I recently noticed a fairly strange and for me unexpected behaviour in Xubuntu 12.04 and 14.04.
I was doing the following:
Testing if my user in in the group users, with
groups $USER
This is not the case by default. So I add my user to this group:
sudo usermod -a -G users $USER
I the can check the file /etc/group and will see my user added in the entry.
I then would like to give the group users access to some files, in my example the www and cgi-bin directory:
sudo chgrp users /var/www /usr/lib/cgi-bin
I also want that my group can write into the directories:
sudo chmod g+w /var/www /usr/lib/cgi-bin
I would assume, that I can now create a file in those directories, but I can't. Neither by commandline, nor by the standard filebrowser from Xubuntu.
Somewhere I read, that I need to logout from the terminal to make it work, so I close and reopen the commandline terminal, but it is still now working.
But: It I reboot the whole system everything works as it should...
Seriously??? Why is this, is it a bug or a feature and are there better ways then restarting the complete OS?
(I thought the strength of Linux is exactly that you don't need to reboot all the time like in other "popular" OS)
(Note: I have not tested this on other systems as e.g. Debian yet...)
Group memberships are inherited from process to process like many other things in a unixoid environment. That means a running shell will not be affected from such changes in the account configuration. Also just opening a new terminal or shell will not show the change, since it is spawned from an already running process, ultimately from the initial process started right after the login.
You have to re-run the login process instead. Either by restarting the graphical environment, or by doing a logout/login sequence when working on the virtual terminals. Also obviously rebooting will lead to a new login process.
The only direct alternative is to spawn a new login shell explicitly: bash -l for example does the trick: it re-executes all stuff run through at login time. But note that this only effects that started shell and processes spawned from it. It does not affect other already running processes. So you have a somewhat mixed environment then...
Related
I'm currently doing a project for my school aimed at creating a server in a VM.
One of the requirements of the subject was to install sudo on an emulated Debian Linux, with the following field in the sudoers config file:
Defaults requiretty
The subject line requiring me to set this:
The TTY mode has to be enabled for security reasons.
I know that tty is a command which prints the filename of a terminal that is currently connected to the standard input. However, what I can't see is how this ensures more security for the server.
When requiretty is set, sudo must be run from a logged-in terminal session (a tty). This prevents sudo from being used from daemons or other detached processes like cronjobs or webserver plugins. It also means you can't run it directly from an ssh call without setting up a terminal session.
This can prevent certain kinds of escalation attacks. For example, if I have a way to modify the crontab for a user who has NOPASSWD sudo permissions, I could use that to kick off a job as root. With requiretty, I can't do that...
...easily. This restriction is not particularly hard to circumvent, and so generally isn't all that useful compared to the valid use cases it breaks. Red Hat used to use it, but removed it a few years ago.
Is there any way to get vs code to work properly in linux? I can't run sudo code . because that gives me an error saying it's not secure to do so, I can't do anything within the editor to force doing things, like staging a file in git, or reloading a newly installed extension. I've googled around, and it seems nobody else has posted about this, and it seems highly unlikely that I'm the first to raise issue about this. (Take it easy on me, I'm a relatively new linux user). I'm trying to figure this out on Ubuntu 18.04 if that's relevant at all. My version of vs code is 1.30.2
I guess my main question is what's the right way to get applications like vs code to be able to perform tasks that required doing things without fighting the OS about sudo and privileges?
Launch via sudo from terminal
To launch VSCode as root --which is highly discouraged-- you must specify an alternate user data directory as follows:
$ sudo code --user-data-dir /path/to/alternate/folder
VSCode will automatically generate the required folders in the selected directory and launch with root privileges.
Change permissions to fix "permission denied" error
The solution in this case is to manually change the permissions of the two directories /home/$USER/.config/Code/ and /home/$USER/.vscode/. Perform these steps:
$ sudo chmod 755 /home/$USER/.config/Code
$ sudo chmod 755 /home/$USER/.vscode
To answer your other question:
If you really need to run several commands as root and you are annoyed by having to enter your password several times (when sudo has expired), just do sudo -i and you'll become root.
If you want to run commands using pipes, use sudo sh -c "comand1 | command2".
You may also want to take a look at this Ask Ubuntu answer about running applications as root.
I solve this problem using:
sudo chown -R YOUR_USER YOUR_PROJECT/
You basically need to tell the OS that you are the owner of the files you create. Use sudo chown <user name> <projects directory>
However, if you already created some files before applying chown, don't forget to change their permission also sudo chown <user name> <projects directory>/<file name>.
While creating an rpm spec file I have created a new user and group in the %pre section. This new user does not however have permission to login from from shell for security purposes. Now when I install the rpm this new user is successfully created. However, I wish to start the installed rpm service with the newly created user. Currently I simply write; 'filePath/file.exe file.cfg' to execute the file.exe with its configuration file i.e. file.cfg in my 'init.d' file to start the service. How can I modify this command to start the same service but with the user that I created while installing the rpm? Basically I want to execute the program in my init.d file but through a different user, like I would have done with sudo if my required user was the super user. Any feedback will be highly appreciated.
Your initial starting point both for installing the rpm and for running the service is privileged. For instance, on my CentOS 6 machine, I see in /etc/passwd
games:x:12:100:games:/usr/games:/sbin/nologin
but running as root, I can do this:
$ sudo -u games /bin/sh
sh-4.1$ echo $PATH
/sbin:/bin:/usr/sbin:/usr/bin
sh-4.1$ id
uid=12(games) gid=100(users) groups=100(users) context=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023
sh-4.1$ cd
sh-4.1$ pwd
/usr/games
In your service script, you can use sudo to run a given process as another user (though a quick check of the same machine does not show this being done).
#msuchy points out that runuser may be preferable. I see that this is relatively recent (according to Ubuntu runuser command?, appeared in util-linux 2.23 -- lacking a date makes release notes less than useful...). The oblique comments in its documentation about PAM make it sound as if this circumvents some of the security checks. Perhaps someone has a better comment about that.
I've got an SVN instance installed on a free EC2 AWS server. In short: I'm using LAMP.
Using what I read in this article and encountered the "you need a TTY" error as mentioned in the comments. I followed the second resource and it cleared the error message, but doesn't seem to be executing the script. When I manually run the script, however, it works.
Any clue what I'm missing?
When I followed the second resource to fix the TTY error I changed the contents of my /svn/repository/hooks/post-commit script from:
#!/bin/bash
sudo /usr/local/bin/svn-post-commit-update-mysite 1>&2
to:
#!/bin/bash
su –session-command=”/usr/local/bin/svn-post-commit-update-mysite 1>&2″ dynamic &
First possible issue:
You cannot rely on the value of the $PATH variable inside the hook. This means you need to specify complete paths for all executables.
In particular, "su" is a program located in "/bin/sh" in most distributions. To be sure, type
type su
Next possible issue:
Is your subversion server being run as root? su will try to ask for password if run by other users, and will fail if it's not being run interactively - even if the user is in the sudoers file!
If you are using Apache+DAV, this means the apache service must be run as root for this to work (instead of www-data), which is a serious security problem.
You probably don't need to use su or sudo at all if all of the files are owned by the same user (www-data, for instance). You can change the ownership of the site files with something like
sudo chown -R www-data:www-data /var/www/<my-project>
And then remove the sudo and su from both the hook and the svn-post-commit-update-mysite file.
My best guess would be that something in your script depends on the PATH environment variable. Subversion runs hooks in an empty environment for security reasons. So you need to either setup the environment in your shell script or use absolute paths.
You might want to read the Subversion book entry on implementing hook scripts. The particular issue I mentioned is explained in the information block.
There is a standard way (working across Linux distributions) to launch a process (from another application) asking for the root password in order to elevate privileges?
I tried to use gksudo (it is installed in ubuntu by default), but under other distributions (or under other desktop manager) it may not be installed.
I would recommend looking at PolicyKit which is what most modern distros are using to accomplish this.
That works everywhere but does not cache the password and asks for the root and not the user password (as sudo does):
su - -c command
EDIT: Not on ubuntu where the root-account is disabled. Probably you need something like that:
test -x /usr/bin/sudo && sudo command || su - -c command
The only default thing is text mode su. Most distros have also sudo installed.
Now, in KDE based distros you'll have kdesu, while in GNOME based it'll be gksu and gksudo. Machines in Kerberized domains have ksu.
You might try to use /etc/sysconfig/desktop to see which is the default desktop.
Traditionally, if your application needs to allow a user to elevate privileges, it installs its own single-purpose setuid executable -- single-purpose meaning that it performs the task needed, instead of acting as a general-purpose launcher.
$ su -
# cp `type -p id` /usr/local/bin/root-id
# chown root:users /usr/local/bin/root-id
# chmod 4750 /usr/local/bin/root-id
$ /usr/local/bin/root-id
... euid=0(root) ...
OTOH setuid executables have also been a common source of security holes too, so exercise care.