|
The homedir thing is easy, we're all actually in the same AD domain (we use QAS to auth against AD) and access the same storage, it's just an admistrative problem. Mainly, module files for department A not existing when you log into Department B, causing terminal errors. Software is a lot of engineering and math and statistics software. Matlab, Mathematica, ProE, Octave, etc etc. A lot of these clock in at 5-10GB a piece, which is why it's prohibitive to just install them all locally on the machine. I was thinking of just putting symlinks in /usr/bin or /usr/local/bin to the binary and calling it a day. Part of the problem is that some people just absolutely hate change, so the way we've always done is the way we should continue to do it going forward. Also, because we've done it this specific way for so long (each piece of software has its own automount map, each module adds that specific automounted directory to your path) a lot of people aren't able to separate the technologies in their head (if we add something to the path it has to be an automounted directory, if we automount a directory we have to add it to the path).
|
# ? Jun 13, 2014 18:16 |
|
|
# ? May 31, 2024 00:14 |
|
evol262 posted:You probably uninstalled your kernel, and the grub triggers removed it. Pretty soon, you'll also be able to do this on Fedora (kernel is currently handled specially in yum, though RPM will happily do it). Progress! Boot into it with a rescue CD and reinstall it. Grab packages and do it manually with dpkg -i if you have to. By "reinstall it" do you mean the kernel? How do I use dpkg to install into another system? Once I boot in with my recovery USB I assume if I do dpkg it will only install in the temporary USB system. And won't there still be that circular dependency issue?
|
# ? Jun 13, 2014 18:38 |
|
SurgicalOntologist posted:By "reinstall it" do you mean the kernel? How do I use dpkg to install into another system? Once I boot in with my recovery USB I assume if I do dpkg it will only install in the temporary USB system. And won't there still be that circular dependency issue? You'll want to use chroot, Polygynous fucked around with this message at 23:33 on Jun 13, 2014 |
# ? Jun 13, 2014 23:27 |
|
Somebody (I forget who, sorry!) asked for an effortpost on Wayland, so I went ahead and made one. It might be a bit too in depth, but hopefully it's enjoyable to read! I have some earlier effortposts if you want to read them as well, just go back a bit in the blog history. http://blog.mecheye.net/2014/06/xdg-shell/
|
# ? Jun 13, 2014 23:31 |
|
spoon0042 posted:You'll want to use chroot, Thanks, should be interesting to play around with, even if ultimately I end up flattening.
|
# ? Jun 13, 2014 23:47 |
|
the posted:I'm trying to verify my provider account on espn3 to stream the world cup. Apparently it depends on HAL for the DRM. See this post for details. In short, add the ppa:mjblenner/ppa-hal repository and install the hal package. Although this still hasn't fixed it for me. I can now get a pop-up of the event window, but I'm told that "WatchESPN cannot be accessed outside the US." And I am very much in the US. So, I don't know, maybe it'll work for you. e: vbcode is hard
|
# ? Jun 13, 2014 23:51 |
|
I'm trying to do something dumb with gnome. I have this DP>2xHDMI adapter for my laptop, which basically enumerates my two screens as a siingle 3840x1080 display. This causes issues such as windows maximizing across both screens. This is pretty annoying, but I've found that I can use CompizConfigSettingsManager and manually set the display geometry to 1920x1080+0+0 and 1920x1080+1920+0. The only problem is that this means I have to turn off display autodetect and manually fix them when I want to go to just using my laptop screen or anything else. What I would like to do is make some kind of script that detects display events, checks what displays are connected, and then modifies this field accordingly. There's really only three states--monitors attached and lid closed, monitors attached and lid open, and monitors detached and lid opened. Although, failing back to autodetect when it's not in one of those states would be awesome too. And getting it to work in lightdm would be really cool too, although probably less important. The problem is I can't find where this setting is actually stored. It's adjustable in CompizConfigSettingsManager, but it's not anywhere in gconftool, gconftool2 or gconf-editor. grepping ~/* for "3840" or "1920" doesn't seem to show anything either. I'd also like to know where I can put in hooks for the script. It's clear that when things change about the display configuration (i.e I open the lid or pull a cable) something gets run, so I'd like to just tweak this functionality. And no, fakeXinerama doesn't work. edit: this is ubuntu 14.04, using unity or whateverthefuck.
|
# ? Jun 14, 2014 03:08 |
|
So I hosed up pretty badly. I installed Windows 7, made a partition for Mint and followed a guide to install it. I started up my computer and Mint and Windows wouldn't start. I took out the HD, formatted the Linux partitions thinking I could get rid of grub but no luck That was pretty stupid, I know. Now I'm stuck in grub rescue command prompt with no idea how to log back into Windows. Any suggestions for a complete beginner? I copied stuff from my external HD back to my computer after installing Windows and I need those files.
|
# ? Jun 14, 2014 23:53 |
|
Suspicious Dish posted:Somebody (I forget who, sorry!) asked for an effortpost on Wayland, so I went ahead and made one. It might be a bit too in depth, but hopefully it's enjoyable to read! I have some earlier effortposts if you want to read them as well, just go back a bit in the blog history. This is pretty cool if a bit over my Head!
|
# ? Jun 15, 2014 01:21 |
|
Suspicious Dish posted:Somebody (I forget who, sorry!) asked for an effortpost on Wayland, so I went ahead and made one. It might be a bit too in depth, but hopefully it's enjoyable to read! I have some earlier effortposts if you want to read them as well, just go back a bit in the blog history. 5 star post
|
# ? Jun 15, 2014 01:43 |
|
.
astral fucked around with this message at 01:00 on Mar 3, 2016 |
# ? Jun 15, 2014 01:57 |
|
huhu posted:So I hosed up pretty badly. I installed Windows 7, made a partition for Mint and followed a guide to install it. I started up my computer and Mint and Windows wouldn't start. I took out the HD, formatted the Linux partitions thinking I could get rid of grub but no luck That was pretty stupid, I know. Now I'm stuck in grub rescue command prompt with no idea how to log back into Windows. Any suggestions for a complete beginner? I copied stuff from my external HD back to my computer after installing Windows and I need those files. If you still have your means of installing Mint just use the Live session to reinstall Grub or Mint.
|
# ? Jun 15, 2014 09:30 |
|
Suspicious Dish posted:Somebody (I forget who, sorry!) asked for an effortpost on Wayland, so I went ahead and made one. It might be a bit too in depth, but hopefully it's enjoyable to read! I have some earlier effortposts if you want to read them as well, just go back a bit in the blog history. Great post! 1 little nitpick I noticed: why is it xdg-shell with a dash while xdg_surface and all the rest is with an underscore?
|
# ? Jun 15, 2014 09:42 |
|
spankmeister posted:Great post! At a guess, xdg-shell is the name of the protocol, so it can be called anything, while xdg_surface is a C type and thus can't use hyphens in its name.
|
# ? Jun 15, 2014 14:55 |
|
ToxicFrog posted:At a guess, xdg-shell is the name of the protocol, so it can be called anything, while xdg_surface is a C type and thus can't use hyphens in its name. Oh yeah that makes sense.
|
# ? Jun 15, 2014 15:00 |
|
spankmeister posted:1 little nitpick I noticed: why is it xdg-shell with a dash while xdg_surface and all the rest is with an underscore? It's actually both. It's "xdg_shell" in the code, but the file is called "xdg-shell.xml". Either is correct, but xdg-shell looks better than xdg_shell to my eyes, so I use xdg-shell. I don't really care which you use. http://cgit.freedesktop.org/wayland/weston/tree/protocol/xdg-shell.xml#n2
|
# ? Jun 15, 2014 15:36 |
|
fourwood posted:Apparently it depends on HAL for the DRM. See this post for details. In short, add the ppa:mjblenner/ppa-hal repository and install the hal package. Although this still hasn't fixed it for me. I can now get a pop-up of the event window, but I'm told that "WatchESPN cannot be accessed outside the US." And I am very much in the US. So, I don't know, maybe it'll work for you. This fixed it, thanks so much!
|
# ? Jun 15, 2014 20:06 |
|
FISHMANPET posted:Software is a lot of engineering and math and statistics software. Matlab, Mathematica, ProE, Octave, etc etc. A lot of these clock in at 5-10GB a piece, which is why it's prohibitive to just install them all locally on the machine. I was thinking of just putting symlinks in /usr/bin or /usr/local/bin to the binary and calling it a day. I am trying to remember how this was handled at my university many years ago. I think it was a case of having each group getting an nfs share mounted under /usr/local/<group name> on each relevant machine. For instance Mathematica and Matlab would be installed under /usr/local/math with symlinks pointing there for files which really needed to be seen elsewhere. You of course then would need to ensure the computers were configured with the correct NFS mounts and licensing requirements. Seemed to work fine from a users point of view, I don't know how many headaches it caused for the people taking care of the backend however.
|
# ? Jun 15, 2014 23:18 |
|
Any idea why xfce4-terminal would be starting every time I log in on Debian? I checked Session and Startup in settings and couldn't find anything in there, and save session on log off is disabled. EDIT: Turns out session caches are a thing, and hitting the "clear saved sessions" button in the session tab of session and startup did the trick. I'm now working with a tricky OpenGL issue that I posted in the OpenGL thread if anyone knowledgable is able to help! Link Jehde fucked around with this message at 19:07 on Jun 17, 2014 |
# ? Jun 16, 2014 08:28 |
|
Suspicious Dish posted:Somebody (I forget who, sorry!) asked for an effortpost on Wayland, so I went ahead and made one. It might be a bit too in depth, but hopefully it's enjoyable to read! I have some earlier effortposts if you want to read them as well, just go back a bit in the blog history. How far are away are we from a usable/practical a gnome/wayland desktop?
|
# ? Jun 17, 2014 00:56 |
|
Any idea what that transparent box in the middle of the screen is and how I can get rid of it? It just showed up today and hovers on top of any application. click I've looked in System Monitor but honestly I don't really know enough about Linux to say what should or shouldn't be in there.
|
# ? Jun 17, 2014 16:42 |
|
looks like a broken tooltip, I see those in windows occasionally too. Usually need to just restart the session.
|
# ? Jun 17, 2014 16:54 |
|
RFC2324 posted:looks like a broken tooltip, I see those in windows occasionally too. That fixed it, thanks.
|
# ? Jun 17, 2014 17:00 |
|
peepsalot posted:How far are away are we from a usable/practical a gnome/wayland desktop? You can launch one today. Build a development copy of GNOME and try it out. Your help and testing will make it a success.
|
# ? Jun 17, 2014 17:44 |
|
evensevenone posted:I'm trying to do something dumb with gnome. I have this DP>2xHDMI adapter for my laptop, which basically enumerates my two screens as a single 3840x1080 display. This causes issues such as windows maximizing across both screens. This is pretty annoying, but I've found that I can use CompizConfigSettingsManager and manually set the display geometry to 1920x1080+0+0 and 1920x1080+1920+0. Anyone have any ideas for this? I feel like it's doable if I could just find where the pieces live.
|
# ? Jun 17, 2014 21:45 |
|
I tried looking for you in the Compiz source and got stuck in a maze of twisty plugins, all alike. Sorry.
|
# ? Jun 17, 2014 22:16 |
|
I'm working on a new architecture for an environment and one thing I'm focusing on is security. We have an web application that can allow you to access the local filesystem, the first thing I'm doing is making sure it has a service account for it to run under. The other thing I'd like to do is set it up to run in a jail. The web application runs on tomcat and it installs as a completely standalone product so java/tomcat/libraries are all bundled together and can run that way, the only dependency is R. I've done a little reading on setting up a jail and one thing that's really stressed is copying libraries into the jail, do I need to also move R into the jail? Does anyone have experience setting this up? I've looked at this Oreilly guide and it doesn't look like it will be too tricky, I just need to automate the process. I'd be running three instances of this web application on a server. Ashex fucked around with this message at 10:58 on Jun 18, 2014 |
# ? Jun 18, 2014 10:46 |
|
I need some Apache help, it's not going well. I'm trying to unfuck my Apache setup, I was doing DNS redirects to a couple test wordpress installs. I've set them up as Apache Vhost files and they don't work. The first one, eleganttailoring, works on IE (well, it pulls index.php which isn't working, but I'll deal with that later), but in Chrome it just pulls /var/www/index.html. The second one grabs index.html no matter what the browser. I've a2ensite them both, no errors. What the hell am I doing wrong? Apache2.conf Vhosts
|
# ? Jun 18, 2014 16:43 |
|
I found that running 'httpd -S' can sometimes reveal how things are ordered or domains are being loaded. It might not help, but worth a try.
|
# ? Jun 18, 2014 17:38 |
|
Pudgygiant posted:I need some Apache help, it's not going well. I'm trying to unfuck my Apache setup, I was doing DNS redirects to a couple test wordpress installs. I've set them up as Apache Vhost files and they don't work. The first one, eleganttailoring, works on IE (well, it pulls index.php which isn't working, but I'll deal with that later), but in Chrome it just pulls /var/www/index.html. The second one grabs index.html no matter what the browser. I've a2ensite them both, no errors. What the hell am I doing wrong? Do you have the namevirtualhost directive anywhere ? I can't see it in your files ?
|
# ? Jun 18, 2014 18:17 |
|
Yeah, it's in conf.d/virtual.conf, sorry. NameVirtualHost *:80 is the full contents of that file. Right now the only enabled vhost is optionsrealty. Somehow it's loading /var/www/index.html even when I directly type index.php into the browser. Might be some hosed up caching issue, this should work.
|
# ? Jun 18, 2014 18:50 |
|
Pudgygiant posted:even when I directly type index.php into the browser. Might be some hosed up caching issue. Always test this kind of thing with wget/curl/netcat.
|
# ? Jun 18, 2014 19:11 |
|
Pudgygiant posted:Yeah, it's in conf.d/virtual.conf, sorry. NameVirtualHost *:80 is the full contents of that file. Right now the only enabled vhost is optionsrealty. Somehow it's loading /var/www/index.html even when I directly type index.php into the browser. Might be some hosed up caching issue, this should work. Seems to work for me, I'm getting an unstyled wordpress site quote:Options Realty Have you checked the wordpress mod-rewrite rules are correct ? If you generated the .htaccess file then changed the vhost config at all, the rewrite base could be wrong or something similar.
|
# ? Jun 18, 2014 19:35 |
|
It totally is a caching thing, must be this lovely Afghanistan internet doing local caching. Thanks for all the help though
|
# ? Jun 18, 2014 19:43 |
|
I had a bash script today w/ a variable containing "-E". I wanted to echo this variable using the echo command. I couldn't do it without using some other tool because no matter what I did echo ate the -E as an argument. Is echo broken? code:
|
# ? Jun 18, 2014 21:49 |
Wow that's pretty strange! code:
|
|
# ? Jun 18, 2014 22:30 |
|
If the variable contains nothing but "-E" then yeah, you're literally passing -E as an argument to echo. Here's a lazy fix: add a space in the variable, or in the quotes when you're calling echo. ie: code:
code:
|
# ? Jun 18, 2014 22:45 |
|
Dren posted:I had a bash script today w/ a variable containing "-E". echo -e "\x2DE"
|
# ? Jun 18, 2014 22:45 |
|
fletcher posted:Wow that's pretty strange! Everything but -E works here because nothing but -E is a valid argument to echo. -E is a valid argument, so you have to do stupid things with unicode (like you did) or hexcode (like I did)
|
# ? Jun 18, 2014 22:47 |
|
|
# ? May 31, 2024 00:14 |
|
Dren posted:I had a bash script today w/ a variable containing "-E". http://unixhelp.ed.ac.uk/CGI/man-cgi?echo It'll do the same for strings with "-n" and "-e" because they are command flags for echo. edit: well beaten
|
# ? Jun 18, 2014 22:48 |