[01:51] <bn_work> is it possible to manually upgrade PHP to v7 on an older Ubuntu 14.04LTS server outside the package manager?  (I'm assuming that's not advised?)
[01:52] <leftyfb> bn_work: upgrade your OS to one that is supported and you'll get a supported version of php along with it
[01:52] <bn_work> leftyfb: that's a lot of stuff to upgrade just for PHP
[01:52] <leftyfb> bn_work: you're running an outdated and unsupported OS
[01:58] <sarnold> bn_work: you could also run a newer instance in an lxd or something
[02:00] <bn_work> sarnold: thanks, that's an interesting idea.  Trying to evaluate what the least painful way to support this web server is.
[02:03] <sarnold> bn_work: there's only 13 more months of ESM support for 14.04; moving to 22.04 would give you four more years before it even enters ESM
[02:04] <leftyfb> a fresh install of 22.04 and restoring from backup is a lot easier than you think
[02:05] <bn_work> leftyfb: no, it's not
[02:07] <sarnold> dealing with apache authz/authn changes is probably a bit annoying, if you're using apache
[02:07] <leftyfb> authz/authn?
[02:08] <bn_work> sarnold: is that something that changed on Apache2 between 14.04 LTS and 22.04 LTS?
[02:08] <sarnold> leftyfb: authorization and authentication
[02:08] <leftyfb> bn_work: I've been hosting sites for over 20 years. 9 of those was were working for the largest shared hosting company in the world. It's really not that hard. Even so, running a public web server should never be a "Set and forget" deal. You should either be able to rebuild it in your sleep or have it well documented on how to rebuild it
[02:10] <sarnold> bn_work: I can't recall when exactly those huge changes went through, but it wasn't great fun for folks who were surprised by it :)
[02:10] <bn_work> sarnold:  "benefits" of updating frequently I guess
[02:11] <sarnold> bn_work: yeah.. what would you rather do, deal with small changes every two years, or huge changes every decade? :)
[02:11] <leftyfb> sarnold: IIRC, it was mainly php/mysql differences. Not apache. I ported a bunch of sites from 14.04 just a couple years ago and didn't have any issues with apache
[02:12] <sarnold> leftyfb: oh nice. maybe it was pre-trusty then
[02:12] <bn_work> sarnold:  yeah.. I guess the latter, because by then workarounds will be known, I don't want to be the guinea pig
[02:13] <sarnold> bn_work: well, that decade is almost up for 14.04 .. so .. it's about time to pay the piper.
[02:13] <leftyfb> bn_work: there's been 4 LTS releases since 14.04.
[02:13] <leftyfb> each one with 5 years of support
[02:13] <leftyfb> 10 if you pay Canonical
[02:14] <bn_work> sarnold: yeah, it will happen at some point, just trying to make sure there is no "easier" way that my boss is thinking exists.
[02:14] <leftyfb> bn_work: easy != secure nor clean nor stable
[02:15] <arraybolt3> And once an OS is entirely EOL, you have security holes, security holes equals hacked systems, and hacked systems usually equals big problems.
[02:15] <leftyfb> there's almost always an "easy way. But you end up paying for it elsewhere
[02:16] <leftyfb> bn_work: either way, you're running an unsupported release of ubuntu. Good luck
[03:24] <voidah> test
[04:25] <hassletime> how is everyone today
[06:59] <RupeLubuntoo> hello
[06:59] <RupeLubuntoo> is there anyway to make my VM OS screen larger than it is. I dont mean the VM screen but the OS within
[07:01] <RupeLubuntoo> Anyone?
[07:05] <arraybolt3> RupeLubuntoo: Can you just increase the resolution inside the VM?
[07:06] <arraybolt3> (I'm assuming you're using Ubuntu inside your VM, in which case this should be doable.)
[07:06] <RupeLubuntoo> hi array, yes Ubuntu, well Lubuntu but same same
[07:08] <supay> hi all,  i seem to have a ghost partition that still shows ubuntu 20.04 even after removing the my only ssd from my laptop and replacing it with a fresh one! :O how is that possible?
[07:10] <RupeLubuntoo> @arraybolt3 its ok I had to fiddle with the display settings, its ok no just a bit of jiggling, but done now!
[08:00] <EriC^^> supay: share 'sudo parted -ls' output
[08:01] <EriC^^> supay: what do you mean by ghost partition? i think you mean some boot menu entry?
[08:59] <rafayet> hello
[08:59] <rafayet> suggest me a best editor to develop c and cpp
[09:02] <rafayet> hello
[09:02] <rafayet> help me bro
[09:02] <rafayet> @hamburgler
[09:03] <hamburgler> What's up?
[09:08] <rafayet>  @hamburgler suggest me a code editor
[09:10] <hamburgler> still using a combination of vim and atom even though atom is not maintained, VScode maybe?
[09:11] <soundmodel> VSCodium
[09:12] <rafayet> @hamburgler how to customize vim ?
[09:15] <hamburgler> I don't
[10:08] <Sevandro> Hi
[11:25] <stofflserlt> Hi! I've got a fresh install of Ubuntu Server 22.04.1 and Froxlor. The install of froxlor worked fine - but now all I'm getting is error 500 on the default html file and 404 on the froxlor setup. What went wrong?
[12:06] <WeeBey> Hello World.
[12:08] <WeeBey> I'm looking at the "passwords and keys" app trying to find the password for my HexChat nick (this one). If it's not there, any idea where it would be stored? HexChat config?
[12:14] <WeeBey> I guess we'll never know. lol
[12:17] <WaV> A bit of googling suggests It is probably in ~/.config/hexchat/servers.conf
[12:19] <WaV> or servlist.conf
[12:20] <guiverc> WeeBey, i wrote an answere here - https://askubuntu.com/questions/1339682/how-do-i-connect-to-libera-chat-with-hexchat, but too long ago for me to remember (had no reason to change it since then)
[12:21] <WeeBey> guiverc, Ha!
[12:22] <WeeBey> It looks like it's in cleartext in the config but masked in the settings. :D
[12:22] <pyqrcode> hi
[12:25] <guiverc> WeeBey, look in servlist.conf
[12:25] <WeeBey> guiverc, I think I found it. Thank you.
[12:26] <WeeBey> If anything it seems like I could change my password cos I was already identified.
[12:26] <WeeBey> I appreciate your help. :)
[12:26] <guiverc> You're most welcome
[13:08] <supay> EriC^^: sorry for the late response, i didn't see it until now. you're exactly right. someone from #linux suggested it must be an EFI entry written to the EEPROM. thank you so much :)
[13:22] <BrianMiller> Where can I find the latest/newest repositories to add to my /etc/apt/sources.list?
[13:22] <BrianMiller> I'm running Ubuntu 18.04.5 LTS
[13:24] <ravage> the repo URLs dont change
[13:24] <ravage> so you always have the latest update for your Ubuntu release
[13:24] <BrianMiller> ravage: I don't have any in my sources.list right now
[13:25] <ravage> did you delete them?
[13:25] <BrianMiller> ravage: Yes
[13:26] <ravage> https://gist.github.com/rhuancarlos/c4d3c0cf4550db5326dca8edf1e76800
[13:28] <BrianMiller> ravage: Why do these repos have old versions of apache? Like 2.4.29. I want to install the latest stable which is 2.4.55
[13:29] <ravage> in the lifecycle of an Ubuntu release you only get security and bug fixes
[13:29] <ravage> no new versions
[13:29] <ravage> so a newer version of Ubuntu usually also has newer package versions
[13:30] <ravage> https://packages.ubuntu.com/jammy/apache2
[13:46] <EriC^^> supay: no problem, glad you solved it :)
[14:30] <raub> Trying to temporarily disable screensaver. After a while, I have resorted to kill xscreensaver. But that only makes the time until screen blanks longer instead of stop trying to blank it. Is there another process name I should be looking at?
[14:51] <tomreyn> raub: i probably don't know your answer, but since you don't seem to be running the default gnome-shell / mutter desktop (which i think is incompatible with xscreensaver), it would be good to provide more detials on what you run there.
[14:53] <ogra> yeah, we have not shipped xscreensaver in about a decade
[14:53] <ogra> (in the default install that is, i think lubuntu, mate and xubuntu still use it)
[14:54] <Sircle> what will this do `sudo mount -t tmpfs /tmp`
[14:54] <ogra> it will mount a tmpfs over your existing /tmp directory
[14:55] <Sircle> what will change/ what would be the benefit of this ogra
[14:55] <ogra> tmpfs lives in ram, not on a disk ... so it is faster but steals memory ...
[14:55] <Sircle> oh
[14:56] <ogra> the moun command itself, when run manually means you are completely hiding what is in /tmp currently ...
[14:56] <Sircle> so it will put all data on ram instead of disk?
[14:56] <ogra> so dont do that on a running system, things might start to break
[14:56] <Sircle> ok, when should I run it?
[14:57] <ogra> if you want to switch to /tmp in tmpfs, do it though fstab or through a systemd mount unit and reboot ... or if you want to manually do it, shut down all graphical services, check what is left in /tmp (also check hidden fils) and make sure to stop everything utilizing bits in /tmp first
[14:58] <jhutchins> ogra: What's the alternative to xscreensaver?  Is something built in to gnome?
[14:58] <leftyfb> Sircle: why do you think you need to run this command? What is your end goal exactly?
[14:58] <ogra> jhutchins, gdm
[14:59] <Sircle> leftyfb pasting the initial question where it started:
[14:59] <ogra> in gnome 3 onwards screen blanking and locking is done through gdm (which is the logical place for the locking part at least)
[15:00] <Sircle> leftyfb I have high OS usage (apps running, database etc) on a disk. I already added a new disk for all file i/o intesive things. Question: Is it best to put database data (postgres) on another drive, OS on separate drive OR  make raid 0 of 2 drives and keep all on both drives (seeming like one drive)?
[15:01] <Sircle> btw, I am already using 10gb of swapspace
[15:01] <Sircle> ram slots are ful
[15:01] <jhutchins> The age of my systems' configurations is starting to show.
[15:01] <Sircle> I have 4x16g= 64. MB can have 128GB.
[15:01] <leftyfb> Sircle: that might be a question better suited for #postgres or a postgres forum
[15:02] <Sircle> ok
[15:02] <raub> tomreyn, ogra: using lxqt as I could never get the screensaver in the default desktop to work, and have been told lxce is deprecated
[15:03] <Sircle> ogra I think this would be better and safer? `systemctl enable tmp.mount`
[15:03] <Sircle> ogra right?
[15:03] <raub> Since this is my desktop, I installed barebones server and then added the rest
[15:04] <leftyfb> raub: gsettings set org.gnome.desktop.session idle-delay 'uint32 0'
[15:05] <freeman_h> hi all ubuntu noob here. i am looking for some kind of tiling/grid application for my ubuntu any recommendation? I looked at gtile but it wouldnt let me open it to do the install
[15:05] <leftyfb> raub: That will disable the screen blanking. Then to re-enable it, change 0 to the number of seconds you'd like to enable the screen to blank after inactivity
[15:06] <geosmile> I've a bug in my ubuntu service, which reboots the machine as soon as it boots (or something else is wrong). How do I get into grub screen? I tried holding the Shift key - but no grub. Any other ideas?
[15:07] <ogra> Sircle, right, that should enable it aftr a reboot
[15:07] <geosmile> is there a keyboard shortcut to log into an ubuntu box in a safe mode of some sort?
[15:08] <ogra> freeman_h, there is some tiling built ito gnome ... you can try super (the windows key) and a cursor key at the same time to tile windows on screen ... beyond that, thre are a few gnome extensions too i think
[15:09] <raub> leftyfb: thanks. Interestingly enough, it is currently set to 300s while the screensave preferences control panel claims to be 180s (3min), but cycling after 300s
[15:09] <raub> (per gsettings get org.gnome.desktop.session idle-delay)
[15:09] <leftyfb> raub: where do you see this "Screensaver" setting? Since that's not default in ubuntu
[15:10] <tomreyn> freeman_h: i think gtile is a gnome-shell extension. you'd need to install it using either the gnome shell extension web repository at https://extensions.gnome.org/ or the CLI tools which are named similarily
[15:10] <raub> xscreensaver preferences
[15:10] <raub> (running that as mentioned in original question)
[15:10] <geosmile> I now have a grub command line - how do i get a shell so i can remove a service.
[15:10] <ogra> in grub ?
[15:11] <leftyfb> raub: that is a separate application that isn't installed by default and doesn't integrate into gdm/gnome
[15:11] <ogra> there are no servces in grub
[15:11] <raub> leftyfb: that makes me think there is another screensaver running, hence original question
[15:11] <tomreyn> freeman_h: be careful, though, as extension versions are only compatible to some gnome-shell versions and some extensions are of very low quality (not this one, i think)
[15:12] <leftyfb> raub: ubuntu uses integrated gsettings to adjust the screen blanking. It does not use screensaver. Again, that is a separate application that does not integrate into the gnome settings
[15:12] <leftyfb> raub: I would recommend removing the screensaver application
[15:14] <raub> leftyfb: If I remove it, screens will no longer go blank. I spent a few days trying to get that to work last year, including posting here. Going to xscreensaver reliably caused my monitors to go blank at 3min
[15:14] <Sircle> I have 4x16g= 64. MB can have 128GB.  No idea what to do? Sell my old ram, buy 4x 32g = 128g  or any other option in which I do not sell my old ram?
[15:14] <raub> I was tired of going to bed and finding out my 3 monitors stayed up all night
[15:14] <leftyfb> raub: that is simply not true
[15:14] <jhutchins> raub: Do the monitors not have power switches on them?
[15:15] <raub> jhutchins: one is part of the computer
[15:15] <freeman_h> tomreyn, what about i3?
[15:15] <tomreyn> geosmile: you can choose a recovery boot from grubs advanced menu to boot to recovery and login to the root shell there
[15:15] <tomreyn> freeman_h: what about it?
[15:16] <Sircle> ogra $ sudo systemctl enable tmp.mount
[15:16] <Sircle> Failed to enable unit: Unit file tmp.mount does not exist.
[15:16] <raub> leftyfb: I will not doubt you must be right even though I am the one sitting in front of the computer
[15:16] <geosmile> tomreyn, how do i get to the menu, I'm in the command line - "grub>"
[15:17] <tomreyn> geosmile: that's where you get after pressing escape once too much, you can try typing "normal" which may bring up the menu, or reboot and hit escape just once
[15:17] <freeman_h> ogra, not sure if you have used the dell manager before it has a really good for the windows guys. when you press the window key and s it shows a grid and you can drag an application to the grid and it will auto expand the application to fill that gride
[15:17] <freeman_h> grid*
[15:17] <freeman_h> I am after something like that
[15:20] <ogra> Sircle, systemctl status tmp.mount ?
[15:20] <ogra> it is proably already running ?
[15:22] <geosmile> tomreyn, It wont even let me type using "normal". The keyboard is acting up when i try.
[15:22] <tomreyn> freeman_h: i think gtile actually supports that. tiling assistant seems to be another gnome extension to support that.
[15:22] <ogra> freeman_h, well, then you likely want to use some extension or switch to a true tiling WM like sway (or i3wm if you are fine with using old X11 stuff)
[15:23] <tomreyn> geosmile: then just hit ctrl-alt-del to reboot, and repeat until the grub menu show up
[15:24] <Sircle> ogra this worked. Is it ok? sudo cp /usr/share/systemd/tmp.mount /etc/systemd/system/
[15:24] <Sircle> systemctl enable tmp.mount
[15:24] <ogra> no 1
[15:24] <ogra> !
[15:24] <tomreyn> freeman_h: shellTile is another for older ubuntu releaeses (it's said not to work on 22.04).
[15:24] <ogra> what did the status command above return
[15:25] <ogra> tmp.mount is a generated unit ... do not touch it by hand
[15:25] <Sircle> ogra Active: active (mounted)
[15:25] <ogra> so your /tmp is *already* in a tmpfs
[15:25] <ogra> nothing to see here, just move on
[15:25] <Sircle> ogra no, I ran tthose 2 line
[15:25] <Sircle> then it became active. I think'
[15:26] <freeman_h> ogra, i will like ot use sway but am not too familar with linux(am using a vm) the git repo does not do it justice on steps to just get it up and running
[15:26] <ogra> dunno ... better check after a fresh boot ...
[15:26] <ogra> the tmp.mount unit gts generated by an fstab generator ... th fact you had it on your system indicates it was already generated and runnng
[15:27] <ogra> freeman_h, git repo ?
[15:28] <Sircle> ogra no, I did sudo cp /usr/share/systemd/tmp.mount /etc/systemd/system/
[15:28] <Sircle> and then ran systemctl enable tmp.mount
[15:28] <Sircle> ogra did I do it correct?
[15:28] <ogra> Sircle, yes, something you should never do with generated units
[15:28] <ogra> remove it again
[15:28] <Sircle> so should never had used the cp command?
[15:28] <ogra> reboot ... then check with the status command
[15:29] <ogra> neither of them
[15:29] <ogra> neither cp nor systemctl enable
[15:29] <Sircle> oh
[15:29] <ogra> revert that change and reboot
[15:29] <Sircle> then how to enable tmpfs then?
[15:29] <ogra> then check with systemctl status tmp-mount
[15:29] <freeman_h> ogra, git repository for sway vw
[15:29] <ogra> freeman_h, this is ubuntu. you install stuff from the archive with apt, not with git
[15:30] <ogra> apt show sway ...
[15:30] <Sircle> ogra revert by sudo systemctl stop tmp.mount ?
[15:31] <ogra> Sircle, did you run systemctl start on it as well ? (that has the same effect as runint the mount command while things are still actively using /tmp ... bad idea)
[15:32] <Sircle> ogra no i didn't run start
[15:32] <Sircle> ok I am rebooting
[15:33] <leftyfb> Sircle: wouldn't it be easier to just configure postgres to write to /dev/shm?
[15:33] <Sircle> leftyfb whats shm?
[15:33] <leftyfb> Sircle: https://www.cyberciti.biz/tips/what-is-devshm-and-its-practical-usage.html
[15:34] <Sircle> ok
[15:34] <Sircle> reading after reboot
[15:35] <asaill3> Hello
[15:37] <Sircle> ogra rebooted:  sudo systemctl status tmp.mount
[15:37] <Sircle> Unit tmp.mount could not be found.
[15:38] <Sircle> ogra https://askubuntu.com/a/1232182
[15:38] <leftyfb> Sircle: https://www.cyberciti.biz/tips/what-is-devshm-and-its-practical-usage.html
[15:38] <Sircle> leftyfb yes I got it
[15:38] <ogra> Sircle, add that lne to fstab: "tmpfs /tmp tmpfs defaults,noatime,mode=1777 0 0" and the service will magically appear
[15:39] <Sircle> leftyfb yes would need lots of ram for this
[15:40] <leftyfb> Sircle: it's exactly the same
[15:40] <leftyfb> Sircle: /dev/shm is exactly the same as mounting /tmp with tmpfs
[15:40] <leftyfb> it's all tmpfs (memory)
[15:40] <Sircle> right
[15:40] <Sircle> what happens when ram gets full?
[15:41] <leftyfb> you run out of memory and bad things happen
[15:41] <leftyfb> don't run out of memory
[15:42] <leftyfb> Sircle: if you have 128GB of memory and 10GB of swap and you're hitting the limits, there's no amount of tweaking /tmp or tmpfs that is going to help you
[15:42] <Sircle> but the priority for /tmp would be ram
[15:43] <Sircle> other things will go to swap
[15:43] <Sircle> so a good deal?
[15:43] <leftyfb> Sircle: at that point, you look at what is running and put some limits on it. Or you add more resources in the form of memory or faster drives
[15:43] <Sircle> my slots are full. can i have some way to add ram  some othe rway?
[15:43] <leftyfb> Sircle: buy bigger sticks of memory
[15:44] <Sircle> dam. $
[15:44] <Sircle> the old ones will get useless
[15:44] <Sircle> ogra any official docs to "how to enable tmpsf on ubuntu"?
[15:44]  * leftyfb sigh
[15:46] <ogra> Sircle, add that lne to fstab: "tmpfs /tmp tmpfs defaults,noatime,mode=1777 0 0" and the service will magically appear ...
[15:46] <ogra> there ... thats your doc ...
[15:47] <Sircle> ogra Ya I respect that but curious why no dcos
[15:47] <tmus> Hi guys... I'm having some problems getting 802.1x to work with iwd (23.04 development edition) and there must be something i'm missing. iwd says "unsupported_certificate" but the certificate looks fine (checked with TLS debug to gte the server cert) etc. Any ideas?
[15:47] <tmus> This is PEAP btw
[15:47] <ogra> no idea, there are probably odcs on the old wiki ...
[15:49] <leftyfb> !ubuntu-next | tmus
[15:49] <lotuspsychje> !next
[15:49] <tmus> lotuspsychje, thanks :)
[15:50] <Sircle> ogra so as leftyfb said, if the ram is low, there is no point in making a tmpfs anyway?
[15:52] <leftyfb> Sircle: tmpfs = ram
[15:53] <Sircle> ya. /tmp would just get priority to be in ram (if and whenever available), else go in swap
[15:53] <leftyfb> Sircle: if you're currently hitting the limits of your memory usage, sticking more things in memory is not going to help you
[15:54] <Sircle> I se
[15:54] <Sircle> leftyfb what options do I have to increase memory except buying more rams (as I told you I gor 4 slots full with smaller sized ram)
[15:55] <leftyfb> buying more "rams"
[15:55] <Sircle> ya,:)  other than that?
[15:55] <Sircle> because I won't have slopts to put them in
[15:55] <leftyfb> Sircle: or reducing the footprint of your application by optimizing it or purging data
[15:55] <Sircle> can't do that to
[15:56] <Sircle> too*
[15:56] <Sircle> any second best thing to ram? like a 7GB/s nvme?
[15:56] <leftyfb> Sircle: buy x4 64GB or 128GB sticks of memory
[15:56] <leftyfb> Sircle: a hard drive doesn't help with memory limitations
[15:57] <Sircle> my MB supports 128 Max. I have 4x 16g already = 644G
[15:57] <leftyfb> Sircle: try posting on postgres forums to increase the performance and decrease the footprint of your postgres db
[15:58] <Sircle> brb 10
[15:58] <Sircle> its not about pg
[15:58] <Sircle>  aloneits not about pg
[15:58] <leftyfb> then what is it?
[15:58] <Sircle> its not about pg alone *
[15:58] <leftyfb> at 128GB, it should be pretty obvious what is using that much memory. It's aint the OS by itself
[16:10] <Sircle> leftyfb true. What is the second best thing after ram?
[16:12] <leftyfb> There isn’t, not when your limitation is memory
[16:12] <leftyfb> Reducing your ram footprint or increasing your ram. Those are your only options
[16:16] <tomreyn> Sircle: note that if you run DB's professionally, you would usually host databases on their separate server (or rather, multiple).
[16:17] <tomreyn> (ideally dedicated, or at least RAM optimized VMs)
[16:19] <tomreyn> put differently: mixing databases with other workloads on the same host is something you wouldn't do as projects reach a certain size
[16:19] <jhutchins> What does Ubuntu use in place of rc.local in these systemd times?
[16:20] <tomreyn> a systemd unit :)
[16:21] <jhutchins> There's no collective file that you can just add links to?  unit file for each program?
[16:21] <tomreyn> there'S an example somewhere in /usr/share/doc if you reeeeally want to bring back /etc/rc.local
[16:22] <jhutchins> Yeah, there are lots of examples, and LOTS of armwaving, and no two do it the same way.
[16:22] <Sircle> tomreyn yes, I am on that path. Not there yet. ture
[16:22] <jhutchins> Mostly it seems like a way for geeks to show off their obfuscation skills.
[16:22] <Sircle> true*
[16:22] <Sircle>  What is the second best thing after ram?
[16:23] <tomreyn> jhutchins: if it helps: i agree that this is functionality that's nice to have and i feel should be available by default. but that's just you and me, and a bunch of other systemd users who probably think weird. ;-)
[16:24] <tomreyn> Sircle: good copy and paste job there, do you want me to copy and paste the previously provided correct answer, too?
[16:24] <ravage> jhutchins: a default jammy install comes with /lib/systemd/system/rc-local.service
[16:27] <tomreyn> whoops, ravage knows better, thanks!
[16:29] <Sircle> tomreyn why a person asks twice?
[16:35] <noarb> is there a way to determine if a wireless card is being picked up by a kernel module and not and installed driver? I'm having difficulty getting an Realtek RTL8814AU working correctly
[16:35] <leftyfb> ravage: that just called whatever is in /etc/rc.local
[16:36] <ravage> that is correct
[16:36] <ravage> and i think that was his question?
[16:37] <jhutchins> I was hoping for something simpler.
[16:37] <leftyfb> yes and no. The question was is there a single file in systemd to add "links"(scripts?) to. Technically you could say that is rc.local if you enable the service with systemd. But systemd doesn't have anything built-in
[16:39] <leftyfb> jhutchins: systemd files really aren't all that complicated. They're a hell of a lot easier to deal with than sysVinit. And they give you a lot of flexibility. There's advantages to splitting out services into their own files
[16:39] <jhutchins> A service file probably makes the most sense, especially if you don't have to test for what default target and set a bunch of variables.
[17:23] <Macwinner> I use to be able to type "cat $ENVVAR/bl" and then hit the tab key, and then the shell would expand the $ENVVAR.. when I've moved to ubuntu, it is not escaping the dollar sign and turning it to "cat \$ENVVAR/bl".   Just moved from centos8 to ubuntu jammy.  So it's probably some default tweak I need to do to the shell
[17:25] <leprechaun> How do i reset to the default premissions in the users home directory
[17:26] <leftyfb> leprechaun: there is no "reset"
[17:27] <leprechaun> there used to be a string of values for that one
[17:28] <matrixy> you have to do chmod and chown
[17:28] <leftyfb> leprechaun: there was never a reset. You can set them to whatever you want. The home directory itself is typically drwxr-xr-x or 755
[17:29] <leftyfb> leprechaun: as for the contents, that's up to you. Feel free to chown them all to your user
[17:30] <leftyfb> leprechaun: if you have changed the actual permissions on everything in your home, you are either going to need to restore from backup or mess with permissions on a per application(settings) basis
[17:30] <LDericher> hey, I want to mount a disk at both /foo/path/to/mnt1 and /bar/path/to/mnt2, such that mnt1 and mnt2 share the disk's capacity but shouldn't see each other's contents. I could mount the disk at /mnt/mydisk and setup bind mounts from /mnt/mydisk/mnt1,mnt2 but I feel like there must be a more elegant way
[17:31] <LDericher> hmmm ... maybe this is better suited for #linux anywaya
[17:32] <leftyfb> LDericher: that is not possible. You can partition the disk into separate partitions to be mounted at separate locations. But mounting the same partition in multiple places will show all the contents in both places
[17:33] <LDericher> leftyfb, so bind mounts it is? I had hoped for some btrfs magic or smilar
[17:33] <leftyfb> regardless if it's a bind mount or not. A bind mount just allows you to mount any location to any other location. There's no magic in how it deals with the contents
[17:33] <leftyfb> LDericher: a bind mount will still show the contents of the partition in all mount points
[17:34] <LDericher> sure, but I can create directories /mnt/disk/mnt1 and /mnt/disk/mnt2 and then bind those to /foo/path/to/mnt1 and mnt2 respectively
[17:34] <LDericher> bee there, done that, but it doesn't feel right in 2023
[17:35] <leftyfb> right, so you'll need 3 mounts of the same partition
[17:35] <leftyfb> of/off
[17:35] <leftyfb> and one of the 3 will show all the contents
[17:39] <LDericher> exactly^^ and I would simply ignore the one with all contents for daily use, which is why I hoped for a more up-to-date approach
[17:46] <Sircle> How can I compress data in ram so less ram is used for same data?
[17:47] <Habbie> zram
[17:47] <Sircle> Habbie its an app?
[17:47] <Habbie> no
[17:48] <Habbie> zram (and zswap) are kernel features
[17:48] <leftyfb> Sircle: mind you, you'll be trading memory for cpu if you do that. And you can only compress the data so much.
[17:48] <Habbie> you might also be trading it for disk access, which can be a win
[17:48] <Habbie> but we can't predict the outcome, indeed
[17:49] <leftyfb> they've already got a 10GB swap
[17:49] <Sircle> Habbie trading for disk access? how?
[17:50] <Habbie> allocating a bit of memory to act as compressed swap -may- reduce use of on-disk swap
[17:51] <Habbie> first paragraph of https://en.wikipedia.org/wiki/Zswap
[17:51] <Habbie> > esulting in a significantly reduced I/O for Linux systems that require swapping; the tradeoff is the need for additional CPU cycles to perform the compression.
[17:52] <Habbie> +R
[17:52] <Sircle> "allocating a bit of memory to act as compressed swap -may- reduce use of on-disk swap" <-- this is a good thing. no?
[17:52] <Habbie> it -may- be a good thing
[17:53] <Habbie> i cannot judge the balance for you
[17:57] <leprechaun> sudo chown -R $USER:$USER /$HOME/$USER is it safe to do then?
[17:57] <leftyfb> leprechaun: no
[17:57] <leprechaun> ok thx
[17:57] <leftyfb> leprechaun: $HOME = /home/$USER
[17:58] <leftyfb> leprechaun: so your command would attempt to change ownership of /home/$USER/$USER
[17:58] <moonmoon> would it?
[17:58] <leprechaun> ok thx i se i did it wrong then
[17:58] <leftyfb> leprechaun: personally, I wouldn't play with vars for this sort of thing. Just use the proper username and locations
[17:59] <jhutchins> leftyfb: I always use full pathnames and explicit settings instead of variables.
[18:00] <leftyfb> jhutchins: right
[18:00] <moonmoon> also depends on env_reset/env_keep settings in /etc/sudoers whether $HOME, $USER, etc. are available at all
[18:00] <leftyfb> moonmoon: we tend to troubleshoot here assuming a default ubuntu installation
[18:00] <moonmoon> and the last thing you want is to expand to an empty string and then recursively chown /
[18:00] <moonmoon> fair
[18:01] <jhutchins> Empty "find ... exec" strings can be fun too.
[18:17] <Sircle> Habbie I think zram is working for me. Whats the simple way to change compresstion to zstd?
[18:23] <sarnold> Sircle: there's a few different ways to enable it, depending upon how you set up your zram config https://wiki.debian.org/ZRam
[18:23] <Sircle> sarnold I just sudo service zram-config start
[18:23] <Sircle> $ cat /sys/block/zram0/comp_algorithm
[18:23] <Sircle> lzo [lzo-rle] lz4 lz4hc 842 zstd
[18:24] <Sircle> I think lzo, lzo-rle gives 3:1 ration. Do not know about zstd?
[18:25] <Sircle> sarnold in btrfs, there are compression levels from 0-15 for zstd. Not sure what is in zram?
[18:26] <Sircle> sarnold this sudo zramctl  just shows the details. How do I edit those?
[18:27] <jhutchins> Compression for active storage is a bad idea.
[18:27] <arraybolt3> jhutchins: Several of us use zram to enable RAM compression, which works just fine.
[18:28] <leftyfb> jhutchins: I think at this point it's just best for them to learn on their own :)
[18:28] <arraybolt3> (At least I know I do this, and have seen others do it.)
[18:28] <sarnold> ± apt-file show zram-config
[18:28] <sarnold> zram-config: /etc/init/zram-config.conf
[18:28] <jhutchins> arraybolt3: It's not a matter of how it runs, but how it fails.
[18:28] <sarnold> Sircle: try editing that file ^^ ?
[18:29] <leftyfb> sarnold: what a poor location for a config file :)
[18:29] <Sircle> really? by puttin bracket around zstd?
[18:29] <sarnold> leftyfb: yes
[18:29] <arraybolt3> jhutchins: I mean, this is a specifically supported Linux kernel feature. Unless you've had it actually fail on you in the past due to some reason other than damaged RAM, maybe it's not as bad as you think?
[18:29] <Sircle> sarnold manually edit this? /sys/block/zram0/comp_algorithm
[18:29] <sarnold> leftyfb: but I thought it looked better than the first webpage I found that suggested editing the shell script directly :)
[18:29] <sarnold> Sircle: if you like to do that every boot, sure
[18:30] <Sircle> jhutchins Compression for active storage?
[18:30] <Sircle> jhutchins you mean compression via zram is bad vs using an nvme drive for swap?
[18:30] <jhutchins> Sircle: It's fine for archival storage.
[18:30] <leftyfb> Sircle: "files" in /sys aren't really meant to be edited. There are usually preferred methods to achieve the same goal
[18:31] <arraybolt3> jhutchins: If you could provide any reasoning to back up what you're saying, please do.
[18:31] <leftyfb> jhutchins: the biggest downside to zram is a hit to cpu performance. If they have room to spare, then it should be fine
[18:31] <Sircle> sarnold oh I missed. you meant to change this file?  /etc/init/zram-config.conf
[18:31] <jhutchins> arraybolt3: I think we're far enough off topic that I'll just drop it.  Compress all you want.
[18:31] <sarnold> Sircle: yeah, see if that works
[18:32] <Sircle> sarnold there is  no such file.
[18:33] <Guest3> Hello. Can somebody please help me with understanding how the Files application work. I've been trying whole day to make my PC automatically mount SMB shares from my NAS on user login. The username and password configured on NAS is the same as username and password on my PC. So when i press "Other locations" in Files app and then select an SMB
[18:33] <Guest3> share, it automatically mounts using current user credetenials. I want it to happen when i log in to system. I've found that it uses 'gio mount', but trying to make a bash script with that command to autorun on login fails because 'gio mount' asks for user credetentials to connect to SMB share instead of using those of a user that runs it. Is there
[18:33] <Guest3> a way to make it use the same credetentials?
[18:33] <Sircle> sarnold ther is one sudo nvim /usr/bin/init-zram-swapping
[18:33] <sarnold> Sircle: weird. dpkg -L zram-config and see if it moved between my system and your system?
[18:34] <leftyfb> Guest3: use fstab or a systemd mount file
[18:34] <leftyfb> Sircle: that is not a config file
[18:34] <Sircle> sarnold it has echo $mem > /sys/block/zram0/disksize
[18:35] <Sircle> should I do dpkg -L zram-config?
[18:36] <Guest3> leftyfb i tried using fstab but it requires writing password in fstab or storing it in separate file. I don't see the need to do that because SMB user/password is the same as user/password with which i am logged in and running everything already.
[18:36] <Sircle> sarnold https://termbin.com/h5tt
[18:36] <leftyfb> Guest3: the fact that the credentials are the same doesn't matter. They are separate authentication systems
[18:38] <Guest3> leftyfb but Files application somehow uses credetentials from one authentication system (Linux) for the other authetication system (SMB) - it doesn't ask for password when i click on SMB share in "Other Locations". How does it do that?
[18:38] <leftyfb> Guest3: right, those smb credentials are stored in your keychain agent. fstab won't use those
[18:39] <leftyfb> Guest3: AFAIK, nautilus doesn't have a way to perm mount network shared. You need to use fstab or systemd for that
[18:40] <Sircle> sarnold am there?
[18:41] <Guest3> leftyfb: then what will use that 'keychain'? As i've written above, apparently Files uses 'gio mount', but there isn't an option to use keychain with it when i run it from terminal.
[18:41]  * Sircle rebooting
[18:41] <leftyfb> Guest3: nautilus uses it
[18:41] <leftyfb> Guest3: You need to use fstab or systemd
[18:43] <Guest3> leftyfb: I tried different ways to do it with fstab but it doesn't work. All I really need is to do the same thing that happens when I click on SMB share in Files, but from terminal. How can it be impossible?
[18:44] <leftyfb> Guest3: You need to use fstab or systemd. Both will work just fine. Try following this and if you need help, let us know https://michlstechblog.info/blog/systemd-mount-examples-for-cifs-shares/
[18:48] <Guest3> leftyfb This method involves specifying password in plain text in .mount file, same as with fstab, which doesn't sound too secure.
[18:53] <Guest3> leftyfb: I did it with systemd like in that tutorial you linked but i cannot write to SMB share.
[18:54] <leftyfb> Guest3: do you have it setup like that right now?
[18:57] <Guest3> leftyfb: Yes. I specified it to mount to /mnt/share in systemd mount file, enabled and started it. It shows up as active and i can access files using that path. However, I cannot modify anything. But when I open the same share through Files app's "Other Locations" section, I have write permissions (can create folders, rename files, etc.). It's
[18:57] <Guest3> weird because the username and password i specified in mount file are the same as user i am currently logged on as.
[18:58] <leftyfb> Guest3: forget about the credentials you are logged in with, this isn't Windows. They have no relation
[18:58] <leftyfb> Guest3: ok, now unmount it
[18:59] <leftyfb> Guest3: let me know when it's unmounted
[19:00] <Guest3> Guest3: but the credetentials i am logged in with are used when i access share through Files app, no? I unmounted it now - the folder it mounts to now shows up empty.
[19:00] <leftyfb> n
[19:00] <leftyfb> no
[19:00] <leftyfb> Guest3: ok, now try to write a file to /mnt/share/
[19:00] <Guest3> leftyfb: i cannot
[19:00] <leftyfb> Guest3: and try to write to /mnt/
[19:01] <Guest3> leftyfb same
[19:01] <leftyfb> Guest3: there's your first problem
[19:02] <leftyfb> Guest3: set the appropriate permissions on your mount point first, then mount
[19:02] <arraybolt3> Something like "sudo chown Guest3:Guest3 /mnt/share", replacing Guest3 with your username.
[19:03] <Guest3> leftyfb: i changed owner and group of /mnt/share from root to my current user and still cannot write to the mounted share
[19:09] <Guest3> going to reboot now
[19:13] <leftyfb> Guest3: ,uid=1000,gid=1000
[19:13] <leftyfb> Guest3: add that to your Options
[19:13] <leftyfb> replacing 1000 with the uid and gid of the user
[19:13] <leftyfb> the user you want to have access to write
[19:16] <leftyfb> Guest3: you can also do something like this to make your plain text password slightly more secure then just being written to the systemd file https://unix.stackexchange.com/a/309772
[19:20] <Guest3> leftyfb: thank you a lot! after addin uid and guid to options i n the mount file I can finally write to share!
[19:20] <Guest3> I will also look into storing password in a separate file
[19:35] <mybalzitch> why does goa-daemon need to crash every time external internet goes down
[19:53] <gok> Hey all. Im trying to install some deps on an M1 mac and I'm getting 404's on a small number of packages. This was working earlier. Wondering if packages were removed from the host in the short time between my builds?
[19:53] <gok> Error:
[19:53] <gok> https://pastebin.com/bDNKeXDv
[19:58] <RKelly> :D
[19:58] <sarnold> gok: ugh
[19:58] <sarnold> gok: thanks
[20:00] <user____> hi
[20:04] <gok> sarnold: np:)
[20:04] <sarnold> gok: hopefully we'll get that sorted soon. I don't have an eta to suggest.
[20:05] <gok> gok: no worries. thanks for helping out!
[20:42] <sarnold> gok: try again?
[20:45] <gok> sarnold: works!! thank you very much, good sir.
[20:45] <sarnold> gok: thanks for the easy report :)
[21:25] <webchat574> hello world
[22:26] <bparker> what would cause a crontab to only work after the user logs in? (no encryption is being used anywhere)
[22:37] <sarnold> bparker: hmm, what's the problem you're trying to solve? are you creating a user and then a crontab immediately, and it doesn't run until after the user logs in? or does the user log in and out all the time, but the crontab only works when they're logged in?
[22:38] <sarnold> bparker: I don't actually like my first guess, https://www.freedesktop.org/software/systemd/man/loginctl.html#enable-linger%20USER%E2%80%A6  -- so I'm hoping there's something else that makes more sense :)
[22:50] <bparker> sarnold: after booting up the system, the cron tab simply will never run until I login as that user. I don't know if logging out would stop it or not
[22:51] <leftyfb> bparker: is the crontab the user's crontab or root?
[22:52] <bparker> the user's
[22:52] <bparker> and I don't know if lingering is related to cron at all
[22:52] <bparker> I wouldn't think so
[22:59] <Jim`> Is there any reason to upgrade from a stable 20.04.1 yet? Or just let the LTS upgrade in April ?
[23:00] <Habbie> Jim`, upgrade to what?
[23:00] <Jim`> Habbie: latest Lubuntu version
[23:00] <Habbie> which version do you mean, just to be clear?
[23:00] <Jim`> Habbie: It downloaded from the 20.04.1 repo so I assume I got that on there
[23:00] <Habbie> !lubuntu
[23:00] <Jim`> I aint messed with the machine in about a year
[23:01] <Jim`> Turned it on and updating it now
[23:03] <Jim`> I suppose I'll just wait for the LTS -> LTS
[23:03] <Habbie> if you're on 20.04, the next LTS is 22.04, which came out a year ago
[23:03] <Jim`> New release '22.04.2 LTS' available.
[23:03] <Jim`> Run 'do-release-upgrade' to upgrade to it.
[23:03] <leftyfb> and will not automatically upgrade on it's own
[23:03] <Jim`> Looks like it wants me to do that after it updated
[23:07] <jhutchins> Jim`: Well of course there's a reason.  The whole reason you're running linux is because you enjoy tackling new problems with the OS, finding them and solving them.  You don't actually use the system to do work.  Maybe to run Windows games off-license.
[23:08] <Jim`> jhutchins: I figure I might as well update that machine since it's a backup machine.
[23:08] <leftyfb> jhutchins: that statement isn't appropriate for everyone
[23:08] <Jim`> while I got it on
[23:08] <Jim`> For when the day comes I need to use it for real while I work on the WIndows machine..
[23:08] <Jim`> It'll be ready rather than need more work
[23:08] <jhutchins> leftyfb: No, and niether is upgrade fever.
[23:09] <jhutchins> https://arstechnica.com/information-technology/2023/03/unkillable-uefi-malware-bypassing-secure-boot-enabled-by-unpatchable-windows-flaw/
[23:10] <jhutchins> Be sure to enable SecureBoot!
[23:11] <Jim`> I thought Secure Boot was for Win OS' ??
[23:12] <jhutchins> Jim`: Some people bought the hype and made it "work" with Ubuntu.
[23:12] <ash_worksi> uh oh
[23:13] <Jim`> I remember having to turn Secure Boot off TO PUT Ubuntu on a machine!
[23:13] <Jim`> It wouldn't boot with it left on
[23:13] <ash_worksi> Jim`: I remember that
[23:13] <jhutchins> Jim`: That's why it was developed.
[23:13] <leftyfb> Jim`: that's fine. It's pretty irrelevant here
[23:13] <leftyfb> jhutchins: feel free to discuss it elsewhere
[23:13] <ash_worksi> peppridge farm remembers too
[23:13] <leftyfb> ash_worksi: can we help you with something?
[23:14] <Jim`> It's probably this same machine I just turned on
[23:14] <Jim`> For the first time in a while
[23:14] <ash_worksi> leftyfb: I figured it out just a moment ago
[23:14] <ash_worksi> I was going to say that I was concerned because my fan speed kicked into high gear and when i looked at top it was a php process started by root
[23:15] <ash_worksi> so I was going to ask about hunting down why that got respawned
[23:15] <ash_worksi> but it was just me starting a container
[23:16] <ash_worksi> dunno why it would run that hot, but whatever