Ubuntu Security :: Exploits In Kernel - Possible To Bypass IPTables Firewall
Jul 4, 2010When there is an exploit in the kernal, can the iptables firewall get bypass? If yes,how do you know? Otherwise how can you find out.
View 9 RepliesWhen there is an exploit in the kernal, can the iptables firewall get bypass? If yes,how do you know? Otherwise how can you find out.
View 9 RepliesA few days ago I installed F12 and it was working fine very well up until today when I booted my computer from a perfect working order state yesterday to this. Well my wireless was still being sniffed and slowed down to dial up speed but what's new thats been consistant for at least 3 months I can't really do much about it since my brother doesn't like changing the password.
I recently logged onto my new fedora 12, 64-bit, system encrypted (all partitions effected by install), selinux enforced install to find myself in tty4 and some "other" users logged on to the other terminals. My folders would have lock icons on them after opening, my notication menu/toolbar crashed and hasn't returned on system reboot, some data transfers between removable storage returned input output errors while others worked fine(?). I also recieved this kernel bug output from the bug reporting tool but I have no idea what it means.
Also I was not loose with the security either I had removed unconfined login types (After setting up the system as I needed) meaning I couldn't even run root or sudo and neither could anyone else (asfar as I was aware). I pretty much increased selinux to its maximum boolean strictness and limited the _default_(Me included) account to a user from a _default_ unconfined (to actually be able to log in with the selinux boolean in place). Meaning they "the exploiters" were able to bypass selinux as a user account? How is that possible and even if you do root logon is disabled by selinux too?
At the moment I'm on a live cd trying to look for a way to custimise them as it seems it may be my only option.
Just a side note you can't just log in to tty4 by default without actively taking up spaces either by other users or your own use. Meaning since the tty login is automated 3 terminals were in use tty1, tty2 and tty3.
Which commands should I run to find out what is being done?
Edit: Just had my F12 x64 live cd taken down twice and had to hard reset as the toolbar disappeared. Took a photo of the last error message. I was just reading a pdf and using firefox at the time.
Is fedora usually this easy to hack?
Can we use iptables as firewall instead of Juniper firewall
View 2 Replies View RelatedIn an effort to learn more about firewalls and iptables I have left behind gui set-up tools and have setup a firewall using iptables that logs to its own file. The firewall is as follows:
Code:
*filter
:INPUT DROP [0:0]
:FORWARD DROP [0:0]
:OUTPUT ACCEPT [0:0]
:TCP - [0:0]
[Code]...
So I know Linux has iptables, I'm rather new to linux, and I'm wondering, are the stock settings with Ubuntu/Kubuntu safe? Is there anything I need to do make them more secure? I tried adding rules myself for some things but ended up just not being able to do anything so I had to reset back to stock with iptables -F. Should I be safe running as-is?
View 4 Replies View RelatedI have a server with 14 IP's on eth0. I'm using virtual interfaces to handle the IP's, but the iptables don't seem to work on the virtual interface. It blocks ports that I want open. I'm not that great with iptables, I use what I have because it works for me, but as far as tweaking it, I'm pretty lost.
My iptables:
# Simple Firewall configuration
#
# Set default policies --------
*filter
:INPUT ACCEPT [0:0]
:FORWARD DROP [0:0]
:OUTPUT ACCEPT [0:0]
#
# Internal Networks -----------
#-A INPUT -s <private.class.C>/24 -d <private.class.C>/24 -i eth1 -j ACCEPT
#
# Loopback --------------------
-A INPUT -s 0/0 -d 0/0 -i lo -j ACCEPT
#
# Accept established connections
-A INPUT -m state --state ESTABLISHED,RELATED -j ACCEPT
#
# Services --------------------
#
# For SSH gateway
-A INPUT -p tcp -s 0/0 -d 0/0 --dport 22 -m state --state NEW -j ACCEPT
#
# For SMTP gateway
-A INPUT -p tcp -s 0/0 -d 0/0 --dport 25 -m state --state NEW -j ACCEPT
#
# For FTP server
-A INPUT -p tcp -s 0/0 -d 0/0 --dport 20 -m state --state NEW -j ACCEPT
-A INPUT -p tcp -s 0/0 -d 0/0 --dport 21 -m state --state NEW -j ACCEPT
-A INPUT -p udp -s 0/0 -d 0/0 --dport 53 -m state --state NEW -j ACCEPT
#
# HTTP services
-A INPUT -p tcp -s 0/0 -d 0/0 --dport 80 -m state --state NEW -j ACCEPT
#
# HTTPS services
-A INPUT -p tcp -s 0/0 -d 0/0 --dport 443 -m state --state NEW -j ACCEPT
#
# POP-3 services
#-A INPUT -p tcp -s 0/0 -d 0/0 --dport 110 -m state --state NEW -j ACCEPT
#
# IMAP services
-A INPUT -p tcp -s 0/0 -d 0/0 --dport 143 -m state --state NEW -j ACCEPT
#
#PLESK
#-A INPUT -p tcp -s 0/0 -d 0/0 --dport 8443 -m state --state NEW -j ACCEPT
#
#Games
-A INPUT -p tcp -s 0/0 -d 0/0 --dport 28960 -m state --state NEW -j ACCEPT
-A INPUT -p udp -s 0/0 -d 0/0 --dport 28960 -m state --state NEW -j ACCEPT
-A INPUT -p tcp -s 0/0 -d 0/0 --dport 27666 -m state --state NEW -j ACCEPT
-A INPUT -p udp -s 0/0 -d 0/0 --dport 27666 -m state --state NEW -j ACCEPT
-A INPUT -p tcp -s 0/0 -d 0/0 --dport 28961 -m state --state NEW -j ACCEPT
-A INPUT -p udp -s 0/0 -d 0/0 --dport 28961 -m state --state NEW -j ACCEPT
-A INPUT -p tcp -s 0/0 -d 0/0 --dport 28962 -m state --state NEW -j ACCEPT
-A INPUT -p udp -s 0/0 -d 0/0 --dport 28962 -m state --state NEW -j ACCEPT
-A INPUT -p udp -s 0/0 -d 0/0 --dport 27015 -m state --state NEW -j ACCEPT
-A INPUT -p tcp -s 0/0 -d 0/0 --dport 27015 -m state --state NEW -j ACCEPT
-A INPUT -p udp -s 0/0 -d 0/0 --dport 27016 -m state --state NEW -j ACCEPT
-A INPUT -p tcp -s 0/0 -d 0/0 --dport 27016 -m state --state NEW -j ACCEPT
-A INPUT -p udp -s 0/0 -d 0/0 --dport 27017 -m state --state NEW -j ACCEPT
-A INPUT -p tcp -s 0/0 -d 0/0 --dport 27017 -m state --state NEW -j ACCEPT
-A INPUT -p tcp -s 0/0 -d 0/0 --dport 27020 -m state --state NEW -j ACCEPT
-A INPUT -p udp -s 0/0 -d 0/0 --dport 27020 -m state --state NEW -j ACCEPT
# Disallow fragmented packets
-A INPUT -f -j DROP
#
# Log & Block broadcast packets
-A INPUT -d 255.255.255.255/0.0.0.255 -j LOG
-A INPUT -d 255.255.255.255/0.0.0.255 -j DROP
# Log & Block multicast packets
-A INPUT -d 224.0.0.1 -j LOG
-A INPUT -d 224.0.0.1 -j DROP
#
# Log and drop all other incoming packets
-A INPUT -j LOG
-A INPUT -j DROP
#
COMMIT
How come FC11 is set by default to reset the IPTables firewall to ACCEPT across the board each restart?
View 4 Replies View RelatedDoes anyone know any common apache 2.2 exploits and how to stop them? I am setting up a web server and want it to be secure as possible. I currently have a basic lamp server on a ubuntu server.
View 1 Replies View RelatedI am facing a strange problem witht my iptables as there are some firewall entries stored somewhere which is displaying the below firewall entries even after flushing the iptables & when I restart the iptables service then the firewall entries are again shown in my iptables as shown below,
[root@myhome ~]# iptables -L
Chain INPUT (policy ACCEPT)
target prot opt source destination
[code]....
I have some questions about security
1> are the flash exploits are of any use to a Linux operating system like Ubuntu etc. ?
2>are the Microsoft office exploits any risk to libreoffice or open-office software suites?
3>are there exploits for Linux , open-office and libreoffice ?
IPtables creates an error during startup as well as when I try to restart it: Here's the output of:
[Code]....
I have set logwatch to report daily the logs, somehow since last week i get below message. A total of 1 possible successful probes were detected (the following URLs contain strings that match one or more of a listing of strings that indicate a possible exploit):
/cgi-bin/blocked.cgi?clientaddr=192.168.1.108&clientname=&clientident=&clientgroup=limitedaccess&targetclass= untrusted&url=http://adfarm.mediaplex.com/ad/fm/9608-84171-8772-2?
[code]...
I refrained from posting this in the Kernel Vulns thread earlier, due to its zero-day status. But now that the issue has been Slashdotted, there's no use in keeping us from publicly discussing this vulnerability. The link to the article (from which I quote below) is here. Brad Spengler's original announcement on the Dailydave mailing list is here.Quote:A researcher has published exploit code for a new vulnerability he discovered in the Linux kernel. The vulnerability is an especially interesting one in that the researcher who discovered it, Brad Spengler, has demonstrated that he can use the weakness to defeat many of the add-on security protections offered by SELinux and AppArmor.
View 9 Replies View RelatedRecently I had a Java exploit on Windows. Luckily Microsoft Security Essentials identified and removed it. Such things can happen on Linux as well, from what I've heard. Why does Linux offer no such detection?
View 14 Replies View RelatedI have a few mail servers (CentOS 5.5) that are running OSSEC Active Response (2.5.1) on Iptables (1.3.5-5.3.el5_4.1). We are currently having a problem where we get loop hook errors:Jan 24 04:15:03 servername kernel: iptables: loop hook 1 pos 464080 00000022 this is the firewall-drop.sh we are currently using:
Code:
#!/bin/sh
# Adds an IP to the iptables drop list (if linux)
# Adds an IP to the ipfilter drop list (if solaris, freebsd or netbsd)
# Adds an IP to the ipsec drop list (if aix)
[Code]...
I'm using Fedora 10 as a proxy server using squid, but I recently noticed that some users use the IPS's Dns to bypass the proxy and surf the web freely. So my question is, is this a problem with Squid or perhaps I can solve the problem whit IPTables.
View 6 Replies View RelatedI love My linux OS, and I carry It with me all the time in USB. I used to be able to boot from USB in the University computers, but not any more. Now it required Admin password in order to boot from CD or USB. I tried The VMWare, but I didn't like it. Is there any way I can get around it.
View 1 Replies View RelatedMy company web access is behind proxy(http://abc.proxy). Network admin can get to check who is top10 user and web they access. I owned a centos server. I have a thought that create an encrypted tunnel within proxy so the admin cant detect my http address. This is how it going to works
client with OpenVPN -> OpenVPN server(centos with company proxy)-> proxy -> internet
My connectivity in my client are using OpenVPN server as bridge. Hence, no record for client is recorded in my Network admin monitoring list. OpenVPN server's activity can be traced by network monitoring tools, just assume that our ultimate goal is to hide client activity.
I am having a little trouble setting up a NAT firewall using iptables. I have 1 PC dedicated to being the firewall running Ubuntu 10.04 LTS. There are 2 NICs in this PC. One NIC is connected to the modem & the other is hooked into my router, sharing the connection through to the other PC on my LAN. Thing is that I am having troubles setting this up using iptables. I have it sharing the connection, but can't seem to make it forward 2 ports through to my webserver on the LAN. I am also wanting to setup init.d to control iptables. I have been trying to google this, but haven't found anything useful to get this accomplished. I put the following into rc.local to make the forwarding work:
/sbin/iptables -F
/sbin/iptables -N block
/sbin/iptables -A block -m state --state ESTABLISHED,RELATED -j ACCEPT
/sbin/iptables -A block -m state --state NEW -i ! eth0 -j ACCEPT
/sbin/iptables -A block -j LOG
/sbin/iptables -A block -j DROP
/sbin/iptables -A INPUT -j block
/sbin/iptables --table nat -A POSTROUTING -o eth0 -j MASQUERADE
I'm trying to configure a messaging system based on php script with iptables, rsyslog and mysql. In the firewall server Sendmail is istalled so i'd like to use it to receive messages in the main mail server, located in DMZ. In the Lan i've configured another linux server than works perfectly, sending mail messages to the server located in DMZ. After looking for some solution I've tried to configure sendmail to relay mail to the server but the only thing I reached is the following message: "Deferred: Connection refused by mail.server.com" message rest on queue and flushing it reply
[Code]...
Does the rc.firewall script just use iptables or is that something completely different?
View 3 Replies View RelatedI don't know if this is Just my Machine, or not. But here is it:
Ubuntu 10.01
Acer Aspire 7740
When the computer is locked. I can smiply go to switch users. when the list of users logins are shown all i have to do is click on my user name and it allows me into my account without typing in a password. I can lock the computer manually or wait for it to time out it doesn't matter. the switch users method allows me to bypass the password protection.
I have to make my final exam on network & security, my task is to compare Iptables and a firewall program, yeah not a distribution. I use Archlinux so I'd like to continue use it for my project! Anyway my question is: which Network firewall could be the best one? I need this features:
- packet filtering
- HA (High Availability)
- Live (active) connection migration (if one fails...)
- Load balancing (not really important but...)
I know "many" firewalls but they all are distro. I need something to install on a linux machine (as said... better if I can run it directly on Archlinux!)
Recently I have been working on iptables and trying to understand how to use it. Here's a little script I have written to setup a basic firewall for myself:
Code:
#!/bin/bash
if [ `id -u` -ne 0 ]; then
echo "You need root privilege"
exit 1
fi
PROG=/sbin/iptables
$PROG -F
function sethttp {
echo "Opening http port..."
$PROG -A INPUT -p tcp --dport 80 -j ACCEPT
}
function sethttps {
echo "Opening https port..."
$PROG -A INPUT -p tcp --dport 443 -j ACCEPT
}
function settorrent {
echo "Opening torrent port..."
$PROG -A INPUT -p tcp --dport 52413 -j ACCEPT
}
while getopts "hst" option; do
case "$option" in
h) sethttp;;
s) sethttps;;
t) settorrent;;
*) echo "DOH!"
esac
done
$PROG -A INPUT -m state --state ESTABLISHED,RELATED -j ACCEPT
$PROG -A OUTPUT -m state --state NEW,ESTABLISHED,RELATED -j ACCEPT
$PROG -A INPUT -i lo -j ACCEPT
$PROG -A OUTPUT -o lo -j ACCEPT
$PROG -A INPUT -j DROP
$PROG -A FORWARD -j REJECT
echo "Done setting up the firewall! Enjoy :)"
exit 0
OK, this can take 3 arguments that open ports 80, 443 and 52413. And at the end, some default rules are applied. But here's the thing I don't understand:
if I don't give the argument for port 80, I can still view web pages... and also, when I remove the line:
Code:
$PROG -A INPUT -m state --state ESTABLISHED,RELATED -j ACCEPT
Whether I say it to open port 80 or not, I can't view any web pages.
I am encountering problems to configure my firewall (through iptables) to allow apt-get features, like update and install.I have the latest debian server running in a virtual machine in my windows xp and therefore I have two interfaces in this debian server:- NAT Interfaceinet: 10.0.2.15- Host Only Interfaceinet: 192.168.56.101So far my iptable rules drop all packets for default, in exception icmp and ssh that I allow to ping and connect from my windows xp. Both of them I use only the Host interface (192...) to connect to another 192... interface on my windows.
Those are working fine, but apt is not. I know, in this very moment it shouldn't. But I made a lot of attempts trying to configure the iptables allow connections through the 80 and 21 ports from/to NAT and Host. I think I made all possible combination (or not, because it didnt work). But I'm wondering if someone more experient can help me solve this problem.
I just install 1 firewall using Iptables.
Firewall includes 2 NIC:
NIC1 <IP PUBLIC>
NIC2 192.168.10.1
I installed 1 web server IP: 192.168.10.2
I have some PC IP range: 192.168.10.10->20
I set rules NAT on firewall and PC & web server can connect internet good, but I have problems:
When PC access to web server with IP 192.168.10.2 that ok, but PC can't access to web server when using IP Public. But outside internet, I can access to web server using IP Public.
Rules on IPTables
Code:
# Generated by iptables-save v1.3.5 on Sun Mar 7 21:01:16 2010
*nat
:PREROUTING ACCEPT [950:126970]
:POSTROUTING ACCEPT [89:5880]
:OUTPUT ACCEPT [19:1342]
-A PREROUTING -d 209.99.242.124 -i eth0 -p tcp -m tcp --dport 80 -j DNAT --to-destination 192.168.10.2:80
-A POSTROUTING -s 192.168.10.0/24 -o eth0 -j SNAT --to-source 209.99.242.124
*filter
:INPUT DROP [1599:157409]
:FORWARD DROP [0:0]
:OUTPUT ACCEPT [232:34452]
-A FORWARD -i eth0 -o eth1 -m state --state RELATED,ESTABLISHED -j ACCEPT
-A FORWARD -i eth1 -o eth0 -j ACCEPT
-A FORWARD -i eth0 -o eth1 -d 192.168.10.2 -p tcp --dport 80 -j ACCEPT
-A INPUT -i lo -j ACCEPT
-A INPUT -i eth1 -j ACCEPT
-A OUTPUT -o lo -j ACCEPT
-A OUTPUT -o eth1 -j ACCEPT
COMMIT
I had been using Guarddog for iptables but I updated to KDE4 and guarddog does not work with KDE4, neither does Firestarter. Is there a Graphical interface for iptables available that works with KDE4?
View 2 Replies View RelatedI have setup sshd_config for port 3210! I have difficult setup the iptables firewall to allow ssh on port 3210! i always enter this:
iptables -A INPUT -p tcp --dport 3210 -j ACCEPT
iptables -A OUTPUT -p tcp --sport 3210 -j ACCEPT
but after i restart ssh, the system do not allow port 3210... why? do i forget something?
I'm at my wits end at this point.
CentOS 5
iptables 1.4.1.1
APF 9.7
Webmin 1.4.80 (yes i can be a gui noob at times)
I'm currently running a dedicated server that hosts a couple of sites and runs a game server or two. I was using iptables on its own for a while, but recently I'm a target of all sorts of attacks (typically aimed directly at the gameserver on port 7777. UDP flood attacks, etc). I'm also seeing an spike in foreign spam, SSH brutes, and a few people in Turkey thought it would be cute to download files over and over and over I have decided to start banning entire countries, using the subnets listed here [URL]... I'm trying to block Central and South america (189,190,200,201.x.x.x), China, Ukraine, Turkey, Iran, Spain and Italy. I do this because a majority of the traffic from those areas are usually up to no good.
I installed APF so I could easily add these ranges in deny_hosts.rules and be done with it. I added the ranges, which turned out to be too many, and the system tanked. I decreased the amount of ranges to just 4:
189.0.0.0/8
190.0.0.0/8
200.0.0.0/8
201.0.0.0/8
Restarted APF and it loaded fine. Do an apf --list and iptables --list and it shows those 4 ranges as blocked. The only issue...I have people from 190.x and 200.x connecting to the gameserver and PLAYING. Its as if the firewall isnt there. Also, adding those ranges to /etc/hosts.deny (or whatever) doesn't block them either.
I add one of my own ips and I get blocked instantly. WTF?? I look in the iptables for webmin, and it shows an empty firewall. I do iptables --list and it shows the ranges I added in APF. I'm looking at building (or whatever its called) an fresh iptables with the geoip module added in. [URL]...
I've never done anything like this, and I don't want to kill the box. I also don't want to spend the effort if 1) something is wrong with my system to begin with and 2) the geoip module doesn't work. geoip module aside....how exactly should i configure the firewall? Empty iptables completely and then rely on APF for everything? Oh and heres another tidbit: I tried this before 2 years ago and it used to be that anything I put in APF would show when i looked at iptables using the webmin module. Thats no longer the case now. That was also on CentOS 4 when it did that. I don't know if moving to 5 is whats preventing it now.
In a nutshell, I'm new at this and I'm being inundated with terrible people trying to do terrible things and I'm ready to just give up. Can someone just give me a quick rundown on:
1) how to test that my firewall is actually firewalling
2) how I should configure the/a firewall on this CentOS5. Not too specific, I just want to know if I should empty iptables then load apf, should i not bother with APF (i like it when it works), is there a specific order of doing things?
I have managed to get iodine working between my ubuntu intrepid box and my windows client with a caveat.
The firewall rules allows DNS queries inbound. The client tunnel endpoint gets assigned an IP address and the tunnel is established properly.
However when I try to ping from the client machine, the reply packets are not coming back.
I used TCPDUMP on the Ubuntu box and watch the dns0 tunnel interface, and noticed that the packets are reaching the Ubuntu box from the client, but I don't see ANY ICMP echo replies until I turn off the firewall from Firestarter.
I see that outbound access rule is to allow all.