The Linux Questions Thread: a bunch of pitfalls, but technically it's possible

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > The Linux Questions Thread: a bunch of pitfalls, but technically it's possible

«‹›1031 »

nitrogen: May 21, 2004; Oh, what's a 217°C difference between friends?

apt-cache search libgtk

you probably want libgtk2.0-0, libgtk2.0-common and libgtk2.0-dev.

also, there's apt-file that'll find specific files inside packages.

# ? Jun 24, 2011 09:14

Adbot: ADBOT LOVES YOU

# ? May 21, 2024 01:35

angrytech: Jun 26, 2009

angrytech fucked around with this message at 00:03 on Oct 17, 2013

# ? Jun 24, 2011 20:07

dolicf: Sep 12, 2010

Reverse DNS is almost always handled by your ISP/host. You'll probably need to contact whoever provides your IPs to have them create the PTR record(s).

Edit: FWIW, your name server is responding properly to the PTR request, but it doesn't matter much as requests won't go to your server.

code:

slice:~# dig @30.14.88.69 -x 30.14.88.69 +short
poopybutts.net.

Somebody fucked around with this message at 04:54 on Jan 9, 2016

# ? Jun 24, 2011 20:38

angrytech: Jun 26, 2009

angrytech fucked around with this message at 00:04 on Oct 17, 2013

# ? Jun 24, 2011 21:16

dolicf: Sep 12, 2010

Definitely. Hit up whoever provides the VPS. It should be as simple as opening up a support ticket and asking that they create a PTR record for you.

# ? Jun 24, 2011 21:32

bort: Mar 13, 2003

angrytech posted:

Since I'm running this on a VPS, would that still hold true?

Since they own the IP address allocation, their authoritative DNS handles the PTR records.

# ? Jun 24, 2011 22:41

icantfindaname: Jul 1, 2008

Sorry to keep bothering you guys, but it still doesn't want to compile for me.
When I try to run gcc it says that the package gtk+-3.0 was not found.
The apt-file list shows that package file as being in /usr/lib/pkgconfig, but when I actually look there there's nothing. What's going on?

code:

$ gcc base.c -o base `pkg-config --cflags --libs gtk+-3.0`
Package gtk+-3.0 was not found in the pkg-config search path.
Perhaps you should add the directory containing `gtk+-3.0.pc'
to the PKG_CONFIG_PATH environment variable
No package 'gtk+-3.0' found
base.c:1:21: fatal error: gtk/gtk.h: No such file or directory
compilation terminated.

but apt-file:

code:

$ apt-file list libgtk-3-dev
.
(etc)
.
libgtk-3-dev: /usr/include/gtk-3.0/gtk/gtk.h
.
(etc)
.
libgtk-3-dev: /usr/lib/pkgconfig/gdk-3.0.pc
libgtk-3-dev: /usr/lib/pkgconfig/gdk-x11-3.0.pc
libgtk-3-dev: /usr/lib/pkgconfig/gtk+-3.0.pc
libgtk-3-dev: /usr/lib/pkgconfig/gtk+-unix-print-3.0.pc
libgtk-3-dev: /usr/lib/pkgconfig/gtk+-x11-3.0.pc
.
(etc)

and then the actual directory /usr/lib/pkgconfig:

code:

/usr/lib/pkgconfig$ ls
banshee-collection-indexer.pc  banshee-lastfm.pc       banshee-webbrowser.pc  ginn.pc                
banshee-core.pc                banshee-mono-media.pc   compiz-animation.pc    gnome-screensaver.pc   
banshee-hyena-data-sqlite.pc   banshee-musicbrainz.pc  compiz-mousepoll.pc    gnome-system-tools.pc  
banshee-hyena-gui.pc           banshee-nowplaying.pc   compiz-text.pc         ibus-table.pc          
banshee-hyena.pc               banshee-services.pc     dbus-python.pc         nautilus-sendto.pc     
banshee-lastfm-gui.pc          banshee-thickclient.pc  fontutil.pc            notify-python.pc

(the last column of that is gone b/c of (even more) table breakage. I'm pretty sure it's not important)

ps sorry for the tables

icantfindaname fucked around with this message at 23:40 on Jun 24, 2011

# ? Jun 24, 2011 23:36

dolicf: Sep 12, 2010

Anyone pretty familiar with Ghostscript? I'm not really at all, but am trying to get it installed on a RHEL 5.6 box for a client so they can convert PDF -> JPG. I chose the easy route and went from RPMs, which worked fine with a test PDF I created. The client let me know today that they're not able to convert any of the PDFs they have on hand due to some JPEG2000 issues. I fixed those by scrapping the RPM install for a shiny new source installation, but now I'm getting an even more cryptic error on these PDFs. No problems on my test PDF.

Here's what I get when I simply try to load a PDF into Ghostscript:

quote:

GPL Ghostscript 9.02 (2011-03-30)
Copyright (C) 2010 Artifex Software, Inc. All rights reserved.
This software comes with NO WARRANTY: see the file PUBLIC for details.
Processing pages 1 through 6.
Page 1
Loading NimbusRomNo9L-Regu font from %rom%Resource/Font/NimbusRomNo9L-Regu... 3533528 2103018 4051888 2748176 3 done.
Loading NimbusSanL-Regu font from %rom%Resource/Font/NimbusSanL-Regu... 3566808 2202821 4125536 2824422 3 done.
Loading NimbusSanL-Bold font from %rom%Resource/Font/NimbusSanL-Bold... 3640456 2312750 4219368 2906820 3 done.
Loading NimbusRomNo9L-Medi font from %rom%Resource/Font/NimbusRomNo9L-Medi... 3803528 2476252 4313200 3005539 3 done.
Error: /undefined in --run--
Operand stack:
--dict:7/16(L)-- false --dict:13/20(L)-- --dict:1/1(L)-- --nostringval-- FlateDecode --dict:1/1(L)-- 0
Execution stack:
%interp_exit .runexec2 --nostringval-- --nostringval-- --nostringval-- 2 %stopped_push --nostringval-- --nostringval-- --nostringval-- false 1 %stopped_push 1894 1 3 %oparray_pop 1893 1 3 %oparray_pop 1877 1 3 %oparray_pop --nostringval-- --nostringval-- 2 1 6 --nostringval-- %for_pos_int_continue --nostringval-- --nostringval-- --nostringval-- --nostringval-- %array_continue --nostringval-- false 1 %stopped_push --nostringval-- %loop_continue --nostringval-- 2318189 --nostringval-- --nostringval-- --nostringval-- --nostringval-- --nostringval-- --nostringval-- %array_continue --nostringval-- --nostringval--
Dictionary stack:
--dict:1150/1684(ro)(G)-- --dict:1/20(G)-- --dict:82/200(L)-- --dict:82/200(L)-- --dict:108/127(ro)(G)-- --dict:295/300(ro)(G)-- --dict:23/30(L)-- --dict:6/8(L)-- --dict:24/40(L)-- --dict:1/1(ro)(G)-- --dict:8/18(L)--
Current allocation mode is local
Last OS error: 2
GPL Ghostscript 9.02: Unrecoverable error, exit code 1

I was thinking this might be font related because there's a lot of wackiness associated with the four or five different GS installs that were there previously, but I could be completely off base here. Anyone a GS guru that can lend a hand? The GS IRC channel is a ghost town.

Edit: changed code to quote to prevent breaking hscroll.

# ? Jun 24, 2011 23:51

Bob Morales: Aug 18, 2006; ~~Just wear the fucking mask, Bob~~

I don't care how many people I probably infected with COVID-19 while refusing to wear a mask, my comfort is far more important than the health and safety of everyone around me!

icantfindaname posted:

Sorry to keep bothering you guys, but it still doesn't want to compile for me.
When I try to run gcc it says that the package gtk+-3.0 was not found.
The apt-file list shows that package file as being in /usr/lib/pkgconfig, but when I actually look there there's nothing. What's going on?

Try -I/usr/include/gtk-3.0 so that gcc knows where to find gtk/gtk.h

# ? Jun 24, 2011 23:54

ToxicFrog: Apr 26, 2008

icantfindaname posted:

The apt-file list shows that package file as being in /usr/lib/pkgconfig, but when I actually look there there's nothing. What's going on?

Are you quite sure that you have libgtk-3-dev installed?

# ? Jun 25, 2011 00:17

ShoulderDaemon: Oct 9, 2003; support goon fund; Taco Defender

icantfindaname posted:

Sorry to keep bothering you guys, but it still doesn't want to compile for me.
When I try to run gcc it says that the package gtk+-3.0 was not found.
The apt-file list shows that package file as being in /usr/lib/pkgconfig, but when I actually look there there's nothing. What's going on?

You are missing a file. apt-file confirms that the file you are missing is in the package libgtk-3-dev. You should install that package, so that you will no longer be missing that file.

# ? Jun 25, 2011 00:19

icantfindaname: Jul 1, 2008

ToxicFrog posted:

Are you quite sure that you have libgtk-3-dev installed?

Well, now I feel stupid. Turns out libgtk-3-dev wasn't actually installed. Thanks for the help, I'll crawl back into my hole now

# ? Jun 25, 2011 00:40

Megaman: May 8, 2004; I didn't read the thread BUT...

I've installed debian 6 with dwm on my desktop that has an ATI card, I have 3 questions.

1) how do I install ATI drivers

2) how do I reconfigure dwm

3) how do i disable capslock on login

# ? Jun 25, 2011 18:55

Dinty Moore: Apr 26, 2007

Megaman posted:

I've installed debian 6 with dwm on my desktop that has an ATI card, I have 3 questions.

1) how do I install ATI drivers

I'm not sure about how this is done in Debian; I know in Ubuntu, you just use the "Additional Drivers" tool. You should be able to enable the non-free repository for Debian, run "apt-get update ; apt-get install fglrx fglrx-amdcccle", and then use its command line tool to configure Xorg to use the fglrx driver. You'll probably also need to add your user account to the 'video' group.

Megaman posted:

2) how do I reconfigure dwm

I assume you mean "Compiz" in this case; once you have hardware acceleration working, run "apt-get install compiz compiz-gnome", and once everything is installed, run "compiz --replace &" in a terminal window. That should enable desktop compositing.

Megaman posted:

3) how do i disable capslock on login

On Ubuntu, I do this by choosing System -> Preferences -> Keyboard, going to the Layouts tab, clicking the Options... button, and under "Ctrl key positioning", selecting "Make Caps Lock an additional Ctrl". I believe it should be similar on Debian.

# ? Jun 25, 2011 21:08

ExcessBLarg!: Sep 1, 2001

dolicf posted:

Anyone pretty familiar with Ghostscript?

It looks like it might be coughing on FlateDecode. By chance did you compile Ghostscript without the zlib development package installed? Usually things complain, but Ghostscript might've just disabled support for it since I don't think it's used that extensively by PostScript (but very much so by PDF).

Anyways, the real problem is that your client is dealing with PDF files that require (PDF 1.5) JPXDecode support for JPEG2000-encoded images. Ghostscript (should) support this if compiled with JasPer, which I imagine Ghostscript is optionally dependent on so you'll need to make sure that library (and any dev package) is installed before building.

Depending on how flexible your client is, a better/alternate option might be to install Poppler (forked from Xpdf). In particular, the "pdftoppm" program can be used to rasterize a PDF to ppm or png formats, and you can use ImageMagick's "convert" to make JPEGs out of those. I generally try pdftoppm first as Ghostscript is slow and sometimes buggy, although I've encountered PDFs that Poppler craps out on but Ghostscript handles like a champ.

Anyways, Poppler supports JPXDecode via OpenJPEG, so you'll need the relevant library installed for that if you're building from source again.

# ? Jun 25, 2011 21:15

ExcessBLarg!: Sep 1, 2001

Megaman posted:

1) how do I install ATI drivers

It's been a while since I last used an ATI card, and it's quite old by now (R300)? The reverse-engineered driver worked best, and should "just work" if you have the xserver-xorg-video-ati package installed. You may have to pull firmware-linux-nonfree too. See AtiHowto wiki page.

If you want to use the proprietary fglrx driver, which may have the best support for whatever card you have, but has otherwise always sucked, you'll have to do some magic. In sid, it looks like installing the "fglrx-driver" should take care of everything but fixing up /etc/X11/xorg.conf. The procedure might be different in squeeze. Anyways, see ATIProprietary.

Megaman posted:

3) how do i disable capslock on login

Again it's been a while, but in squeeze and after you should be able to disable capslock system-wide by "dpkg-reconfigure keyboard-configuration" and entering "ctrl:swapcaps" or "ctrl:nocaps" (for swapping with ctrl, or replacing so you have two controls) at the prompt that allows you to enter arbitrary string options. You may have to install the keyboard-configuration package first, as I think it was optional at one point.

# ? Jun 25, 2011 21:25

dolicf: Sep 12, 2010

ExcessBLarg! posted:

It looks like it might be coughing on FlateDecode.

Thanks for the ideas! I actually checked in on the Ghostscript IRC channel and they advised me to open a bug report. It looks like this is actually a known issue in that Ghostscript can't yet handle the case where a PDF's images have multiple decoding filters one of which being JPXDecode, so I guess we're waiting for a patch. I'll see if the client wants to investigate another route.

# ? Jun 25, 2011 22:08

Megaman: May 8, 2004; I didn't read the thread BUT...

Dinty Moore posted:

I'm not sure about how this is done in Debian; I know in Ubuntu, you just use the "Additional Drivers" tool. You should be able to enable the non-free repository for Debian, run "apt-get update ; apt-get install fglrx fglrx-amdcccle", and then use its command line tool to configure Xorg to use the fglrx driver. You'll probably also need to add your user account to the 'video' group.

I assume you mean "Compiz" in this case; once you have hardware acceleration working, run "apt-get install compiz compiz-gnome", and once everything is installed, run "compiz --replace &" in a terminal window. That should enable desktop compositing.

On Ubuntu, I do this by choosing System -> Preferences -> Keyboard, going to the Layouts tab, clicking the Options... button, and under "Ctrl key positioning", selecting "Make Caps Lock an additional Ctrl". I believe it should be similar on Debian.

I didn't clarify, I meant all from the command line, and through the window manager dwm. No compiz.

# ? Jun 26, 2011 04:07

Lonely Wolf: Jan 20, 2003; Will hawk false idols for heaps and heaps of dough.

haha who uses compiz with dwm? You configure dwm by applying patches, editing config.h, sometime writing a little code, and recompiling. Most stuff you can do just from editing config.h and recompiling though.

# ? Jun 26, 2011 09:29

dont skimp on the shrimp: Apr 23, 2008

Megaman posted:

I didn't clarify, I meant all from the command line, and through the window manager dwm. No compiz.

DWM requires you to edit the source files and recompile it. No way around it AFAIK.

You could give awesome a try though, if you don't like recompiling software.

# ? Jun 26, 2011 09:46

Underflow: Apr 4, 2008; EGOMET MIHI IGNOSCO

What would be the most elegant alternative to mountpoint for pre-2004 systems? I want to check for a mounted Samba share in rc.local. Don't like to, but I'm still spamming kernel messages over all terminals.

# ? Jun 26, 2011 15:55

spankmeister: Jun 15, 2008; Can't post for 12 hours!

Underflow posted:

What would be the most elegant alternative to mountpoint for pre-2004 systems? I want to check for a mounted Samba share in rc.local. Don't like to, but I'm still spamming kernel messages over all terminals.

I don't fully understand your question. You want to check if a remote server has a mountpoint available?

Just guessing here, but what if you'd just put it in fstab (fs would be cifs) and use the "bg" option meaning the system will try to mount it at boot but continue booting if it's not available.

e: actually that option may be specific to NFS, need to check up on that.

# ? Jun 26, 2011 16:03

Underflow: Apr 4, 2008; EGOMET MIHI IGNOSCO

spankmeister posted:

I don't fully understand your question. You want to check if a remote server has a mountpoint available?

Just guessing here, but what if you'd just put it in fstab (fs would be cifs) and use the "bg" option meaning the system will try to mount it at boot but continue booting if it's not available.

e: actually that option may be specific to NFS, need to check up on that.

I have an earlier version of Samba cause the share is on a Win95 machine. Cifs only does NTFS, yep.

I do this atm:

mountpoint -q /mnt/X || smbmount //X/pub /mnt/X -o password=Y,username=Z

but mountpoint is from 2004 and I need to do the same thing on older machines. Upgrading is not an option cause of custom hardware drivers that would be a huge pain to rewrite.

e: mountpoint checks if there's anything mounted at all under the specified mountpoint - on the local machine

# ? Jun 26, 2011 16:12

spankmeister: Jun 15, 2008; Can't post for 12 hours!

Hmm you could use stat to see what device a directory is on, if it's mounted it will be different from it's parent. You could also compare the path to see if it's in /etc/mtab, or even do some grepping on the output of "mount".

I like the stat option best.

# ? Jun 26, 2011 16:26

Underflow: Apr 4, 2008; EGOMET MIHI IGNOSCO

spankmeister posted:

Hmm you could use stat to see what device a directory is on, if it's mounted it will be different from it's parent. You could also compare the path to see if it's in /etc/mtab, or even do some grepping on the output of "mount".

I like the stat option best.

Okay, thanks; will do.

# ? Jun 26, 2011 16:31

spankmeister: Jun 15, 2008; Can't post for 12 hours!

Underflow posted:

Okay, thanks; will do.

Did some experimentation:

code:

root@pauperbak:~# stat test
  File: `test'
  Size: 4096            Blocks: 8          IO Block: 4096   directory
Device: fb02h/64258d    Inode: 525013      Links: 2
<snip>

root@pauperbak:~# stat -f test
  File: "test"
    ID: 579e2cbc7e291e70 Namelen: 255     Type: ext2/ext3
<snip>

Now mounted:

code:

root@pauperbak:~# mount -t cifs //localhost/data test/
root@pauperbak:~# mount
//localhost/data/ on /root/test type cifs (rw,mand)
<snip>

root@pauperbak:~# stat test
  File: `test'
  Size: 0               Blocks: 0          IO Block: 16384  directory
Device: 13h/19d Inode: 2           Links: 9
<snip>

root@pauperbak:~# stat -f test
  File: "test"
    ID: 0        Namelen: 4096    Type: cifs
<snip>

As you can see; file system type and device change.

# ? Jun 26, 2011 16:42

Underflow: Apr 4, 2008; EGOMET MIHI IGNOSCO

spankmeister posted:

Did some experimentation:

As you can see; file system type and device change.

Great, thanks a lot!

# ? Jun 26, 2011 16:50

brc64: Mar 21, 2008; I wear my sunglasses at night.

Yesterday I took my old PC, replaced the crappy old small IDE drives with a nice 1TB SATA and installed Ubuntu Server. (I figured Ubuntu is one of the most widely supported distributions around right now for novice-ish users). It's primary purpose, at the moment, is a file server for all of our media.

I used a combination of guides to gets samba working about as easily as possible (basically with a complete lack of security... but the server is only accessible on our LAN, so that's fine with me), but I am running into a slightly annoying problem.

I'm using k9copy to rip our DVD collection, which works fairly well, but I can't get it to output to my samba shared directory. I have to output my ISOs to my home folder and use sudo to move them to the share, so obviously I am not understanding permissions correctly.

The way the guide I used had me set up the shares was with nobody:nogroup permissions, which as best I can tell is essentially a lack of security (equivalent to Everyone in Windows?) But if that were the case, I wouldn't think I would need to use sudo in order to put files there. Windows PCs can write files to the share with no issue.

===

2nd question: Right now, I've just got the one drive because I'm on a budget, but in the future I plan on adding significantly more storage. The Linux file system is still fairly mysterious to me, but I understand it's capable of some pretty magical things. If I add a second harddrive, will I be able to make that space available to my existing samba share, or will I need to create a new share when I get to that point?

# ? Jun 26, 2011 17:24

Notorious b.s.d.: Jan 25, 2003; by Reene

brc64 posted:

The way the guide I used had me set up the shares was with nobody:nogroup permissions, which as best I can tell is essentially a lack of security (equivalent to Everyone in Windows?) But if that were the case, I wouldn't think I would need to use sudo in order to put files there. Windows PCs can write files to the share with no issue.

When your Windows PCs access Samba without authenticating, Samba is mapping the Windows anonymous user to "nobody." They're writing as "nobody."

Your user is trying to write as your user. You aren't "nobody."

brc64 posted:

2nd question: Right now, I've just got the one drive because I'm on a budget, but in the future I plan on adding significantly more storage. The Linux file system is still fairly mysterious to me, but I understand it's capable of some pretty magical things. If I add a second harddrive, will I be able to make that space available to my existing samba share, or will I need to create a new share when I get to that point?

You *can* add a new drive to the existing filesystem, but you *shouldn't*.

Buy two new drives, create a software RAID-1 array, and migrate your shares to the RAID-1. (You don't really care if you lose the root drive, it's easy to backup configs and reinstall the OS.)

When the time comes, you'll turn off samba, move all the data to a new filesystem on the RAID-1, and mount that filesystem at the old mountpoint. If the new LVM volume is foo/bar, and your shares are at /samba-shares, here's a cheap example:

code:

mkfs.ext4 /dev/mapper/foo-bar 
mount /dev/mapper/foo-bar /mnt
cd /samba-shares && tar --one-filesystem -cpf - . | (cd /mnt; tar -xvf -)
rm -rf /samba-shares/*
umount /mnt
mount /dev/mapper/foo-bar /samba-shares

# ? Jun 26, 2011 17:31

brc64: Mar 21, 2008; I wear my sunglasses at night.

Empty Threats posted:

When your Windows PCs access Samba without authenticating, Samba is mapping the Windows anonymous user to "nobody." They're writing as "nobody."

Your user is trying to write as your user. You aren't "nobody."

Okay, that makes some bit of sense. Is there anything I can do to make it equally easy for Windows users and myself to write to that location?

quote:

You *can* add a new drive to the existing filesystem, but you *shouldn't*.

Buy two new drives, create a software RAID-1 array, and migrate your shares to the RAID-1. (You don't really care if you lose the root drive, it's easy to backup configs and reinstall the OS.)

When the time comes, you'll turn off samba, move all the data to a new filesystem on the RAID-1, and mount that filesystem at the old mountpoint. If the new LVM volume is foo/bar, and your shares are at /samba-shares, here's a cheap example:
code:
mkfs.ext4 /dev/mapper/foo-bar 
mount /dev/mapper/foo-bar /mnt
cd /samba-shares && tar --one-filesystem -cpf - . | (cd /mnt; tar -xvf -)
rm -rf /samba-shares/*
umount /mnt
mount /dev/mapper/foo-bar /samba-shares

While this does look like something I will want to do (couldn't afford a second drive for RAID just yet), I'm mostly interested in extending storage. It's starting to become apparent that even 1 TB isn't going to be enough, at least not long term. So at some point I will want to be adding drives not just for redundancy, but for additional storage as well. Given your warning, however, it sounds like it might be safer to just create a second share (on a new filesystem?) when I do that.

# ? Jun 26, 2011 17:43

Underflow: Apr 4, 2008; EGOMET MIHI IGNOSCO

brc64 posted:

Yesterday I took my old PC, replaced the crappy old small IDE drives with a nice 1TB SATA and installed Ubuntu Server. (I figured Ubuntu is one of the most widely supported distributions around right now for novice-ish users). It's primary purpose, at the moment, is a file server for all of our media.

How big are the IDEs, are they okay, and do you still have them?

brc64 posted:

I used a combination of guides to gets samba working about as easily as possible (basically with a complete lack of security... but the server is only accessible on our LAN, so that's fine with me), but I am running into a slightly annoying problem.

I'm using k9copy to rip our DVD collection, which works fairly well, but I can't get it to output to my samba shared directory. I have to output my ISOs to my home folder and use sudo to move them to the share, so obviously I am not understanding permissions correctly.

The way the guide I used had me set up the shares was with nobody:nogroup permissions, which as best I can tell is essentially a lack of security (equivalent to Everyone in Windows?) But if that were the case, I wouldn't think I would need to use sudo in order to put files there. Windows PCs can write files to the share with no issue.

I use this with smbmount mounting a share on an old FAT fs, but the options may still be valid:

-o dmask=770,fmask=660,gid=wheel,password=$your_pwd,username=$your_uid

For total freedom you can set the dmask to 777, the fmask to 666, the gid to users.

brc64 posted:

2nd question: Right now, I've just got the one drive because I'm on a budget, but in the future I plan on adding significantly more storage. The Linux file system is still fairly mysterious to me, but I understand it's capable of some pretty magical things. If I add a second harddrive, will I be able to make that space available to my existing samba share, or will I need to create a new share when I get to that point?

You can make it available as long as the permissions on that drive/partition make it available to who you want (don't know if the initial perms used when you mounted the smb share will carry over) and place a symbolic link in the original Samba share's directory

# ? Jun 26, 2011 17:44

brc64: Mar 21, 2008; I wear my sunglasses at night.

Underflow posted:

How big are the IDEs, are they okay, and do you still have them?

Two 120 GBs and one 250 GB. They're okay. I mainly just pulled them out to clean things up and simplify my life a little bit. In hindsight, I guess I could have kept one and used that for the OS and the entire new drive for data, but whatever.

quote:

I use this with smbmount mounting a share on an old FAT fs, but the options may still be valid:

-o dmask=770,fmask=660,gid=wheel,password=$your_pwd,username=$your_uid

For total freedom you can set the dmask to 777, the fmask to 666, the gid to users.

I barely understand this, but I think you're suggesting I mount my own SMB share on the server? That, or I really don't understand what you said at all.

quote:

You can make it available as long as the permissions on that drive/partition make it available to who you want (don't know if the initial perms used when you mounted the smb share will carry over) and place a symbolic link in the original Samba share's directory

Similar to above, I think I kinda sorta understand this. I think the most difficult thing about coming into Linux/UNIX from a Windows background is understanding the file system, and that's certainly my weakest area right now. It sounds like I've got some reading up to do. I do appreciate the help!

# ? Jun 26, 2011 18:10

Thermopyle: Jul 1, 2003; ...the stupid are cocksure while the intelligent are full of doubt. �Bertrand Russell

Add your user to the nobody group. As long as the group has the permissions you want, you're golden.

# ? Jun 26, 2011 18:38

Megaman: May 8, 2004; I didn't read the thread BUT...

Lonely Wolf posted:

haha who uses compiz with dwm? You configure dwm by applying patches, editing config.h, sometime writing a little code, and recompiling. Most stuff you can do just from editing config.h and recompiling though.

Where is config.h? I did a find from / and no such a file exists. Do I have to make it? And if so where?

# ? Jun 26, 2011 18:57

brc64: Mar 21, 2008; I wear my sunglasses at night.

Thermopyle posted:

Add your user to the nobody group. As long as the group has the permissions you want, you're golden.

I'm now a member of nogroup, but I'm still getting permission denied without using sudo. Here are the permissions for the folder I'm trying to copy into:

code:

drwxr-xr-x 2 nobody nogroup 4096 2011-06-26 12:38 Movies

Group and Other lack write permissions. But hmm, now that I'm a member of nogroup, why isn't 775 sufficient? I only get rid of the permission denied error with 777... I mean, I guess it doesn't really matter... I'm not at all worried about security for my purposes, but I am interesting in learning.

# ? Jun 26, 2011 18:57

Lysidas: Jul 26, 2002; John Diefenbaker is a madman who thinks he's John Diefenbaker.; Pillbug

How much data is on the 1TB drive? Do you have enough scratch space to copy everything off of it, wipe it and reformat it? If so, you should do that.

After copying everything off of the 1TB disk, unmount its filesystem and modify its partition table (assuming that the disk is /dev/sdb):

code:

fdisk /dev/sdb
t
8e
w

After marking the partition as type 8e (Linux LVM), set up a physical volume, volume group and logical volume ("storage_vg" and "some_name" are arbitrary; replace with whatever you'd like):

code:

pvcreate /dev/sdb1
vgcreate storage_vg /dev/sdb1
lvcreate -n some_name -l 100%FREE storage_vg
mkfs.ext4 /dev/storage_vg/some_name
mount /dev/storage_vg/some_name /wherever/it/was/mounted/before

Once you get a new disk (let's say /dev/sdc), you can do this:

code:

fdisk /dev/sdc
n
p
1


t
8e
w
pvcreate /dev/sdc1
umount /wherever/it/was/mounted/before
vgextend storage_vg /dev/sdc1
lvextend -l 100%FREE /dev/storage_vg/some_name
resize2fs /dev/storage_vg/some_name
mount /dev/storage_vg/some_name /wherever/it/was/mounted/before

This should add the physical volume on the new disk to the volume group, then extend the logical volume with the extra space. After that, resize the filesystem to take up the newly-available space on the logical volume. This is not RAID; this is (I think) the equivalent of Windows' dynamic disks. The two devices are concatenated and appear as one big disk. Obviously if one disk dies, you lose half of your data (and recovering the other half might be tricky).

I haven't tested these instructions and typed some of it from memory (and man pages), but this should be pretty painless.

# ? Jun 26, 2011 19:07

Underflow: Apr 4, 2008; EGOMET MIHI IGNOSCO

brc64 posted:

Two 120 GBs and one 250 GB. They're okay. I mainly just pulled them out to clean things up and simplify my life a little bit. In hindsight, I guess I could have kept one and used that for the OS and the entire new drive for data, but whatever.

Yes, you can easily use those for extra space. almost 500Gb isn't bad. To keep things simple, you could mount them under /120a, /120b, /250.

brc64 posted:

I barely understand this, but I think you're suggesting I mount my own SMB share on the server? That, or I really don't understand what you said at all.

You're on a Linux machine trying to access a Windows share, right? The mount command for that is:

mount -t cifs //NameOfTheWindowsMachine/NameOfTheShare /EmptyDirectoryToMountTheShareIn

Now see if the same command will work with the options I mentioned:

mount -t cifs //NameOfTheWindowsMachine/NameOfTheShare /EmptyDirectoryToMountTheShareIn -o dmask=777,fmask=666,gid=users,password=PasswordOnTheWindowsShare,username=YourUsernameOnTheWindowsMachine

Oh and if you want to go the nogroup route: I think I vaguely remember something about group order being important, so maybe it would work if you have 'nogroup' listed before 'users'. Could be nonsense, but worth a try.

e: NB: there's a space between the '-o' and the options that follow

And...

brc64 posted:

I'm now a member of nogroup, but I'm still getting permission denied without using sudo. Here are the permissions for the folder I'm trying to copy into:
code:
drwxr-xr-x 2 nobody nogroup 4096 2011-06-26 12:38 Movies

That's not a 775 dir, that's a 755

Try it after you type this as su: 'chmod 775 /PathOfYourDirectory'.

Underflow fucked around with this message at 20:07 on Jun 26, 2011

# ? Jun 26, 2011 20:02

Notorious b.s.d.: Jan 25, 2003; by Reene

brc64 posted:

I'm now a member of nogroup, but I'm still getting permission denied without using sudo. Here are the permissions for the folder I'm trying to copy into:
code:
drwxr-xr-x 2 nobody nogroup 4096 2011-06-26 12:38 Movies
Group and Other lack write permissions. But hmm, now that I'm a member of nogroup, why isn't 775 sufficient? I only get rid of the permission denied error with 777... I mean, I guess it doesn't really matter... I'm not at all worried about security for my purposes, but I am interesting in learning.

Changes in group membership will not apply to any logged in sessions. Log out and log back in to become a member of the group.

Re: the disks, use one of the old crappy IDE disks for the boot/root/swap filesystems. Root doesn't need to be big or fast. 8GB will do.

Use the entirety of the 1 TB drive for your Samba shares. Be sure to use LVM, in order to ease later migration to software RAID. (It's probably best to use a separate LVM VG, too. You don't HAVE to, but it will be less of a pain in your rear end.)

Honestly, what the hell kind of data can you put on a 1TB drive that isn't worth $80 to protect? Restoring 1TB of data from backup is a huge pain in the rear end. It's hard for me to believe that your time is worth less than the $80 for a second drive.

# ? Jun 26, 2011 20:07

brc64: Mar 21, 2008; I wear my sunglasses at night.

Underflow posted:

You're on a Linux machine trying to access a Windows share, right?

Sort of. The share is on the Linux machine (using samba), but that's moot at this point.

quote:

That's not a 775 dir, that's a 755 Try it after you type this as su: 'chmod 775 /PathOfYourDirectory'.

No, I realized it was 755 after I posted it, so I tried changing it to 775 in the hopes that it would work, but it didn't work until I made it 777... however...

Empty Threats posted:

Changes in group membership will not apply to any logged in sessions. Log out and log back in to become a member of the group.

...I did not realize this.

quote:

Use the entirety of the 1 TB drive for your Samba shares. Be sure to use LVM, in order to ease later migration to software RAID. (It's probably best to use a separate LVM VG, too. You don't HAVE to, but it will be less of a pain in your rear end.)

For what it's worth, when installing the server to begin with, I believe I picked an LVM option for the 1 TB disk. I didn't really know what it meant at the time, but it said something about making it easier to extend the volumes or something like that, so it seemed like the way to go.

quote:

Honestly, what the hell kind of data can you put on a 1TB drive that isn't worth $80 to protect? Restoring 1TB of data from backup is a huge pain in the rear end. It's hard for me to believe that your time is worth less than the $80 for a second drive.

It's all just media from our different PCs, plus our DVD collection I've been ripping, all for streaming to XBMC. It's nothing mission critical or anything that would give me a heart attack in the event of sudden catastrophic harddrive failure. I do intend to add redundancy, but I could only afford to start small and I wanted to get the project moving.

# ? Jun 26, 2011 20:33

Adbot: ADBOT LOVES YOU

# ? May 21, 2024 01:35

Notorious b.s.d.: Jan 25, 2003; by Reene

brc64 posted:

It's all just media from our different PCs, plus our DVD collection I've been ripping, all for streaming to XBMC. It's nothing mission critical or anything that would give me a heart attack in the event of sudden catastrophic harddrive failure. I do intend to add redundancy, but I could only afford to start small and I wanted to get the project moving.

I know the feeling. I ripped ~300 DVDs for the same purpose, with the same application. It took me about 5 minutes of work per DVD. (plus quite a few minutes of waiting)

You can fit ~100 complete DVDs (or ~250 DVD features) inside a "1TB" drive. At five minutes a pop, that's between 10 and 20 hours of work. A second 1TB HDD is about $80.

Is your leisure time really worth less than four bucks an hour?

Notorious b.s.d. fucked around with this message at 21:46 on Jun 26, 2011

# ? Jun 26, 2011 21:38

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > The Linux Questions Thread: a bunch of pitfalls, but technically it's possible

«‹›1031 »