Welcome to the Linux Foundation Forum!

Apache installation

Hi,

ThanX for tryn'a help.

I've installed apache package on my Ubuntu 9.04, and website directory tree is /var/www/. I don't have write permission with my normal user, I mean the user I created during setup. I always need to use sudo to copy my files, but its extremely foolish, because most of the times I need to keep my file open, for example in Text Editor, and update the page frequently to preview the changes.

How can I gain write access to this folder using this user of mine?

(To tell ya I'm a real beginner. I was a Windows user for more than 7 years, its less than 2 months I've been using Linux).

ThanX for you helps.

Comments

  • mfillpot
    mfillpot Posts: 2,177
    Keeping it at limited access is best for security reasons. I do not know the default apache right in an ubuntu installation, but if you see that a group has write rights to the directory then you can just assign your user to that group. My method has also been to launch a file editor with root rights to change the specific files (example "sudo gedit /var/www/index.html").
  • gomer
    gomer Posts: 158
    the files are owned by www-data. So ... you hace a couple of options ...

    you could just sudo -s and do everyting as root (creepy, you shouldn't do this).

    You can make yourself a member of the www-data group (se /etc/group) and then make sure that group write permission is turned on for all dirs and files under /var/www.

    you can su to www-data (sudo su - www-data) and work as the www-data user.

    you can change the ownership on the /var/www directory to your user account (you may have to tweak apache's config to allow your user to execute scipts, etc, e.g. make sure it's not using suexec module).

    and I'm sure there are other options that I didn't think of.
  • ActiveX
    ActiveX Posts: 59
    Thanks so much guys, you really helped.

    But now there's another problem I've been facing: Apache configuration!

    Does anyone know how can I CHANGE the www directory path? I have another directory, which is in my NTFS drive in windows, where I used to work on my web projects, and now I wanna change the www directory path, to /media/archive/htdocs/ .

    How can I configure this? I used to change DocumentRoot option in httpd.conf in windows, but I can't see even a character /etc/apache2/httpd.conf !

    How and where should I change the path?
  • ActiveX
    ActiveX Posts: 59
    I found a part of problem: I compiled and installed apache manually (as opposed to installation using Synaptic) and everything is ok, but there's stil one problem: the apache package that I had installed using synaptic is not completely uninstalled and system paths have not changed, I mean instead of my own compiled package, those installed by synaptic are being executed and I can't start my apache service.

    My terminal:
    activex@activex-desktop:/usr/local/apache2/bin$ ./apachectl start
    (13)Permission denied: make_sock: could not bind to address 127.0.0.1:80
    no listening sockets available, shutting down
    Unable to open logs
    activex@activex-desktop:/usr/local/apache2/bin$ 
    

    I think now you should know what's the problem. Any idea?
  • bubray
    bubray Posts: 6
    Hi,

    I think there is a old process running on port :80

    check whether it is running using
    netstat -plan|grep :80

    kill the process using kill -9 process id

    then use apachectl to start your new installations.
  • ActiveX
    ActiveX Posts: 59
    nope! there wasn't such a process. Even if it existed, I think killing the process would not be the solution. I need to process local web files with my own apache installation permanently, I mean I want my system to be as if I had installed apache by myself from beginning.

    My current problem is that the previous apache package is not completely removed and is not letting my new configuration to be used. System is still using the old httpd.conf.

    I need to remove the previous package completely befire doing anything else, but I don't know how! I don't know where the files are, because I've been a windows user for more than 7 years and it's less than a month that I've switched to linux, so I don't know where do the files go, in / and it's sub directories.

    Now all I need to know is: How to remove apache package from my system manually? (Synaptic didn't do this correctly!!!)
  • ActiveX
    ActiveX Posts: 59
    nope! there wasn't such a process. Even if it existed, I think killing the process would not be the solution. I need to process local web files with my own apache installation permanently, I mean I want my system to be as if I had installed apache by myself from beginning.

    My current problem is that the previous apache package is not completely removed and is not letting my new configuration to be used. System is still using the old httpd.conf.

    I need to remove the previous package completely befire doing anything else, but I don't know how! I don't know where the files are, because I've been a windows user for more than 7 years and it's less than a month that I've switched to linux, so I don't know where do the files go, in / and it's sub directories.

    Now all I need to know is: How to remove apache package from my system manually? (Synaptic didn't do this correctly!!!)
  • mfillpot
    mfillpot Posts: 2,177
    If the package was installed via an ubuntu package, then you can go to the synaptic package manager for the package, right click, go to mark for complete removal. If it does not give that option because it is not installed then install if first so you get the option to do a complete removal.
  • ActiveX
    ActiveX Posts: 59
    I've done so. After uninstallation using Synaptic most of the files, BUT NOT ALL, are deleted. Some of files, like apachectl and httpd.conf still exist.

    I've faced a situation as if I had installed a software by myself and had corrupted it's files manually.

    What should I do to remove those remaining files? What would I have to do if I had compiled and installed a generic linux kernel by myself, not a distro?
  • mfillpot
    mfillpot Posts: 2,177
    That is a strange finding but it does not entirely surprise me. It may have left httpd.conf to allow you to keep the original configuration file in case you need to reinstall the app.

    I have tried the uninstallation in Slackware and found that apachectl has been removed but httpd.conf has not been removed because it is meant to be backup in case you are just reinstalling the app due to other issues. I am assuming that the logic on ubuntu is the same, but I have not yet attempted the operation on ubuntu to test it.

    If you are using a custom Linux distribution without a package manager then tracking of file relationships can be difficult unless you manually setup a function, some apps have a script that tracks installed files and has it's own uninstall script.
  • ActiveX
    ActiveX Posts: 59
    From your post I understand that I must know about Linux directory structure and where every file is put so I can easily track these changes.

    Aynway,I think the best solution for me at this time is reinstall ubuntu. Am I right? so I can have an intact directory structure and know where every file is gonna be put.
  • mfillpot
    mfillpot Posts: 2,177
    I agree with your recommendation. You do not want a system with unknown files floating around, since it is no longer is a state that are comfortable with re-installation can be beneficial.

    I would like to ask what version of ubuntu are you using? Are you using the Desktop version or Server version? For you server operations have you thought about going with a more simplified distro?
  • ActiveX
    ActiveX Posts: 59
    That's what exactly I mean, preventing useless files to exist.

    I use Ubuntu just because I'm new to Linux and I need a GUI. I've decided to install Linux as a generic kernel compilation, later.

    I use Ubuntu 9.04 Desktop Edition. I'm a Web developer, and as in Windows, I need apache to be installed here in Linux to host my local files.

    By the way, is there any way I can keep packages of softwares downloaded by Synaptic so I won't need to download them again, for installation on my fresh Ubuntu installation?

    Since its a simple desktop web server, I don't need any simpler distros, and as I said, I'll compile and use the kernel later, when I get enough professional in Linux! lol!
  • mfillpot
    mfillpot Posts: 2,177
    I found a thread talking about archiving downloaded packages you should check it out. http://ubuntuforums.org/archive/index.php/t-909744.html
  • ActiveX
    ActiveX Posts: 59
    Thanks so much. Thank you all of guys here who helped me.
    I almost don't have any other question in this context.

    Thanks so much again.

Categories

Upcoming Training