Security :: Setting Up Server With IPSec Authentication
Aug 6, 2010
As part of the project I'm working on, I need to set up a server with IPSec authentication only connections to a large number of low bandwidth clients. I'm making use of the PF_KEY interface to populate the keys on the server and while prototyping things I've found that the initial setup is taking longer than I had expected. At the start of my test, entries are being added to the database at a rate of around 30/second, but as time goes on this is dropping significantly. I ran a test up to around 100k entries and by then the rate had dropped to 10/second. It's key to me that if I reboot my server that the Security Associations can be repopulated in a very short period, so I do genuinely need this to be much faster.
Two questions:
1) Does anyone have any experience of running with a large number of SAs set up, and if so what sort of setup rate did you get?
2) Are there things I can do to speed up the provisioning of these SAs? I'd really like to see a rate in the thousands per second.
We've been doing the prototyping on the 2.6 kernel.
I've searched through google, and all I can find are instructions on how to set up a L2TP/IPSec VPN that works with macs and iPhones. I'm NOT trying to set up an L2TP/IPSec VPN. I'm trying to set up a pure-ipsec vpn.
The iPhone IPSec client is a built-in cisco client, I believe. I'm staying away from L2TP and PPTP because I need multicast packets to go through. *edit: wow, i just noticed that the title says "8.10 LTS". Oops! I obviously mean "8.04 LTS". Gah, the lack of sleep got to me.
My client is on Ubuntu Lucid 10.04, I installed ipsec-tools and racoon from the repositories. The gateway is installed on a CentOS machine. I've configured everything to get a working roadwarrior configuration with authentication_method hybrid_rsa client and server. It's working in aggressive mode, but in main mode I can't get it working. I delivered new CA and certificates several times but I'm still stuck.
It seems that it comes from my client not supporting the certificate sent by the server. The client contains a copy of the CA, whereas server has a private key and a certificate signed by the CA.
I am setting up Fedora-13 with "389 directory server" for authentication. I had performed the following steps.
1. Install FD-13. 2. Yum install 389-ds. 3. Run script to configure. 4. Start 389-condole and create few group and user for testing.
I can see these user with "ldapsearch" and with "phpldapadmin". It looks my server is responding. However, I am unable to see any user name with "getent passwd". also "ssh server_user@server" is not able to login. Whereas "getent passwd" shows local user and "ssh local_user@server" is able to login. Also note that I am not using ssl, so want to avoid ssl.
I want to setup firewall protection with iptables to support IPSec tunnels. That is, the firewall will drop anything from any host if it is not from an established IPSec tunnel. And it will accept anything (any protocols) if it's from an IPSec tunnel.
That is, I need also to open up ping to make ping work. But if I open up icmp, I cannot prevent pings from hosts that's outside my IPSec tunnels. This defeats my purpose.So if my purpose is to allow "anything" within the tunnel and disallow/drop anything outside the IPSec tunnels, how should I setup the iptables rules?
I've set up two security associations(in and out) on two hosts, and then set up two policies per host that should filter traffic to those SA's. Yet when I try to ping one host from the other I get no response, meaning that the filters on one side work and drop unprotected packets, but both hosts are configured to communicate using ipsec. Can anyone point me in the right direction?
Code: ip xfrm state add src 192.168.77.23 dst 192.168.77.24 proto esp spi 0x53fa0fdd mode transport reqid 16386 replay-window 32 auth "hmac(sha1)"
Above mentioned is my configuration. when i try to connect client with SSH it is not sending a request for authenticating user to RADIUS server. what else configuration i have to do, or if there are any mistakes in my configuration
I'm new in UNIX & trying to access the server using SSH but I encounter this error PAM Authentication Error. I use edit /etc/ssh/sshd_login & set the PermitRootLogin to yes. But didn't work. I used this command ps -ef | grep sshd & saying Process environment requires procfs(5). I don't know what to do now. What I want is access it by SSH but I got Access Denied. [MOD]Pruned from [URL]. create your own thread instead of resurrecting a five year old one.[/MOD]
I have been trying to setup IPSEC encryption between two linux boxes. I have a server application which runs on Linux Box A and a client application which runs on Linux Box B. The client sends the data to server. I have captured wireshark logs at both server and client end. In the wireshark logs I can see that the Box B send ESP packets to the Box A.
But the server Application running at Box A is is not able to get any packets. If I turn the policy off at Box B, Box B sends normal UDP data packets to Box A, but still the Server Application running at box A doesn't get any packets.( Expected behavior since policy at Box A enforces that all packets coming from Box B should be encrypted.)
If I turn the policy off at Box A and Box B both, the server application receives the unencrypted data which is also expected behavior. But when the policy is turned on at both the boxes the encrypted packets reach the Box A but are not delivered to the server application. If anyone has faced such issue please help me to debug this issue. I have attached the ifconfig and policy settings at Box A and Box B for your reference.
I need to make a choice on what authentication protocol I want to use for Authentication and Authorization. I was looking at Radius and then literature suggested that Diameter was a better protocol. Keep in mind I need this on a hetrogeneous setup ( linux & windows together). Diameter seemed like a good fit until I discovered that the open source code no longer seems to be maintained ( C/C++).
I was also looking at Kerberos as an option though there is alot overhead with the server. SSL/TLS or EAP? I am looking for simple but secure and am new at the security protocols.
Has anyone got experience connecting a linux machine to a Microsoft VPN server using RSA authentication? What puzzles me perhaps most about this topic is the absolute dirth of information. If it is not possible, can anyone tell me why?
I followed this How To (https://help.ubuntu.com/community/Postfix) in order to add smtp authentication to my Postfix installation used as spam filter for my exhange server, and it'seem all ok; the only thing that I don't understand is where I list all the users (with passwords) that I authorize to send mail through my server...
I recently set up a ftp server in my house running a dyndns service so I can get to it from the outside. I called my isp to get some help in setting up the router to forward port 21 from the outside to that box, and in short we had some problems. Long story short, they ended up bypassing the router itself, and now the line running to the box is its own fixed external ip. Naturally I want a pretty darn good iptables setup for this. The box runs proftpd and so far my iptables only accepts local loopback and port-21. (I left port 80 closed as its only purpose is to be a standalone ftp server) But I know there must be a safer rule for port 21, as right now its just wide open. Anyone have any ideas on how to make this a bit safer? Also would that command be fine for any of the linux machines im connecting to it from the outside too?
how to lock down individual users from setting a proxy server. Its a server not a WS so it should never go to the internet. I want to lock down the system side and firefox 5 settings.
I wanted to set up Computer Lab. loading Fedora 11 OS and one system acting as a Server to store Users(Student) Login Informations. When students do a programs, all programs (eg, C++ programs) files should be saved in the local fedora system but when login to the system, the login should be validate by a Server System.
I have a problem with ldap client authentication in ubuntu. I am using rhel5 as openldap server and I configured ubuntu as client, when I am trying to login the following message is coming."su: Authentication service cannot retrieve authentication info. Sorry" But when I do search through "ldapsearch" command output is coming without any errors, Can anybody explain what would be problem.
i want to setup a proxy authentication on fedora 15 for minimum of 5 users using squid but i had successfully setup a proxy sever on my system and its working perfecly.# vi /etc/squid/squid.conf this was the output.
I need to send authenticated mails from a RHEL machine to a relay that only accepts authentication mails from domain users. I've never done this so I'm a bit lost. I read some sendmail docs and I uncommented these optins in sendmail.mc file
I've been trying to set up my printer that I'm sharing through a desktop running Windows 7 and am having trouble getting the authentication details to stay saved, or even work, for that matter.
No matter whether or not I tell the New Printer program to remember my authentication details, it doesn't. If I tell it to prompt me if it needs details, I get no prompt. So, in order for me to print a document, I have to tell it to print, then go into the print queue and authenticate it from there.
1. IIs there a way, either manually or through GUI, to either save my authentication details or just have it prompt me upon printing to save me from going into the print queue?
2. This isn't exactly on-topic, but it came up while I tried to set up the printer. Has anyone else had their SMB URI come up incorrectly with the spaces incorrectly rendered as "20" instead of "%20:?
My work has got a new Sharp MX-1800N network printer. The setup requires user authentification, one password for printing b&w, another for printing color. However when they try to set up my netbook running ubuntu desktop 10.04 they don't get it to prompt for a password, which makes the printer reject the printout. How do you fix this?
I was trying to configure user authentication in SSH using certificate method.As u all know the usual way of authentication is using the ssh-keygen method. But i want the another method where we create a certificate key and send it to the CA, which signs it and send back etc etc.I cannot find any unique procedure in the net to configure this method.
I have squid proxy authenticating Internet users with LDAP. It's working well. But I have problem when I authenticate to squid proxy to login to Yahoo Messenger. Each time, I login to YM application, the squid proxy popups many authentication windows. These confuse users when they you YM. I checked in squid access log and see that: when users use YM application, the application requests the following links:
[code]...
With each link, squid requires one authentication window. Do you have any ways to squid require only one authentication window when users use YM?
I'm working on a work project related to Web (Client) authentication and DOD Common Access Cards. But I'm having difficult getting the details about what happens on the CAC side of things.
I familiar with the PKI system as it applies to e-mail. (Correct me if I err, of course.) If you want to sign an e-mail (i.e., so it can be authenticated by the receiver) you use your private key to add a digital signature to the message. Then, the receiver uses your published public key to determine if the digital signature is valid, i.e., was created using your private key (even though the receiver never actually has access to your private key).
So... my questions:
1) When a person with a DOD CAC visits a CAC-enabled web site, and the server grants access after the CAC is inserted, is the authentication process fundamentally the same as what happened with the e-mail authentication?
2) If the private key is used in this process (it would have to be, correct?) is the signature created on the CA Card electronics (i.e., the private key remains on the CAC)? Or is the private key copied onto the computer, which uses it to create the signature?
My server is connected to the Internet for ssh on port 22 with root logon disabled, a single non-dictionary word user name allowed, and pki authentication only (about as secure as I can make it). I've previously run fc5 and 9 servers using the same sshd config since 2006 and had no security troubles, so I'm happy, but.. After the fc13 install and configuration, logins from a host on a remote network are taking about 1m 30s to complete! A (partial) console output for ssh -vv appears below. The lines marked with "**" were the lines after which significant pauses happen. This is fully repeatable.
Code:
debug1: Next authentication method: publickey debug1: Offering public key: debug2: we sent a publickey packet, wait for reply
