignore my response, i misunderstood what x-udisks-auth does [06:38] hi all, anyone having suggetions for cloud storage or automatic backup of data? [06:39] ...Borg+rsync.net? [06:39] if you want opinions about software try #ubuntu-discuss === beaver_ is now known as pong [07:19] when using a network adapter, is it normal for ubuntu to say I am connected to my wifi from two different network devices? It says my internet network card is connected as well as my external card. Could this pose any security/privacy issue? [07:21] where / how do you see that "ubuntu [says that you are] connected to [your] wifi from two different network devices"? [07:22] On the drop down on the top right of the screen. It says they are both connected. [07:22] I am able to connect to the same network twice [07:25] can i disable the internal network card to make sure its only going through the external adapter? === jclsn5 is now known as jclsn [07:34] solved === yaeiou1 is now known as yaeiou === rootkea_ is now known as rootkea [10:03] 守望天空 [10:10] Heya. I have a question about a missing package: libengine-tpm2-tss-openssl. It was in Jammy a few weeks ago, but has now disappeared. Where can I find more information about this? I was very happy to see it in jammy which meant that I could stop installing it from bullseye-backports... === desnyt628 is now known as bonjour [11:47] noent, indeed, removed? https://launchpad.net/ubuntu/+source/tpm2-tss-engine [11:47] https://bugs.launchpad.net/ubuntu/+source/tpm2-tss-engine/+bugs?field.status:list=NEW [12:04] oerheks: yep. That bug references the package which existed a few weeks ago [12:05] !next | noent [12:05] noent: Jammy Jellyfish is the codename for Ubuntu 22.04. For technical support, see #ubuntu-next. For testing and QA feedback and help, see #ubuntu-quality. [12:05] ty [12:06] Why Jammy Jellyfish when it's a wombat on the official wallpaper ? [12:07] NeoFAT32: really? it's a wombat? lol [12:08] https://www.how2shout.com/linux/wp-content/uploads/2021/11/Ubuntu-22.04-LTS-Jammy-Jellyfish-Gnome.png [12:08] O.O [12:08] not a biologist, but pretty sure that's not a jellyfish :) [12:12] NeoFAT32, that is not the original, dude [12:12] see the date .. fail [12:13] oerheks: Oh ok ! [12:15] Can the next be a... monkey ? I like monkey ! But I want a monkey which looks like Goku in SS 4. [12:16] no. [12:16] Release on 21/04 [12:16] LTS. [12:18] oerheks: Please, please, please ! Why did they use such a "Mangekyo Sharingan" as logo ? :/ [12:18] NeoFAT32, please troll in some other channel, thanks. [12:18] !ot [12:18] #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! [12:19] I'm not trolling. [12:19] But bye ! [12:38] Hello, is it just me or is ubuntu.com down? [12:39] its just you [12:39] i get a "504 Gateway Time-out" [12:39] (on the wiki) [12:40] it is up, hit F5 ? [12:40] Doesn't help. [12:40] the wiki seems kind of down here too [12:41] https://status.canonical.com/ [12:41] yes, wiki down/maintenance .. [12:42] ok. so lets wait until some sysadmin restores a VM backup :) [12:42] Got to bookmark this page ^^' thanks. [12:42] no, the cleaning lady is hoovering the carpet, ..? [12:43] and it is up again, thanks! [12:45] the history show some wiki downtime lately :D [12:46] RiFo, another bookmark; https://status.snapcraft.io/ [12:47] I am trying to upgrade a old ubuntu lts, and it shows "no new release found" on do-release-upgrade . Any idea what the reason could be? (Version is Xenial - 16.04 lts) [12:48] !eol | RiFo [12:48] RiFo: End-Of-Life is when security updates and support for an Ubuntu release stop. Make sure to update Ubuntu before it goes EOL so you get updates promptly for newly-discovered security vulnerabilities. See https://help.ubuntu.com/community/EOL and https://wiki.ubuntu.com/Releases for more info. Looking to upgrade from an EOL release? See https://help.ubuntu.com/community/EOLUpgrades [12:48] the old-releases trick might work, i would do a fresh install [12:51] can anyone tell me the correct way to add a dns search suffix in 20.04 ? [12:51] also, if you know, you might answer this thread on the subject: https://askubuntu.com/questions/1304452/adding-a-domain-suffix-to-the-dns-search-in-ubuntu-20-04 === bonjour is now known as desnyt628 [12:51] ill try, thanks [12:53] Windy, dns is usually set in your router, but you can add it to your networkmanager profile. [12:53] this is an ec2 instance, not a lot of access to the DHCP configuration [12:54] "nmcli: command not found" ? [12:54] ec2, maybe in your control panel? [12:55] in every other linux distribution i've used, i could add a search suffix in the resolver config of resolv.conf [12:55] there's got to be a way to do it from the system itself... [12:55] nmcli is standard in ubuntu, not sure about ex2 [12:57] https://ubuntu.com/server/docs/network-configuration this suggests i need to make a change to my netplan configuration file, not sure where that is [12:57] Windy: i bet that you can both set a search domain on Ubuntu 20.04 aws cloud images and configure AWS to pass that to your EC2 instances (via a Route53 hosted private zone, or resolver forwarding) [12:57] ... via dhcp [12:59] Windy: usually, system service configurations are stored in /etc, in either a subdirectory or file named just like the service [12:59] so is there /etc/netplan* ? [12:59] gotcha. yeah, looks like you can set an override in /etc/netplan https://aws.amazon.com/premiumsupport/knowledge-center/ec2-static-dns-ubuntu-debian/ [13:00] fo aws support, try #aws [13:00] netplan.io/examples is quite handy too [13:00] ##aws rather [13:00] nmcli is provided by network-manager. network-manager is not installed by default on ec2 Focal instances. [13:00] oh oke, thanks jchittum [13:01] tomreyn and ahasenack are correct -- netplan.io is the default network management tool on Ubuntu ec2 instances === lotuspsychje_ is now known as lotuspsychje [13:02] thanks all. writing a netplan config. i'll probably end up injecting it from userdata. [13:02] at least until we get around to configuring ansible for post-provisioning [13:03] isn't there a cloud-init module specific for this? [13:03] then you wouldn't have to manually mess with config files and netplan commands [13:07] not even that is needed since resolver configurations are passed into EC2 instances via dhcp. and there are "dhcp option sets" in AWS. but that's rather a topic for ##aws [13:07] hmm, yeah, probably. the primary netplan does mention it's configured from cloud-init\ [13:07] that's probably cleaner than running a bash script. [13:07] you can override it still [13:08] use just the nameserver: stanza, followed by search:, even when using dhcp [13:10] Hi, we are trying to disable 20.04 Desktop from having the screen blanking on the login screen do to inactivity. Our searching has come up with a few ideas, but nothing seems to stick, and the screen goes blank. These are being used in a lab so we want users to know the computers are on and able to walk right up and use them. Thanks in advance! [13:18] maybe consoleblank=0 ? https://askubuntu.com/questions/1323258/screen-timeout-for-ubuntu-server-login-screen [13:19] dconf ; org.gnome.desktop.session to 0 only works after login, AFAIK [13:37] thanks. I am trying to set dconf values but not seeing them apply. From what I can tell they should be stored in /etc/dconf/db/local.d and I should run dconf update to make them apply [13:39] misilot, that dconf example is for when you have logged in, not for the login screen timeout [13:40] consoleblank=0 on your grubline, and run update-grub [13:40] oerheks: yes, but trying other settings as well [13:40] !grub [13:40] 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 [13:41] sorry, I haven't tried the consoleblank=0 since I assumed that was for one of the terminals, and not GDM [13:44] hmm, another try; edit the file /etc/login.defs and change the line; LOGIN_TIMEOUT 60 to 0 ? === withered_wolf is now known as life_alert === life_alert is now known as withered_wolf [14:45] how can I change ownership of all files and folder in a directory to "root"? [14:46] zen_coder: Why? [14:48] I want to archive some files in a tar archive [14:48] cd to folder; sudo chown root * [14:48] zen_coder: that doesn't explain why you need to change their ownership to root [14:49] and with -R it does folders too [14:49] can I tar file without ownership? [14:49] files [14:50] zen_coder: as root, yes. As other users, depends on their permissions [14:51] adm I do: `tar cvzf my_targ.tar.gz *` [14:51] so when I do `sudo tar cvzf my_targ.tar.gz *` [14:51] man tar; --no-same-owner [14:52] why do we make man pages? [14:55] `--no-same-owner` is only for extracting [14:55] but not for packaging [14:56] zen_coder: just tar it up as root, it will keep the permissions [14:58] leftyfb: now I got a protected tar archive [14:58] I want to distribute this archive [14:59] not only me should open it [14:59] zen_coder: how will you be distributing it? [14:59] I upload it to a server [14:59] and will be downloaded by other and extracted [15:00] zen_coder: it's not "protected", it's probably just currently owned by root as you wanted the files to be. [15:00] zen_coder: upload how? scp? ftp? smb? nfs? https? === cris_ is now known as cjdc [15:01] https [15:02] zen_coder: ownership of the tarball is not going to translate being uploaded via http. As long as your user has access to the file and can upload it, it'll be whatever ownership your uploaded systems sets it to be [15:02] The lurking question is what should the ownership and permissions be on the target system. [15:02] zen_coder: I still don't understand why you needed to chown the contents to root [15:03] There is the point that root is likely to be the only consistent UID across systems. [15:03] Hi all [15:04] jhutchins: given the current level of knowledge here, I don't think that's being planned out [15:09] so the normal case is that you just tar the files with whatever owner it is just doing it, then you extract it to your local system and by default the extracted files get owner of the current user extracting it? [15:11] zen_coder: creating the tarball will preserve the ownership. Extracting requires setting --same-owner but usually also needs to be done as root because your user might not have permissions to set ownership on the files as they are being extracted [15:12] zen_coder: but this is only if you need to preserve the exact same ownership as they were. If you only need the end user to have access, that will be the case by default with no additional steps [15:24] okay, following case, I tar the files as arbitrary user [15:25] then I want that root can extract them as root [15:25] how to do this? [15:25] will this work by default? [15:25] because I read: [15:25] zen_coder: yes [15:25] zen_coder: you know you could easily test this [15:25] --same-owner => Try extracting files with the same ownership as exists in the archive (default for superuser). [15:26] superuser = root, right? [15:27] zen_coder: what is the end goal here? Does the end user who is receiving your file need to have them owned by root for some reason? Does this end user have access to sudo or to login as root on their end machine? === ullbeking_ is now known as ullbeking [15:30] zen_coder: Yes, superuser = root [15:41] h [16:01] when click on .torrent file it save this not open aplication , why this ? [16:02] aguitel: did you install a torrent client that sets itself as the default application to open .torrent files? [16:04] leftyfb, yes [16:04] aguitel: which application? [16:04] qbittorrent [16:04] aguitel: also, if that were true, you wouldn't be having this issue [16:05] maybe reinstall it ? [16:06] aguitel: firefox? [16:06] aguitel: change the default action of your browser [16:06] firefox and chrome [16:06] aguitel: https://www.linuxbabe.com/desktop-linux/how-to-change-the-default-torrent-program-on-linux [16:07] cbreak: they said clicking on a .torrent file, nothing to do with a browser [16:07] well... where else would it "save" a torrent file? [16:08] in cache [16:08] aguitel: after saving it, open it up with qbittorent [16:10] leftyfb, after saving and click in .torrent file ,qbittorrent open [16:10] aguitel: you have to change the action your browser performs when encountering those files. [16:11] I think saving is the right thing, but if you don't like that, in firefox, you can change that in the settings [16:11] cbreak, HOW DO IT [16:11] That can be an issue on the server as well. [16:11] That can be an issue on the server as well. [16:11] by selecting your desired program instead of save [16:11] aguitel: open the settings [16:11] scroll down to "Applications" [16:12] search for torrent [16:12] change action [16:13] cbreak, no torrent aplication appears , only magnet [16:15] weird. [16:16] https://support.mozilla.org/en-US/kb/applications-panel-set-how-firefox-handles-files [16:16] that article says: "The Applications panel has limited functionality for editing. You can change the action for an existing file type but you cannot add or remove file types. Entries are added automatically when you download files and select actions for them. See Adding download actions. " :( [16:17] so it should ask you when you download a torrent file for the first time [16:18] magnet links are blocked here from opening WITHOUT saving too.. [16:20] saving magnet links is weird... [16:22] when I go to https://torrent.ubuntu.com/kubuntu/releases/impish/release/desktop/ and click on the torrent file, it downloads fine or opens in transmission remote fine, depending on what I configure. [16:29] leftyfb: but do I read the docu correclty, when root is extracting a tar file, it will preserver the owner rights (which in my case leads to file owned by some arbitrary user?) [16:32] that means, when I want that root is extracting the file as themself, he should use "--no-same-owner" [16:32] zen_coder: maybe. Depends of the UID of the original user exists and matches the same user on the new system [16:32] I tested it, switched to root [16:33] and only with option "--no-same-owner" the files are shown to me as root [16:45] Hello! I'm using Ubuntu Ubuntu 21.10. I have `x-udisks-auth` in /etc/fstab for one of my partitions but whenever I click on the partition name (on the left hand side under "Devices") in file manager Thunar it simply gets mounted without asking password. Why is that? I expected a prompt asking password. [16:46] Same with nautilus. Why isn't `x-udisks-auth` producing a password prompt? [16:50] by putting it in fstab, it is authorized alrady? [16:50] c/already [16:50] try to mount it from a live iso, or remove it temporary from fstab to check ? [16:52] oerheks, I used gnome-disks utility which edits /etc/fstab on it's own. [16:54] interesting, never seen that [16:55] In /usr/share/polkit-1/actions/org.freedesktop.UDisks2.policy there's `` with default value `auth_admin_keep`. Make a backup of that file. Change all 3 values there from "auth_admin" to "no"; reboot and check if the new value is respected (i.e. you can't mount it) [16:55] If that's the case, it means that you already entered a polkit authentication somewhere and it remembers it [16:56] (which is what auth_admin_keep means, vs plain auth_admin) [16:57] alkisg, will it then ask password before mounting? [16:58] rootkea: if you verify the above steps, and THEN you change auth_admin_keep to plain auth_admin, then it will always ask for a password before mounting, even if you had put your password already a while before clicking mount [16:59] alkisg, Okay. BTW after editing that file, shouldn't it be copied in /etc somewhere for it to take effect? [17:00] No, it takes effect immediately [17:00] (or at least it should; if not, just reboot; no /etc is involved) [17:01] rootkea, my 2 cents : 'Note that this authorization is only applicable to non-administrators of the PC. Admin users will be able to mount the partition without needing to enter the password. This also means that if there is only one account in your PC, which implies it’s that of the administrator, this setting does not affect.' [17:01] ioria, source for the quote? [17:02] rootkea, https://www.fosslinux.com/4216/how-to-automount-hard-disk-partitions-in-ubuntu.htm [17:02] BTW, my current user is in sudo [17:02] so ... [17:04] oh, gnome disks does this not by itself :-D [17:06] hi [17:07] hi [17:08] alkisg, rebooting now... [17:15] The more we automate basic tasks, the worse off people are when the automation fails. [17:16] alkisg, I put "no" in those three lines and rebooted the machine but still when I click on partition name in file manager it simply gets mounted without asking password. [17:16] It's as if the file in /usr/share has no effect [17:22] This AskUbuntu answer says to edit another system file: https://askubuntu.com/questions/417685/how-to-require-mount-with-password-of-other-internal-hdd-partitions-for-all-moun [17:23] But I don't want to make changes in files outside of my home directory so I'm giving up... [17:24] After selecting "Require additional authorization to mount" in gome-disks I expected it to just work. [17:25] s/gome-disks/gnome-disks === _miruoy is now known as miruoy [17:44] rootkea: Gnome knows what's best for you. [17:52] Hi all [17:52] hi [17:52] Hello, anybody knows how to modify taskbar ? Right click is not working. Or just how to let toolbar to show all programs instead only active window. Ubuntu 20.04.4 LTS [17:53] Hi all, needed suggestion for this problem: [17:53] Looking for Centralised Linux Device/Desktop Management/Fleet Management solution [17:53] Challenge is about managing about 300+ devices [17:53] Developers are running multiple Linux Flavours [17:53] Mostly/half of the users are using Ubuntu Desktop [17:53] Followed by considerable amount of Arch Linux [17:53] Debian [17:54] Karthik1: What items do you want to manage? [17:54] !enter | Karthik1 [17:54] Karthik1: Please try and keep as much of your info as possible on ONE line - easier to follow for everyone. [17:55] ok got it [17:56] Karthik1: define manage [17:56] updates, authentication, deployment, etc [17:58] mostly patch applying, remote management (power off/on), push packages remotely, encrypt the local drives, installing agents like Nessus (security/compliance) management [17:58] Reporting about users/OS version/Hardware [17:58] As of now, its unmanaged fleet, so need essential fleet management capability/bare minimum management [17:58] Karthik1: ansible. Though that" encrypt local drive" is a pretty loaded ask [17:58] yes leftyfb [17:59] you mean ansible playbooks for Linux fleet/desktop management [18:00] Karthik1: though this based on my personal preference and experience. It's kind of beyond the scope of this channel. You should try #ubuntu-offtopic for opinions on software. This is more focused on Ubuntu support questions [18:01] ok fine [19:02] anyone know if it's possible to install ubuntu (server or whatever) without a gui. I need to via serial. [19:03] Yep, ubuntu server [19:04] DivanSantana: https://github.com/ynkjm/ubuntu-serial-install [19:04] DivanSantana: maas can probably make that task pretty easy; you can probably get there via pxe booting, https://ubuntu.com/server/docs/install/autoinstall-quickstart , kind of thing [19:05] you could also install the thing somewhere else or via ipmi, and then just use the image without GUI at all [19:05] awesome. thanks. good ideas. [19:06] I'm trying on openbsd's vmd. Which only supports a single core. When it boots the iso I get Loading install... failed: No such file or directory [19:07] Simarly rhel/rocky doesn't work on vmd because it needs more then one core. Perhaps ubuntu is the same. So far I've only been able to install alpine on vmd. Guess I should check in #openbsd. [19:42] I have installed emacs 27.2 via snap, but doing `emacs --version` says it's 26.3 [19:43] why is this happening? snap list shows emacs 27.2 as well [19:43] TikityTik: type "which emacs" [19:43] cbreak, latest/beta classic alexmurray [19:43] TikityTik: type -a emacs [19:43] TikityTik: remove the apt package [19:43] 26.3 must be a deb package? [19:43] TikityTik: it should tell you a path [19:44] if you want to be sure you use the one from snap, start it via /snap/bin/emacs [19:44] I did sudo apt remove emacs too before the snap [19:44] but "which" will tell you which one you're using via unqualified lookup [19:44] leftyfb, what you mean `-a emacs`? [19:45] TikityTik: "type -a emacs" [19:46] getting /usr/bin/emacs /bin/emacs /snap/bin/emacs [19:46] so, yes, you have other emacs in your $PATH [19:46] either start it as I said above via full path [19:46] or change your $PATH, or remove the other emacses [19:47] why do I still have emacs if I did `apt remove emacs`? [19:47] This is a metapackage that will always depend on the latest [19:47] recommended Emacs variant (currently emacs-gtk). [19:47] TikityTik: apt-cache policy emacs | nc termbin.com 9999 [19:48] i mean why do i still have the old emacs if i did `apt remove emacs`, when i try typing it again it says it's not installed but it's in my /usr/bin/emacs and /bin/emacs [19:48] TikityTik: because you don't understand what the 'emacs' package actually provided :) [19:49] TikityTik: try "apt-cache show emacs" and "dpkg -S /usr/bin/emacs" and you'll probably be enlightened :) [19:49] reboot, and try again? [19:49] that would clear stuff [19:49] ... rebooting shouldn't delete files from /bin ... [19:50] sarnold, no path found for that dpkg [19:50] sigh that might be the usrmerge nonsense.. how about dpkg -S /bin/emacs ? [19:50] TikityTik, emacs is a meta package ... it only depends on the versioned packges that actually carry the binaries [19:50] cbreak, then emacs is not uninstalled, likely [19:50] TikityTik: what about "apt-cache show emacs\*"? [19:50] sudo apt purge emacs* [19:50] oerheks: I think it is indeed not uninstalled [19:51] that should get rid of everything emacs ... [19:51] cbreak, I see a package for emacs for that one [19:51] i also tried sudo apt install emacs, then sudo apt purge emacs [19:51] TikityTik: apt-cache policy emacs | nc termbin.com 9999 [19:51] chances are you don't have emacs anymore [19:51] TikityTik, put an astersik at the end [19:51] you have emacs-xxxxx or something [19:51]