How can I get rid of all policykit restrictions that fedora 13 has? I just upgraded from fedora 10 and of course my freenx sessions are again unable to do anything useful like mounting a drive. Difference is no GUI now to help fix this. So I would like to get rid of all restrictions.
I have searched for days on Google and can't find a clear answer to my question. I have a NT4 PDC which I am migrating to Samba 3 (Version 3.4.2-47.fc12) on FC12 with kernel(2.6.31.5-127.fc12.i686). I am using tdbsam as my passdb backend.I setup Samba as a BDC and then joined to NT4 Domain succesfully. When I go to vampire the accounts I get lots of errors and some user accounts get transfered over. It turns that all the user accounts that transfer are those that don't have a capital letter in their username on the NT4 domain server. Most do and don't get transfered. There seems to be errors with my groups and Computer accounts.Is there a way to change the requirements in Fedora 12 for username, groups and computernames?
I have a work network of about 20 boxes most of which are running Windows 7 and one of them is a file server using linux and another is Windows server 2003. Now the local IP is distributed by the router, and no regulation of internet access is done by any of the servers.What I need to do is restrict internet access to select domains, which would probably need DHCP through linux(I think, not really sure), and I need something simple like a 'blabla.conf' file with the allowed websites that I can edit. need to know how to regulate IP addresses through the linux box (all details if possible, I never tried to do that before), and how to restrict internet access also through linux.
I am looking for PolicyKit rpm for my RHEL 5.4 64 bit machine. I looked in most of the places but couldn't find one. Can somebody point me to this? Even the source rpm would be fine.
I am trying to make the edit referenced on this page. But obviously it is for an older Ubuntu version because the /etc/PolicyKit/PolicyKit.conf file does not exist on my 10.04 system.
How do I go about making this edit for however 10.04 handles PolicyKit calls now?
It details I should be replacing a line that looks like
I can't seem to find the "Authorizations" GUI that was present in earlier ubuntus for configuring system policies. It used to be in System - Administration - Authorizations. Which package does it come in? What's the console command for it?
I am trying to make the edit referenced on this page. But obviously it is for an older Ubuntu version because the /etc/PolicyKit/PolicyKit.conf file does not exist on my 10.04 system.
How do I go about making this edit for however 10.04 handles PolicyKit calls now?
It details I should be replacing a line that looks like
Code: <match action="org.freedesktop.hal.storage.mount-removable"> With Code: <match action="org.freedesktop.hal.storage.*">
Looks like its some sort of PolicyKit issue. Documentation (and I suspect the system itself) all assume that I have only one user. But my computer is a two-seat, my girlfriend and I are logged in at the same time. Only I am able to mount and write to USB drives. I suspect its because my account is always logged in first.
Dolphin gives the following error: org.freedesktop.Hal.Device.PermissionDeniedByPolicy blah blah org.freedesktop.hal.storage.mount-removable auth_admin_keep_always
So my guess is that this "auth_admin" for some reason only refers to my user. I opened up the "Global Configuration" application (something that typed up when I filtered for 'policy' in the application launcher), and added my user, my girlfriends user and the 'user' group as "System Administrators". I logged out and logged back in on her account and this did nothing.
why can I not allow remote sessions admin rights? why is there a remote exception thrown when Gnome auto mounts my cdrom? Here is a valid use case for auto mounting on a headless server. Server live in external room, admin pc lives in different part of same building. Server needs new software on CD admin walks to server puts cd in server walks to admin pc installs software. Currently you need to mount manually. not a big deal but could be so much easier if I didn't have to do this.who can a moan to in order to get policykit changed.
P.S. I'm a developer willing to put my time where my mouth is and change this.
after a recent update my policykit permissions stopped working, and I can't understand why.An example: mounting a flash drive requires password, hinting on rg.freedesktop.udisks.filesystem-mount. /usr/share/polkit-/actions/org.freedesktop.udisks.policy has the following:
Code: <action id="org.freedesktop.udisks.filesystem-mount"> <description>Mount a device</description>
I try to understand the reasons for restricting DVD and MP3. My conclusion so far is that DVD is restricted due to software patents (and the DMCA). The software decoding DVD is open software, though. What is the situation regarding MP3? It is also restricted. Is it due to the same kind of software patents? As I understand, the MP3 codecs are not free software (like DVD)?
I usually use .htaccess to restrict access to directories. But what if I just wanted to secure a single php file? Is there some sort of code that would allow me to say ONLY THIS IP can access this PHP file?
I have a server with two active network interfaces. On one, I need ssh open for all users (it's running LTSP, and as I learned the hard way today, blocking ssh kills LDM access).
On the other interface (which connects to the rest of the network), I only want to allow a few administrative users to connnect.
Is there a way to do this cleanly using sshd_config or PAM? I don't want to do something hacky like running dropbear.
I have an Apache, PureFTPd, PHP5, and MySQL server setup and running. I'm running several scripts that require folder access of "var/www" in order to accomplish the scripts duty. How do I remove and/or work around the security measure?
i have jsut setup a kvm virtual machine on my server. to connect to the VM from outside of the network i use ssh tunneling. what i would liek to know is if there is any way to create a new user with jsut ssh access. i dont want people to be able to edit files in ~/ or such. jsut need the user to estabilish the connection to the server
I've got Ubuntu server 10.04 set up and I wanted to make a few restrictions. It's pretty much just acting as a VMware server at the moment, and there are some users I've created who I only want to be able to be able to log into the VMware infrastructure web interface. I want to make sure these users can't log in via SSH, FTP, or the console itself. I understand how to block them from logging in via SSH by using DenyUsers, and I added these users to the /etc/ftpusers file to lock them out of FTP, but how can I block them from logging in at the console itself?
I tried locking the user out by editing the /etc/passwd file, but the problem is that by doing this, it also prevents the user from being able to log into the VMware web interface.
The user's entry in /etc/passwd looks like this: bsmith:*:1005:1005:Bob Smith,,,:/home/bsmith:/bin/bash
I have a computer with two interfaces (eth0 and eth1), eth0 is connected with a local network and eth1 is connected to the internet, also it implements a NAT in the interface eth1. Nevertheless, I'm trying to create spoofed packets with sockets raw in the computer that runs the NAT and send the packets to the interface eth1. The problem is that the NAT is changing the IP source to the real one before send the packets. So, anyone have any idea how can I implements the NAT in eth1 but only apply the NAT to the packets that are from/to eth0? I was thinking in something like (I am really newbie with iptables):
While I successfully configured an IPsec-VPN (I use a similar tho modified setup like this:[URL].. I am now stuck on the next steps. While I can connect to everything I want, I need to configure "access-groups" and/or "users".
The scenario is similar to this: Lets say Host A, B and C allow SSH-Connections and some weird non-standard UDP-Connection from Host-VPN, and are also accessible on other ports with public IP's (like http).
I now want to limit, that an admin-user has access to all of them, while trainee-admin only can access everything on Host B and C, and CEO only can connect via telnet to Host C - and all users can be roadwarriors
(I made this example up to give you an idea what i'm trying to do - hope it makes sense). Now my question is, if someone can point me towards a direction, as I'm quite clueless at the current moment as to what to try. I know that commercial IPsec-Implementations can do this, but can OpenSWAN/... give me something similar?
I am using internet web control through squid... All is working fine only some little bit issues.
(1) Sometime when i tried to open google.com or any site I got message (The requested URL could not be retrieved) Screen Shot Attached.) but again after sometimes same websites will open.
url
(2) I would like to block word 'sex'.. So I have edit squid.conf with the following acl
acl Blockword url_regex sex http_access deny Blockword
but problem occur in some websites where 'sensex' word found in url. Then squid block 'sensex' url content website also..
I have a user that has already used up a demo 24hr trial on my website. At present, I only check the customer id and the IP address to search for duplicates. On the whole this works but it's not foolproof. We now have 1 user from China that is changing their IP address everyday to get access to the free trial. Any options on what to do? I thought of downloading a cookie to their computer that the website could pick up - again not foolproff but most people don't disable cookies. Any other options?
I could ban China temporarily until the user gives up but if they find another proxy to chain then their IP address will be different again.
I have searched other post on here and they appear to be relevant but when I enter in the exact same commands it denies relay access to everyone. I have also used the postmap command to refresh the database.
Below is my main.cf # See /usr/share/postfix/main.cf.dist for a commented, more complete version # Debian specific: Specifying a file name will cause the first # line of that file to be used as the name. The Debian default # is /etc/mailname.
I seem to be able to install / configure Postfix server in 10 minutes as an MTA for a single domain but my struggle is really understanding the maps / restrictions which even after reading "The Book of Postfix" is not very clear to me:
[Code]....
My question is between those commonly used three maps above, what are the difference between them and how do I know when to use one over the other? Can someone clearly explain them to me? Here's what I have in my 'main.cf' but honestly I couldn't tell you if they're correct or now:
I'm trying to configure notifying 'sys admins' cellulars about some events in clients IT infrastructure. In linux env it would be ease, just to use 'mail' cmd but need solution for windows env in this case.
I have RAID Controller running under windows xp and its raid array sotfware has feature to notify about events by email. I found some free solutions on the internet but most of them are shareware and if are freeware, they have limitations e.g. nr of sent emails per day.
Is there some free smtp server that I can use with no limitations?
Can I limit the system resources that a process can use on Linux? I want to configure the system to avoid that some specified processes use some system resources:
choose if a process is allowed to use network and Internet.
choose which files and folders that a process can read, write or execute.
choose if a process is allowed to use sound and graphics output, and printer.