General :: Compiling And Installing Programs As Non-root In Home-dir ?
May 31, 2011
I'm trying to compile programs under Solaris in my Home directory.
Ive had experience with simple installs which work with just
But with more involved programs, with dependencies on special libraries, how do I have the system look into my home directory for these? and how would I compile them into my home dir?
and if i didnt customize partitions when i installsed and just used the whole drive - they're both together in the same partition right, the home just goes into the root? but it still creates a 2-3 gig swapfile?
do you recommend creating separate partitions for the home and root folder during install? is there any benefit besides being able to reinstall ubuntu without losing your data?
I am new to C programming and I need a way to compile and run my code. I've made a little research and I found that eclipse is a nice IDE. But I don't know how to configure it and generally. What do I need to download and install to make it work. I've also tried downloading easy eclipse that doesn't need any configuration or extra files but the link on the site was broken. Is there any other C IDE like easy eclipse available?
I am trying to study gtk programming and get stucked before some errors in compiling them. Iam using ubuntu 9.10 and i didnt installed any additional libraries or packages for running these gtk programs assuming gtk is already installed in ubuntu. I am newbie to gtk I tried compiling some example programs from my tutorial and it ended up in errors. So do I need any additional packages to run these programs.when compiling it shows package config cannot find any gtk+-2.0 packages and to try specifying .pc in environment path. I don't know how to set these path stuff.
I am new to Slackware and I am wondering how you set the ARCH variable. On the multilib wiki page it says:
"Rather than editing the script and change/remove CFLAGS definitions, you can set the ARCH to 'x86_64' and save yourself some time."
So from this I understand that they advise against editing the Slackbuild script. How do I set the ARCH variable on the command line or is it a configuration file I need to edit? I am running Slackware 13.1 64.
I am suppose to explain the dependencies that exist and each of the following lines of makefile. Lists orders to be executed as a result of running the make utility on it.
I'm trying to get my head around compiling programs, and need some things clarified. Can anyone confirm or deny my understanding so far:
1) Program is written in high level language, such as C. File(s) of the program are the 'source code'. 2) Program is then 'compiled' to turn it into low-level code that is both hardware - i.e. Intel/AMD/ARM processor specific, and also Linux distribution specific. This resulting files are called the binaries. 3) I can go from source code to installed and usable program by doing these steps: ./configure make make install
I am trying to write modular program but when I try to compile I come to know I don't have any files in /usr/src/linux. I have installed fedora 12 which has 2.6.31.5-127.fc12.i686.PAE. I am including in my code:
linux/kernel.h linux/module.h
Those are not present directory in /usr/src/linux is not present. My goal is to compile modular programs.
I would like to run Blender (3D modelling software) on RedHat Enterprise. I have a problem with missing libraries (libavutil50.so). Second problem is that I have no root access so I cannot install rpm package with libavutil50.so. Does anybody knows how can I manage this? Is it possible to install somehow libavutil50.so in my /home folder, that Blender will have an access to this library ?
I have downloaded the newest most stable Linux kernel, 2.6.33.2. I thought I would test this using VirtualBox. So I create a dynamically sized hard disk of 4 GB. And installed CentOS 5.3 with just the minimum packages, just the development libs and development tools. I unzipped in the /usr/src directory. I setup the make menuconfig with just the default settings.
After that I ran make and got the following error: Code: net/bluetooth/hci_sysfs.o: final close failed: No space left on device make[2]: *** [net/bluetooth/hci_sysfs.o] Error 1 make[1]: *** [net/bluetooth] Error 2 make: *** [net] Error 2
The amount of space I have left is after it give that error is: Code: # df -h Filesystem Size Used Avail Use% Mounted on /dev/mapper/VolGroup00-LogVol00 3.3G 3.3G 0 100% / /dev/hda1 99M 12M 82M 13% /boot tmpfs 125M 0 125M 0% /dev/shm
So my virtual harddisk size is 4gb and the actual size if 3.5gb. So what happened to the other 500mb? How much size should I give when compiling and installing a Linux kernel? Are there any guidelines to follow when doing this? This is my first time, so just experimenting with this.
I have downloaded the newest most stable Linux kernel, 2.6.33.2.
I thought I would test this using VirtualBox. So I create a dynamically sized harddisk of 4 GB. And installed CentOS 5.3 with just the minimum packages.
I setup the make menuconfig with just the default settings.
After that I ran make and got the following error:
The amount of space I have left is:
My virtual size is 4 GB, but the actual size is 3.5 GB.
How much size should I give when compiling and installing a Linux kernel? Are there any guidelines to follow when doing this? This is my first time, so just experimenting with this.
I am compiling kernel to learn it and as well as for edubook, a netbook I am working on. I wanted to put the compilation under my home directory and followed the syntax suggested in the README in /usr/src/linux,like,
Quote:
'make o=/home/pbhat/temp/kcompile menuconfig'
Like this I have gone through full compilation successfully,but see no output going to the specified folder.Could anybody point out where I went wrong? I was compiling under /usr/src/linux and in Opensuse 11.2.I followed the README apparently written by Linus where I saw the above make syntax.Maybe that is old and hold good no more.
I am trying to install google chrome on my computer, it is in a .deb package and I am using dreamlinux. Earlier today I installed a .cbr/.cbz file reader from root, that was also a .deb and it worked just fine. now I go to install google chrome and it says (Blue is my command, and red is the system response):
I checked under my USER GROUPS and Root is still set as root, and I haven't been using it unless I need to install with it to other directories, Should I be using another command to install? or is it a problem that may affect other aspects of my system? ... My biggest concern is not getting google in, I can wait for that ... I just wanna know why ROOT is under the impression it's not a superuser.
I have a machine which has only /opt with some decent amount of space where I can install a software. /opt belongs to root:root. The software I want to install cannot be installed as root user.
So lets say I create a directory called /opt/install1 and then chown -R install1 to belong to user1. And now I install the software under /opt/install1 with user as user1.
Is this a best practice violation? There could potentially be just /opt/install1 belong to user1 and in future everything else created under /opt belonging to root..
so i've just been sitting here, compiling and running some c++ programs (with g++ from the terminal), I looked away for one minute to open a file that had just been written out (opened it with gedit if that makes a difference), now my terminal says:kbuildsyscoca4 running...so I'm wondering if anyone knows what this is about?hy it suddenly started running and why it has appeared on my terminal. I'm basically just curious as I though bash shells were entirely separate virtualizations of shells, and unrepresentative of other processes. i'm also wondering what kbuildsyscoca4 is/does (i've got gnome going on and generally gnome varieties of programs). I found a post from about a year ago mentioning the same thing, though no replies.
I am doing my project which detects and displays the error message along line number when compiling c programs under Linux. Using gcc command..cc command which we generally use. For which my program must act as a compiler and generate error messages when user program is written and compiled. My idea was to develop a shell script program using name suppose ncc that would be..
$ vi ncc.sh echo "compiling program $1" $cc MyPgm.c $.a/.out <esc> :wq Now user program pgm1.c can be compiled using $ sh ncc.sh pgm1.c
This will invoke my compiler program. But the thing is how to pass user program pgm1.c as input to my compiler program...
Code: [kuba@Kuba-fedora uget-1.8.0]$ ./configure checking for a BSD-compatible install... /usr/bin/install -c checking whether build environment is sane... yes checking for a thread-safe mkdir -p... /bin/mkdir -p checking for gawk... gawk checking whether make sets $(MAKE)... yes checking for style of include used by make... GNU checking for gcc... gcc checking whether the C compiler works... yes checking for C compiler default output file name... a.out checking for suffix of executable .....
Configure: error: Package requirements (libnotify) were not met: No package 'libnotify' found Consider adjusting the PKG_CONFIG_PATH environment variable if you installed software in a non-standard prefix. Alternatively, you may set the environment variables LIBNOTIFY_CFLAGS and LIBNOTIFY_LIBS to avoid the need to call pkg-config. See the pkg-config man page for more details. But I have the newest version of libnotify.
I tried to compile C program that uses Openssl libraries on shell but got this error. I guess libraries are not linked properly. undefined reference to SSL_library_init()
[URL] I'm currently running simplyMEPIS 8.0.15 (rel 32). It's booting automatically off a USB. The way I have it configured right now, it just loads right up without a root menu or boot-options, just basically going straight to "hey, u ROOT or DEMO?" I'm not sure if there's a way to get back to the bash and load directly to RAM at this point.
In another thread I was looking into burning backups from k3b as I run live because my DVD-player isn't behaving right now under windows... 'still working on that I want to get a backup done before I do a hard install. It's my goal to get either Kismet, wireshark, or Airsnort/crack/peek running on this machine so I can start gathering packets and doing analysis. Is that a way that (atleast for the timebeing until I can work out backups) that I can install any of these programs directly to the USB (by editing the ISO somehow?), so that they'll be available when I liveboot?
l my root (/) partition has 11G free space and my /home is only left with 5g around and /usr has around 8g in my fedora 13 .So is there any possibility to "resize" the root partition and add it to home partition bcoz i see the opposite in the threads(resize home to add space to root).My home has nothin more than a movie which is 700MB and i've installed some new application yesterday. But it shows half of the space is almost used!!!
I am operating Debian 5.0 kernel 2.26. In trying to clean up (I believe I was using a root terminal), I erased the directory /home/root. Since I have done that, I can not get to a root terminal, can not use synapse and a host of other administrative processed.
Now, when I boot the system, I can log in as a user (my version of Debian does not allow me to login as the administrator). Then, logged in as the user, I previously could select for example 'synapse', I would be given a screen to enter the Administrator password and then get synapse. Now, I am given the screen to enter the Administrator password, but once I enter it, I do not get synapse. I do get an error message 'can not find /home/root/.synapse' And of course, I erased the /home/root directory. Now I might just mkdir /home/root - except to do that I need a root terminal and I can not get that either - probably for the same reason.
I have tried using the boot disk and going to the rescue mode - except for some reason I can not mount a root directory. I believe it is because I am using LVM2 and my root directory must be in an LVM volume, but I can not figure out how to get to that using the rescue mode.
i tried to run lame install. on debian lenny amd64.
sudo sh /home/levi/Desktop/lame-3.98.4/install.sh
it returned errors, and i gave up on that. but it has placed a bunch of files and directories in my home folder(not in Desktop)which are owned by root. how may i delete them?
folders are ACM debian Dll doc
[Code].....
mv: cannot move ..... to ... No such file or directory. which i at least FEEL is certainly not the case.
My total filesystem capacity:39.9 GB(used 4.2GB,available:35.7 GB) Currently,i have only single partition. i wanna make again a new partition from the single existing partition where root(/) folder stored.
my aim is to separate the home folder from the existing partition to the new partition.
Or would this sacrifice security in some way? I've been using root only, and am ready to have a seperate account now. It's the dotfiles for GUI apps that I'm concerned about:
Code: -rw------- 1 root root 98 Feb 13 16:23 .Xauthority -rw------- 1 root root 6392 Feb 12 18:13 .bash_history drwx------ 5 root root 4096 Jan 13 17:47 .config drwxr-xr-x 4 root root 4096 Dec 29 21:36 .fvwm drwx------ 4 root root 4096 Nov 7 19:55 .mozilla -rw------- 1 root root 218 Jan 26 10:04 .recently-used.xbel -rw------- 1 root root 98 Feb 13 16:23 .serverauth.17096 drwxr-xr-x 2 root root 4096 Dec 25 12:42 .tuxcmd drwxr-xr-x 2 root root 4096 Feb 12 17:25 .xine
In regards to other distribution cds: I'd like to learn how to install a 'root' and 'grub' boot loader so that I can cleanly install other distributions. I'm new to the file system. Could I make a bootable usb with simple installers? or command line codes?
I'm trying to compile a "faster", more optimized kernel. It compiles ok, but can't mount the root filesystem.Here is my kernel configuration for the 2.6.33.4 kernel. My boot partition is ext4 on a two-disk FakeRAID array totaling 1TB.When trying to boot in recovery mode, it says I could mount /dev/sdaX, etc. but can't recognize my dmraid array. BTW,
I have 84GB free space on this hard drive and want to install another distro. Will I be able to create another / and /home partitions for the new distro?