General Linux Chat and Small Questions v. I broke my Arch Install - Hardware & Software

It's actually incredibly common to have a group for each user. Also reduces the risk of accidentally making all home dirs available to all unprivileged users or whatever.

Anyone have an alternative to puush (image cropping + automatic uploading) that works on linux?

[QUOTE=Hng;46203252]Anyone have an alternative to puush (image cropping + automatic uploading) that works on linux?[/QUOTE] [Url=https://github.com/naelstrof/slop]Slop[/URL], [Url=https://github.com/naelstrof/maim]maim[/URL], and pretty much [url=https://aur.archlinux.org/packages/imgurbash/]any image uploader from the AUR[/URL] combined into a simple bash script would do the trick.

gyazo works too in a pinch by default it links you to a page with the image embedded but if you want to make it link directly to the image you just edit the main ruby script so it puts '.png' on the end of the url in your clipboard like so: [img]http://gyazo.com/a0bda67079517ead18cc84b7d9e50ed5.png[/img] [editline]12th October 2014[/editline] this isn't as comprehensive a solution as the one posted above but it can be set up very quickly

[QUOTE=Hng;46203252]Anyone have an alternative to puush (image cropping + automatic uploading) that works on linux?[/QUOTE] ScreenCloud works rather nicely for me, it also allows you to connect to your own FTP server. There's no messing with shell scripts for it either.

[QUOTE=Hng;46203252]Anyone have an alternative to puush (image cropping + automatic uploading) that works on linux?[/QUOTE] Shutter if you need uploading. Gnome Screenshot is also good for quick screenshots if you're saving locally

[QUOTE=Little Donny;46207266]gyazo works too in a pinch by default it links you to a page with the image embedded but if you want to make it link directly to the image you just edit the main ruby script so it puts '.png' on the end of the url in your clipboard like so: [img]http://gyazo.com/a0bda67079517ead18cc84b7d9e50ed5.png[/img] [editline]12th October 2014[/editline] this isn't as comprehensive a solution as the one posted above but it can be set up very quickly[/QUOTE] That said, don't get gyazo-git off AUR I have no idea of my credentials and it's been flagged out-of-date and I don't even have it installed anymore [editline]11th October 2014[/editline] Er, I'm supposed to be the maintainer, that is

[QUOTE=Hng;46203252]Anyone have an alternative to puush (image cropping + automatic uploading) that works on linux?[/QUOTE] I used to use Shutter, but I found it very clunky and tedious to use compared to ShareX so I went with bash. with the help of Naelstrom, we cooked this up [code]#!/bin/bash _date=`date +%d-%m-%Y-%H:%M` _file="/home/vlad/scrots/$_date.png" scrot "$_file" notify-send -i /usr/share/icons/Numix-Circle/48x48/apps/xfce4-clipman-plugin.svg Snapshot 'Uploading...' scp -P port "$_file" apache@pred.me:~/public_html/pics notify-send -i /usr/share/icons/Numix-Circle/48x48/apps/xfce4-clipman-plugin.svg Snapshot 'Uploaded!' echo http://pred.me/pics/$_date.png | xclip -selection clipboard[/code] I have another script which does selections instead, which is pretty easy to do. replace scrot with maim using the -s flag, and you got a script that does selections as well. this script assumes you have a server that you have SSH access to + its private key and a web server though

Day/Month/Year in file names... eww.

I just use the UNIX timestamp for my screenshot names. Nice way of organising them chronologically but looks random enough to not bother me. And if I really want to know when I took it, I can just feed the filename to date.

i just toss it in /tmp before i sftp it to my domain or upload it to jesusfuck.me (when it isn't inexplicably down) since i'm using tmpfs there it gets deleted on shutdown without me even thinking about it

With all this talk about screenshots I decided to write up a better README.md for maim. It includes examples on how to use maim to take screenshots in a variety of ways, how to name each screenshot dynamically, and how to automatically upload the screenshots to Imgur. Some of it is applicable to scrot or import so some of you may be interested: [url]https://github.com/naelstrof/maim[/url]

[QUOTE=esalaka;46194027]It's actually incredibly common to have a group for each user. Also reduces the risk of accidentally making all home dirs available to all unprivileged users or whatever.[/QUOTE] Alternatively one could make home dirs + content correctly umask'd including in the users .profile and .*shrc

Need linux help here I'm running a TF2 dedicated server with 10 player MvM plugin and a plugin to increase visible maxplayers to 10, and the output spams my terminal with Setting sv_visiblemaxplayers to 6 for MvM The page says that using [code]./srcds_run -game tf +sv_pure 1 +map mvm_ghost_town.bsp +maxplayers 32 -debug +port 27015 | grep -v Setting\ sv_visiblemaxplayers\ to\ 6\ for\ MvM [/code] should remove it, but it didn't. I have tried [code]echo 'Setting sv_visiblemaxplayers to 6 for MvM' | grep -v Setting\ sv_visiblemaxplayers\ to\ 6\ for\ MvM[/code] and the grep successfully removes it. But why it doesn't remove the text from dedicated server output?

Every other kernel update breaks my mousepad. 3.15 - broken 3.16 - worked 3.16 some patch - broken 3.17 - working My touchpad man.

[QUOTE=Abaddon-ext4;46230809]Need linux help here I'm running a TF2 dedicated server with 10 player MvM plugin and a plugin to increase visible maxplayers to 10, and the output spams my terminal with Setting sv_visiblemaxplayers to 6 for MvM The page says that using [code]./srcds_run -game tf +sv_pure 1 +map mvm_ghost_town.bsp +maxplayers 32 -debug +port 27015 | grep -v Setting\ sv_visiblemaxplayers\ to\ 6\ for\ MvM [/code] should remove it, but it didn't. I have tried [code]echo 'Setting sv_visiblemaxplayers to 6 for MvM' | grep -v Setting\ sv_visiblemaxplayers\ to\ 6\ for\ MvM[/code] and the grep successfully removes it. But why it doesn't remove the text from dedicated server output?[/QUOTE] The only thing I can think of is srcds outputting to stderr instead of stdout. Try putting 2>&1 before the pipe. If that fails, try grep aaaaa (without -v) and check if it still outputs anything at all. (It shouldn't)

I'm setting up a stupid simple web server using lighttpd+CGI+Ruby+SQLite (no Rails). Is there a "proper" location for the .db files? It works fine if I put them in /srv/http but that seems wrong and it would make it really easy for someone to download the whole database if I fuck up the server config.

[QUOTE=Larikang;46233732]I'm setting up a stupid simple web server using lighttpd+CGI+Ruby+SQLite (no Rails). Is there a "proper" location for the .db files? It works fine if I put them in /srv/http but that seems wrong and it would make it really easy for someone to download the whole database if I fuck up the server config.[/QUOTE] /srv/http is hardly a standard location in the first place. Anyway, just make the document root a subdirectory of /srv/http, give the webserver permissions for both directories, and store your private files in /srv/http directly, outside of the web root.

My btrfs broke =( Everything still runs, but I have a few snapshots that are corrupt that I can't access nor delete. [code][ 506.737115] BTRFS error (device sda2): Error removing orphan entry, stopping orphan cleanup [ 506.737118] BTRFS critical (device sda2): could not do orphan cleanup -22 [ 506.737329] BTRFS: bad tree block start 0 62368972800 [ 506.737416] BTRFS: bad tree block start 0 149782528 [ 518.799008] BTRFS: bad tree block start 0 62368972800 [ 518.799628] BTRFS: bad tree block start 0 149782528 [ 546.541329] BTRFS: bad tree block start 0 62368972800 [ 546.541468] BTRFS: bad tree block start 0 149782528 [ 546.909539] BTRFS: bad tree block start 0 62368972800 [ 546.909628] BTRFS: bad tree block start 0 149782528 [ 547.266649] BTRFS: bad tree block start 0 62368972800 [ 547.266764] BTRFS: bad tree block start 0 149782528 [ 547.898513] BTRFS: bad tree block start 0 62368972800 [ 547.898638] BTRFS: bad tree block start 0 149782528 [ 549.748352] BTRFS: bad tree block start 0 62368972800 [ 549.769652] BTRFS: bad tree block start 0 62368972800[/code] [code]% ls ls: cannot access home_2014-10-12_03:00:03: Cannot allocate memory ls: cannot access home_2014-10-14_03:00:08: Cannot allocate memory home/ home_2014-10-12_03:00:03/ home_2014-10-14_03:00:08/ incbak* root/[/code] [editline]14th October 2014[/editline] Spoke too soon. Whole filesystem just shit itself [editline]14th October 2014[/editline] Best I can tell so far is that all metadata is just _gone_. It's recognised as a btrfs volume but that's as far as anything gets.

I finally got around to setting up an OS on my little Thin-ITX Celeron 847 to use on my workbench. Instead of being boring and shoving Windows on it, I had a copy of Elementary OS sitting around. I've used it briefly on my Thinkpad X201, though now that I've sat down and used it for actual work, I really like it. It's just so PRETTY.

I like certain parts about Elementary. Pantheon Terminal is a nice, clean terminal, if a bit too barebones. Gala is a nice WM. I wish there were more GTK3 compatible WMs. I really like Maynard a lot. [media]http://www.youtube.com/watch?v=irrv8OA30Bs[/media] Some combination of Gala/Maynard + XFCE/MATE (maybe tint2 idk) seems like the best bet ATM for good looks + speed, which you might want on a Thinkpad. You can get the same look if you use their apps or other similar GTK3 apps [editline]15th October 2014[/editline] Keep in mind tho that Maynard is a Wayland WM

What about xfce4-terminal? Clean interface with tabs.

[QUOTE=Darkwater124;46231168]The only thing I can think of is srcds outputting to stderr instead of stdout. Try putting 2>&1 before the pipe. If that fails, try grep aaaaa (without -v) and check if it still outputs anything at all. (It shouldn't)[/QUOTE] I have tried [code]./srcds_run -game tf +sv_pure 1 +map mvm_ghost_town.bsp +maxplayers 32 -debug +port 27015 2>&1 | grep -v Setting\ sv_visiblemaxplayers\ to\ 6\ for\ MvM[/code] but sv_visiblemaxplayers still spams the console I tried [code]./srcds_run -game tf +sv_pure 1 +map mvm_ghost_town.bsp +maxplayers 32 -debug +port 27015 | grep aaaaa [/code] and it still output everything although it shouldn't

So, I forced my mother to use Linux (Mint) to browse the internet now, since this was kind of all she was doing anyways on the laptop. All because I am working around 500 km away from home, cannot be give tech support when ever something happens and she manages to break the windows installs on regular basis, even with just normal user accounts. I thought she wouldn't mind since she only uses Firefox anyways, but oh boy was I wrong. She absolutely hates it! The reasons are the different UI, a giant box that says it is running in Software Rendering mode and GRUB loader. Every time she starts the laptop, she, for some what ever reason, selects safemode to boot resulting in the huge textbox and just double clicking the Firefox icon on the desktop is too much. I don't really want to set GRUB to autoboot or have a short countdown, because reasons, and I guess I can enable autostart on Firefox. But despite all of these "problems" the OS is still running without actual problems. So far the whole ordeal was entertaining and only sometimes gave me a headache unlike when ever a windows install broke.

If you set GRUB's timer to 0 seconds, you can still access GRUB by holding down shift. I'd do that. Or get a new mother.

[QUOTE=Abaddon-ext4;46238112]I have tried [code]./srcds_run -game tf +sv_pure 1 +map mvm_ghost_town.bsp +maxplayers 32 -debug +port 27015 2>&1 | grep -v Setting\ sv_visiblemaxplayers\ to\ 6\ for\ MvM[/code] but sv_visiblemaxplayers still spams the console I tried [code]./srcds_run -game tf +sv_pure 1 +map mvm_ghost_town.bsp +maxplayers 32 -debug +port 27015 | grep aaaaa [/code] and it still output everything although it shouldn't[/QUOTE] What if you do 2>&1 | grep aaaaa?

[QUOTE=IpHa;46235528]Spoke too soon. Whole filesystem just shit itself [editline]14th October 2014[/editline] Best I can tell so far is that all metadata is just _gone_. It's recognised as a btrfs volume but that's as far as anything gets.[/QUOTE] That's terrifying, considering I have all of my data backed up on a RAID 1 btrfs volume. Any idea what broke it? Kernel update? Drive failure?

I killed a btrfs volume by running an RC kernel that had a bug in it. Never had problems with non-RC releases though.

For those of you using maim or slop: Be weary the next time you upgrade them your scripts will break. I recently changed how they parse arguments to be more standard unix-like. I wasn't sure where to put this warning besides in the github commit log (that I'm sure nobody reads) and here. Hope it doesn't bother anyone.

[QUOTE=Larikang;46240568]That's terrifying, considering I have all of my data backed up on a RAID 1 btrfs volume. Any idea what broke it? Kernel update? Drive failure?[/QUOTE] Best guess right now is that there's a bug in 3.17.0 involving snapshots and/or volume sending. It started as a corrupt snapshot that I couldn't delete, then a few days later suddenly killed the whole FS. Don't upgrade to 3.17.0 if you're using btrfs, and if you're already on 3.17.0 don't create or move subvolumes. [URL]http://www.mail-archive.com/linux-btrfs@vger.kernel.org/msg38049.html[/URL] [editline]15th October 2014[/editline] Looks like a fix didn't make it into 3.17.1 either.