November 2000 - openSUSE Security - openSUSE Mailing Lists

Problems with sendmail & relay
by RoMaN SoFt / LLFB!! 22 Nov '00

22 Nov '00

Hi. I'm running SuSE 6.4 and need to run sendmail (host permanently connected to Internet). By default, SM 8.9.3 comes with relay denied for all. I want to set an acceptable secure sendmail. The scenario is as follows: - the smtp (mta) is aaa.bbbb.com (where aaa & bbbb are ficticious). I want the rely denied by default but also want granted access for users whom From:'s field is: *(a)bbbb.com and *(a)cccc.com (being cccc.com another domain which is NOT local to the machine; bbbb.com is the "local" domain, I mean, there is a MX record pointing to the mta's ip.). The problem is that my users can connect to the smtp machine from *ANY* ip. So the rely-filters only could trust in the "From:" line in header's mail. I know this isn't too much secure, since spammers could send mail spoofing the From: field (which is trivial). But it's more secure than a sendmail running with "promiscuos relay" feature turned on. I'm new to sendmail so I need some help. I've read some docs at www.sendmail.org and have a look to O'Reilly sendmail book. But it still doesn't working. These are the attempts I've made: 1) Using Yast, I created a /etc/sendmail.cf. Then I personalized a little using Yast too and added the domain: bbbb.com and ddd.bbbb.com (which is an alias to aaa.bbbb.com). Afterwards I modified /etc/mail/access and added: cccc.com RELY Finally: # makemap hash /etc/mail/access < /etc/mail/access # /sbin/init.d/sendmail reload (or killall -HUP sendmail) The result is that now I can send to recipients like: user(a)cccc.com. But this isn't the behaviour I want. What I want is that user(a)cccc.com can send (not be sent to) to any other recipient (at whatever domain) using my mta. 2) 2nd attempt: this time I edited /etc/mail/linux.mc and added a line: FEATURE(`relay_local_from') Compiled using: # m4 /etc/mail/linux.mc > /etc/sendmail.cf And reload sendmail: # /sbin/init.d/sendmail reload (Access file kept intact). The result apparently is the same. I cannot send to any arbitrary domain from user(a)cccc.com. I'm quite desperated. What am I missing????? =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= ** RoMaN SoFt / LLFB ** roman(a)madrid.com http://pagina.de/romansoft ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

10 18

SOLUTION: Chroot ssh/telnet login
by Stiefenhofer, Marek ECOFIS 22 Nov '00

22 Nov '00

How to chroot standard login users: You need a program as login shell, that does the following: 1.) Checks if called as login else exits 2.) checks if UID=0 (big security hole?!) 3.) call CHROOT 4.) sets UID=back to normal 5.) calls a standard shell I've found a little c-program that implements all of this features except 1.) here: http://www.phirate.ethos.co.nz/dev/srsh/ Maybe someone will have a look at this and check if the UID=0 part offers standard users possibilitities to break into a root shell or something similar... Marek Stiefenhofer -----Ursprüngliche Nachricht----- Von: Stiefenhofer, Marek ECOFIS Gesendet am: Montag, 20. November 2000 12:19 An: 'Gerd Bitzer'; 'suse-security(a)suse.com' Betreff: AW: [suse-security] Chroot ssh login I know, but that's not sufficiant. rbash does not prevent from accessing files below your home-directory like: ls ~/../../etc -----Ursprüngliche Nachricht----- Von: Gerd Bitzer [mailto:gerd.bitzer@tesion.de] Gesendet am: Montag, 20. November 2000 12:13 An: Stiefenhofer, Marek ECOFIS Betreff: Re: [suse-security] Chroot ssh login may be there's also another interesting possibility, socalled restricted shells. The user is then limited to its own homedirectory, as far as I know e.g. bash supports this mode with "bash -r". Maybe other shells also have this feature "Stiefenhofer, Marek ECOFIS" wrote: > Hi, > > I want to chroot user logins via telnet/ssh to their home directory. I guess > this is a standard procedure, but I'm sort of stuck. I can't chroot the > login shell of a standard user - only root can chroot. > > Examples would be appreciated... > > Kind regards, > > Marek Stiefenhofer > > --------------------------------------------------------------------- > To unsubscribe, e-mail: suse-security-unsubscribe(a)suse.com > For additional commands, e-mail: suse-security-help(a)suse.com --------------------------------------------------------------------- To unsubscribe, e-mail: suse-security-unsubscribe(a)suse.com For additional commands, e-mail: suse-security-help(a)suse.com

2 1

Re: [suse-security] question about logging
by webmaster 22 Nov '00

22 Nov '00

Let me say also: better have /var/log/messages to gather up all from kernel, warn, popper, named, and then have logcheck installed for checking system events. You can have that emailed to root, possibly to another machine. Allow me then to ask the list: the bind8 has user named, and group named - contrary to earlier versions where root was user and group. Which are the files named writes into, so that I should chown and chgrp those files? Besides: Has anyone been ruinning sshd2? /-------------------------/ Antal Leisen webmaster(a)kreorg.hu Kreorg Educational Center Budapest, Hungary

3 2

Securid for suse
by Peter Lange 22 Nov '00

22 Nov '00

Hello all, anyone knows about a ACE-Agent for Securid authentication available for SuSe? Thanks -- Mit freundlichem Gruß i.A. Peter Lange PROFI Engineering Systems AG Geschäftsbereich Antaris Systemtechnik Moosbürger Str.20 D-92637 Weiden Tel. 0 961/89-502 Fax 0 961/89-579 eMail: plange(a)antaris.de

1 0

Re: [suse-security] Authenticated routing?
by brandon＠2i.com 22 Nov '00

22 Nov '00

CheckPoint's FW1 is available for Linux, though RedHat is the only flavor supported. I have the full version running on Solaris and plan to give it a test run on SuSE near January. If you are only concerned about web traffic, you can set up Squid in proxy mode with authentication. Unfortunately you cannot use authentication in transparent mode and this only works for http traffic. I, too, wish there were a clean solution that could tap into the stock Linux firewall. At 01:29 PM 11/20/00 +0100, you wrote: >Hi, >I am looking for a solution to authenticate routing. > > IPSEC CLient (Mostly WinXX boxes) connects to firewall. > Firewall untunnels packets (FreeS/WAN) > Firewall authenticates user ???????? > Successful authentication enables routing of the clients Packets into the > internal net. > >Commercial FWs like FW1 have that feature, but I'd rather run it on Linux. > >Thanks for any pointers >afx > >-- >atsec information security GmbH Phone: +49-89-44249830 >Steinstrasse 68 Fax: +49-89-44249831 >D-81667 Muenchen, Germany WWW: www.atsec.com > May the Source be with you! > >--------------------------------------------------------------------- >To unsubscribe, e-mail: suse-security-unsubscribe(a)suse.com >For additional commands, e-mail: suse-security-help(a)suse.com > > > -- brandon(a)2i.com

1 0

RE: [suse-security] SuSE-Firewall logging
by Volker Kuhlmann 22 Nov '00

22 Nov '00

> With syslog you get not only the "facility" category (i.e. "kern") but > also the "priority" (or "severity") level. I didn't mention it because it's irrelevant here. Yes you can filter via facility as well as severity. It still doesn't allow you to filter by source (as in what program logged the entry). grep at a later stage is the only way. Volker

1 0

RE: [suse-security] SuSE-Firewall logging
by Volker Kuhlmann 22 Nov '00

22 Nov '00

No it's not possible to log only the firewall messages to a separate file. These log entries come from the "kernel", and the filtering which can be done via syslog is not very extensive. You can basically only filter on what category (called "facility") the log entry is to be recorded as, and choices there are very limited. In this case, it's "kernel". That means that everything logged under "kernel" will always go to the same file. Sorry you can't filter on which program the log entry came from. The default setup when you install the SuSE firewal package is to log all "kernel" entries to /var/log/firewall. The rest stays in /var/log/messages. Alternatively, reduce the log level for your firewall setup. Volker

2 1

NFS over VPN
by mute 21 Nov '00

21 Nov '00

Hi, recently at the place where i work a collegue came up with an idea to use nfs over vpn to sync our webservers. i have not done much work with either, and i have two questions about it. first, is this a practical solution? and second, im a bit stumped about resources on the subject (i do have nfs documentation, but not much of vpn resources, and nothing on how to combine the two). sorry is this question might seem trivial/stupid, any help/opinion is greatly appreciated : )

2 1

Re:Re:[suse-security] NET
by Ron Blanchett 21 Nov '00

21 Nov '00

Dustin, Could you send me your proxy config file so that I can have a better idea of your configuration so that I can possibly find a soltion to your problem Thanks Ron _____________________________________________________________________________________ Get more from the Web. FREE MSN Explorer download : http://explorer.msn.com

1 0

Re:Re:[suse-security] NET
by Ron Blanchett 21 Nov '00

21 Nov '00

Dustin, If you are not running more than one proxy I am not sure what else could be giving you that error message. I there server used for anything else (i.e. email server, web server) I will keep looking into your problem The proxy that I have setup at my work hasn't had this message yet so I am not sure what is causing it. Thank You Ron _____________________________________________________________________________________ Get more from the Web. FREE MSN Explorer download : http://explorer.msn.com

1 0

question about logging
by RENE 21 Nov '00

21 Nov '00

hello there some question about the loggin in /var/log/messages how can i sort the messages by file from var/log/messages. spetially: no firewall loging into messages, now on my system: firewall logging also under messages, but this file becomes very big, and is not easy to read, because i loog all traffic on my system thx Rene

2 1

AW: [suse-security] "Submission" port on sendmail
by Stiefenhofer, Marek ECOFIS 21 Nov '00

21 Nov '00

Hi Roman, did I get something wrong? The sendmail in the example is listening on this port. Marek -----Ursprüngliche Nachricht----- Von: Roman Drahtmueller [mailto:draht@suse.de] Gesendet am: Dienstag, 21. November 2000 15:33 An: RoMaN SoFt / LLFB!! Cc: suse-security(a)suse.com Betreff: Re: [suse-security] "Submission" port on sendmail > > emilio:~ # grep submission /etc/services > submission 587/tcp # Submission > submission 587/udp # Submission > > It permits users send or receive mail. Admin tasks are made through > ssh and some logs are sent via syslog. > > Why is sendmail opening "submission" port??? What is it intended for? > Is it insecure??? Could I close it? How? A port as such is never "insecure" (please don't prove me wrong...). sendmail uses port 587 by pure coincidence. Next time it will be another port. Roman. -- - - | Roman Drahtmüller <draht(a)suse.de> // "Caution: Cape does | SuSE GmbH - Security Phone: // not enable user to fly." | Nürnberg, Germany +49-911-740530 // (Batman Costume warning label) | - - --------------------------------------------------------------------- To unsubscribe, e-mail: suse-security-unsubscribe(a)suse.com For additional commands, e-mail: suse-security-help(a)suse.com

6 5

Suse firewall warning.
by Wade Chandler 21 Nov '00

21 Nov '00

Yurri, Does you /etc/syslog.conf file have a line like "kern.* /var/log/firewall" in it. If so then that is the file. If not then add the line and you will be able to have them all stored in one file. Wade Chandler.

1 0

AW: [suse-security] "Submission" port on sendmail
by Stiefenhofer, Marek ECOFIS 21 Nov '00

21 Nov '00

Excerpt from: SENDMAIL RELEASE NOTES $Id: RELEASE_NOTES,v 8.561.2.5.2.80 2000/07/19 20:40:57 gshapiro Exp $ sendmail implements RFC 2476 (Message Submission), e.g., it can now listen on several different ports. Use: O DaemonPortOptions=Name=MSA, Port=587, M=E to run a Message Submission Agent (MSA); this is turned on by default in m4-generated .cf files; it can be turned off with FEATURE(`no_default_msa'). -----Ursprüngliche Nachricht----- Von: Roman Drahtmueller [mailto:draht@suse.de] Gesendet am: Dienstag, 21. November 2000 15:33 An: RoMaN SoFt / LLFB!! Cc: suse-security(a)suse.com Betreff: Re: [suse-security] "Submission" port on sendmail > > emilio:~ # grep submission /etc/services > submission 587/tcp # Submission > submission 587/udp # Submission > > It permits users send or receive mail. Admin tasks are made through > ssh and some logs are sent via syslog. > > Why is sendmail opening "submission" port??? What is it intended for? > Is it insecure??? Could I close it? How?

1 0

"Submission" port on sendmail
by RoMaN SoFt / LLFB!! 21 Nov '00

21 Nov '00

Hi. That's a typical MTA machine: emilio:~ # lsof -i COMMAND PID USER FD TYPE DEVICE SIZE NODE NAME inetd 154 root 4u IPv4 38091 TCP *:pop3 (LISTEN) sshd 181 root 3u IPv4 82 TCP *:ssh (LISTEN) syslogd 14030 root 10u IPv4 27318 UDP *:syslog sendmail 16965 root 4u IPv4 65606 TCP *:smtp (LISTEN) sendmail 16965 root 5u IPv4 65607 TCP *:submission (LISTEN) emilio:~ # grep submission /etc/services submission 587/tcp # Submission submission 587/udp # Submission It permits users send or receive mail. Admin tasks are made through ssh and some logs are sent via syslog. Why is sendmail opening "submission" port??? What is it intended for? Is it insecure??? Could I close it? How? Regards. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= ** RoMaN SoFt / LLFB ** roman(a)madrid.com http://pagina.de/romansoft ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

2 1

Re:Re:[suse-security] NET
by Dustin Huptas 21 Nov '00

21 Nov '00

Hello Ron, no I have only one proxy running. My setup is as follows: Internet | | Cisco Router | |---------------------------------------------- | | | Firewall(SuSE 6.2) Webserver(SuSE 6.3) Proxy (SuSE 6.3) Mailserver | | --- internal network HTH, Rgds Dustin

1 0

Re: [suse-security] NET
by Ron Blanchett 20 Nov '00

20 Nov '00

Dustin, Do you have more that one proxy server running. If you do this may be caused by because one of your "Neighbor" proxies is miss configured. Please give more info on you proxy setup if this is not the case Thank You Ron Blanchett MBT Assistant Network Administrator >From: "Dustin Huptas" <dustin2010(a)gmx.de> >To: <suse-security(a)suse.com> >Subject: [suse-security] NET >Date: Fri, 17 Nov 2000 15:50:21 +0100 >MIME-Version: 1.0 >Received: from [202.58.118.3] by hotmail.com (3.2) with ESMTP id >MHotMailBBDE8EE8004DD82197CECA3A760304760; Fri Nov 17 06:42:48 2000 >Received: (qmail 30881 invoked by alias); 17 Nov 2000 14:48:39 -0000 >Received: (qmail 30874 invoked from network); 17 Nov 2000 14:48:39 -0000 >From suse-security-return-3975-rrcb Fri Nov 17 06:43:14 2000 >Mailing-List: contact suse-security-help(a)suse.com; run by ezmlm >Precedence: bulk >X-No-Archive: yes >list-help: <mailto:suse-security-help@lists.suse.com> >list-unsubscribe: <mailto:suse-security-unsubscribe@lists.suse.com> >list-post: <mailto:suse-security@suse.com> >X-Mailinglist: suse-security >Delivered-To: mailing list suse-security(a)lists.suse.com >Message-ID: <EKEHLKGPPAOBEPKLAJPCGEIOCEAA.dustin2010(a)gmx.de> >X-Priority: 3 (Normal) >X-MSMail-Priority: Normal >X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) >X-MimeOLE: Produced By Microsoft MimeOLE V5.00.3018.1300 >Importance: Normal > >Good day! > >My proxy wrote about 200 of the below messages into /var/log/messages >during >the last days: > >Nov 15 09:34:08 proxy kernel: NET: 166 messages suppressed. >Nov 15 09:34:08 proxy kernel: neighbour table overflow > >The number of suppressed messages is changing with every line. I am >completely stuck with this. Does this mean, someone is trying to do a >buffer >overflow? > >P.S.: I am using SuSE 6.3 on this machine. > >Thank you very much in advance for your help! > >Rgds Dustin ><< DustinHuptas.vcf >> >--------------------------------------------------------------------- >To unsubscribe, e-mail: suse-security-unsubscribe(a)suse.com >For additional commands, e-mail: suse-security-help(a)suse.com _____________________________________________________________________________________ Get more from the Web. FREE MSN Explorer download : http://explorer.msn.com

1 0

apache 3.1.14 update
by Thomas Michael Wanka 20 Nov '00

20 Nov '00

Hi, is there a plan to provide an apache 3.1.14 update? The apache group claims it fixes security holes in mod_vhost_alias and mod_rewrite modules. I assume that the version included in SuSE not to be affected with this problems, but customers request an update as the announcement on the apache homepage indicates that it is. thanks mike

2 1

AW: [suse-security] Chroot ssh login
by Stiefenhofer, Marek ECOFIS 20 Nov '00

20 Nov '00

I know, but that's not sufficiant. rbash does not prevent from accessing files below your home-directory like: ls ~/../../etc -----Ursprüngliche Nachricht----- Von: Gerd Bitzer [mailto:gerd.bitzer@tesion.de] Gesendet am: Montag, 20. November 2000 12:13 An: Stiefenhofer, Marek ECOFIS Betreff: Re: [suse-security] Chroot ssh login may be there's also another interesting possibility, socalled restricted shells. The user is then limited to its own homedirectory, as far as I know e.g. bash supports this mode with "bash -r". Maybe other shells also have this feature "Stiefenhofer, Marek ECOFIS" wrote: > Hi, > > I want to chroot user logins via telnet/ssh to their home directory. I guess > this is a standard procedure, but I'm sort of stuck. I can't chroot the > login shell of a standard user - only root can chroot. > > Examples would be appreciated... > > Kind regards, > > Marek Stiefenhofer > > --------------------------------------------------------------------- > To unsubscribe, e-mail: suse-security-unsubscribe(a)suse.com > For additional commands, e-mail: suse-security-help(a)suse.com

1 0

Chroot ssh login
by Stiefenhofer, Marek ECOFIS 20 Nov '00

20 Nov '00

Hi, I want to chroot user logins via telnet/ssh to their home directory. I guess this is a standard procedure, but I'm sort of stuck. I can't chroot the login shell of a standard user - only root can chroot. Examples would be appreciated... Kind regards, Marek Stiefenhofer

1 0

Martian Source detection
by Rainer.Frohne＠t-online.de 20 Nov '00

20 Nov '00

Hello, in the network of my company the firewall based on SuSE 7.0 the following kernel warning appears from time to time: Nov 14 16:03:35 dbwwall kernel: martian source 4796a8c0 for ff64a8c0, dev eth0 Nov 14 16:03:35 dbwwall kernel: ll header: ff ff ff ff ff ff 00 a0 f9 00 d0 57 08 00 I know what matian sources are. My question is, how I can find out the device or intruder, which causes the trouble. Which tool can I use to find the listed mac-address in my Network? Thanks in advance! Rainer Frohne

3 5

Firewall help needed
by Ralph De Witt 19 Nov '00

19 Nov '00

Hi to all I have been trying to install my firewall on a SuSE 7.0 box, I have one Win ME box connected to my network and would like to allow it to connect to the internet trough the Linux box. My linux box connects to the internet through eth1 and a DSL line. I use RPPOE to enable the ethernet card and connect to the internet. I believe I have configured the SuSE firewall correctly to allow me to connect to the internet and masqurade for the remaining win machine but the only thing I have accomplished to block myself from the net. I was hoping someone slightly more knowlegeable and less tired then I could look at my script and help me out. I have enclosed the script below: # Copyright (c) 1999,2000 SuSE GmbH Nuernberg, Germany. All rights reserved. # # Author: Marc Heuse <marc(a)suse.de>, 1999,2000 # Please contact me directly if you find bugs. # # If you have problems getting this tool configures, please read this file # carefuly and take also a look into # /usr/share/doc/packages/firewals/EXAMPLES ! # # /etc/rc.config.d/firewall.rc.config # # for use with /sbin/SuSEfirewall version 2.6 # # ------------------------------------------------------------------------ # # PLEASE NOTE THE FOLLOWING: # # Just by configuring these settings and using the SuSEfirewall you are # not secure per se! There is *not* such a thing you install and hence you # are safed from all (security) hazards. # # To ensure your security, you must also: # # * Secure all services you are offering to untrusted networks (internet) # You can do this by using software which has been designed with # security in mind (like postfix, apop3d, ssh), setting these up without # misconfiguration and praying, that they have got really no holes. # SuSEcompartment can help in most circumstances to reduce the risk. # * Do not run untrusted software. (philosophical question, can you trust # SuSE or any other software distributor?) # * Harden your server(s) with the harden_suse package/script # * Recompile your kernel with the openwall-linux kernel patch # (former secure-linux patch, from Solar Designer) www.openwall.com # * Check the security of your server(s) regulary # * If you are using this server as a firewall/bastion host to the internet # for an internal network, try to run proxy services for everything and # disable routing on this machine. # * If you run DNS on the firewall: disable untrusted zone transfers and # either don't allow access to it from the internet or run it split-brained. # # Good luck! # # Yours, # SuSE Security Team # # ------------------------------------------------------------------------ # # Configuration HELP: # # If you have got any problems configuring this file, take a look at # /usr/share/doc/packages/firewals/EXAMPLES for an example. # # # All types have to set option 1): set START_FW in /etc/rc.config to "yes" ;-) # # If you are a end-user who is NOT connected to two networks you just have to # reconfigure (all other settings are OK): 2), and maybe 9), 11), and 18). # # If this server is a firewall, which should act like a proxy (no direct # routing between both networks), or you are end end-user connected to the # internet and to a internal network, you have to setup your proxys and # reconfigure (all other settings are OK): 2), 3), 9) and maybe 7), 10), 11) # 12), 14) and 18). # # If this server is a firewall, and should do routing/masquerading between # the untrusted and the trusted network, you have to reconfigure (all other # settings are OK): 2), 3), 5), 6), 9), and maybe 7), 10), 11), 12), 15), 18). # # If you want to run a DMZ in either of the above three standard setups, you # just have to config 4), 9), 13) and maybe 19). # # If you know what you are doing, you may also change 8), 16), 17), 18) # and the expert options 20), 21), 22) at the far end, but you should NOT. # # If you use diald or ISDN autodialing, you might want to set 18). # # To get programs like traceroutes to your firewall to work is a bit tricky, # you have to set the following options to "yes" : 11 (UDP only), 19 and 20. # # If you want to load the full firewall rules for an interface even if it's not # available, configure a static IP and netmask (see 2, 3 and 4 for an example). # # Please note that if you use service names, that they exist in /etc/services. # There is no service "dns", it's called "domain", email is called "smtp" etc. # # If you use a modem/ISDN for connections, put "/sbin/SuSEfirewall" in the # 2nd line of "/etc/ppp/ip-up". This is important!! # # *Any* routing between interfaces except masquerading has to set FW_ROUTE to # "yes" and use FW_FORWARD_TCP and/or FW_FORWARD_UDP # # If you just want to do masquerading without filtering, ignore this script # and run this line (exchange "ippp0" with your masquerade/external interface): # ipchains -A forward -j MASQ -i ippp0 # # ------------------------------------------------------------------------ # # 1.) # Should the Firewall be started? # # This setting is done in /etc/rc.config (START_FW="yes") # # 2.) # Which is the interface that points to the internet? # # Enter all the network devices here which are untrusted. # # Choice: any number of devices, seperated by a space # e.g. "eth0", "ippp0 ippp1" # FW_DEV_WORLD="eth1" # # You *may* configure a static IP and netmask to force rule loading even if the # interface is not up and running: set a variable called # FW_DEV_WORLD_[device]="IP_ADDRESS NETMASK" # see below for an example. Otherwise automatic detection is done. # You will still need to set FW_DEV_WORLD first! # #FW_DEV_WORLD_ippp0="10.0.0.1 255.255.255.0" # e.g. for exernal interface ippp0 # # 3.) # Which is the interface that points to the internal network? # # Enter all the network devices here which are trusted. # If you are not connected to a trusted network (e.g. you have just a # dialup) leave this empty. # # Choice: leave empty or any number of devices, seperated by a space # e.g. "tr0", "eth0 eth1" or "" # FW_DEV_INT="eth0" # # You may configure a static IP and netmask to force rule loading even if the # interface is not up and running: set a variable called # FW_DEV_INT_[device]="IP_ADDRESS NETMASK" # see below for an example. Otherwise automatic detection is done. # You will still need to set FW_DEV_INT first! # #FW_DEV_INT_eth0="192.168.1.1 255.255.255.0" # e.g. for internal interface eth0 # # 4.) # Which is the interface that points to the dmz network? # # Enter all the network devices here which point to the dmz. # A "dmz" is a special, seperated network, which is only connected to the # firewall, and should be reachable from the internet to provide services, # e.g. WWW, Mail, etc. and hence are at risk from attacks. # See /usr/share/doc/packages/firewals/EXAMPLES for an example. # # Special note: You have to configure FW_FORWARD_TCP and FW_FORWARD_UDP to # define the services which should be available to the internet and set # FW_ROUTE to yes. # Very special note: servers/networks in FW_MASQ_NETS may access the DMZ to # the same extent they are allowed to access the internet! No FW_FORWARD_* # needed ... # # Choice: leave empty or any number of devices, seperated by a space # e.g. "tr0", "eth0 eth1" or "" # FW_DEV_DMZ="" # # You may configure a static IP and netmask to force rule loading even if the # interface is not up and running: set a variable called # FW_DEV_INT_[device]="IP_ADDRESS NETMASK" # see below for an example. Otherwise automatic detection is done. # You will still need to set FW_DEV_DMZ first! # #FW_DEV_DMZ_eth1="192.168.1.1 255.255.255.0" # e.g. for dmz interface eth1 # # 5.) # Should routing between the internet, dmz and internal network be activated? # REQUIRES: FW_DEV_INT or FW_DEV_DMZ # # You need only set this to yes, if you either want to masquerade internal # machines or allow access to the dmz (or internal machines, but this is not # a good idea). This option supersedes IP_FORWARD from /etc/rc.config! # # Setting this option one alone doesn't do anything. Either activate # massquerading with FW_MASQUERADE below if you want to masquerade your # internal network to the internet, or configure FW_FORWARD_TCP and/or # FW_FORWARD_UDP to define what is allowed to be forwarded! # # Choice: "yes" or "no", defaults to "no" # FW_ROUTE="yes" # # 6.) # Do you want to masquerade internal networks to the outside? # REQUIRES: FW_DEV_INT, FW_ROUTE # # "Masquerading" means that all your internal machines which use services on # the internet seem to come from your firewall. # Please note that it is more secure to communicate via proxies to the # internet than masquerading # # Choice: "yes" or "no", defaults to "no" # FW_MASQUERADE="yes" # # Which internal computers/networks are allowed to access the internet # directly (not via proxys on the firewall)? # Only these networks will be allowed access and will be masqueraded! # # Please note this config changed in firewals-2.3: You may either use just # hosts/nets to allow all traffic from them to the internet, or use an extended # syntax, to restrict internet access to certain services! # # Choice: leave empty or any number of hosts/networks seperated by a space. # Every host/network may get a list of allowed services, otherwise everything # is allowed. A protocol and service is appended by a comma to the host/network. # e.g. "10.0.0.0/8" allows the whole 10.0.0.0 network with unrestricted access # "10.0.1.0/24,tcp,80 10.0.1.0/24,tcp,21" allows the 10.0.1.0 network to use # www/ftp to the internet. "10.0.1.0/24,tcp,1024:65535 10.0.2.0/24" is OK too. # You may NOT set this variable to "0/0" ! # FW_MASQ_NETS="hosts/nets" # # If you want (and you should) you may also set the FW_MASQ_DEV option, to # specify the outgoing interface to masquerade on. (You would normally use # the external interface(s), the FW_DEV_WORLD device(s), e.g. "ippp0") # FW_MASQ_DEV="$FW_DEV_WORLD" # e.g. "ippp0" or "$FW_DEV_WORLD" # # 7.) # Do you want to protect the firewall from the internal network? # REQUIRES: FW_DEV_INT # # If you set this to "yes", internal machines may only access services on # the machine you explicitly allow. They will be also affected from the # FW_AUTOPROTECT_GLOBAL_SERVICES option. # If you set this to "no", any user can connect (and attack) any service on # the firewall. # # Choice: "yes" or "no", defaults to "yes" # FW_PROTECT_FROM_INTERNAL="yes" # # 8.) # Do you want to autoprotect all global running services? # # If set to "yes", all network access to services TCP and UDP on this machine # which are not bound to a special IP address will be prevented (except to # those which you explicitly allow, see below: FW_*_SERVICES_*) # Example: "0.0.0.0:23" would be protected, but "10.0.0.1:53" not. # # Choice: "yes" or "no", defaults to "yes" # FW_AUTOPROTECT_GLOBAL_SERVICES="yes" # "yes" is a good choice # # 9.) # Which services on the firewall should be accessible from either the internet # (or other untrusted networks), the dmz or internal (trusted networks)? # # Enter all ports or known portnames below, seperated by a space. # TCP services (e.g. SMTP, WWW) must be set in FW_SERVICES_*_TCP, and # UDP services (e.g. syslog) must be set in FW_SERVICES_*_UDP. # e.g. if a webserver on the firewall should be accessible from the internet: # FW_SERVICES_EXTERNAL_TCP="www" # e.g. if the firewall should receive syslog messages from the dmz: # FW_SERVICES_DMZ_UDP="syslog" # # Choice: leave empty or any number of ports, known portnames (from # /etc/services) and port ranges seperated by a space. Port ranges are # written like this, from 1 to 10: "1:10" # e.g. "", "smtp", "123 514", "3200:3299", "ftp 22 telnet 512:514" # FW_SERVICES_EXTERNAL_TCP="SMTP POP" # Common: smtp domain FW_SERVICES_EXTERNAL_UDP="" # Common: domain # FW_SERVICES_DMZ_TCP="" # Common: smtp domain FW_SERVICES_DMZ_UDP="syslog" # Common: domain syslog # FW_SERVICES_INTERNAL_TCP="ssh smtp pop www irc icq news rplayer" # Common: ssh smtp domain FW_SERVICES_INTERNAL_UDP="" # Common: domain # # 10.) # Which services should be accessible from trusted hosts/nets on the internet? # # Define trusted networks on the internet, and the TCP and/or UDP services # they are allowed to use. # # Choice: leave FW_TRUSTED_NETS empty or any number of computers and/or # networks, seperated by a space. e.g. "172.20.1.1", "172.20.0.0/16" # FW_TRUSTED_NETS="" # # leave FW_SERVICES_TRUSTED_* empty or any number of ports, known portnames # (from /etc/services) and port ranges seperated by a space. # e.g. "25", "ssh", "1:65535", "1 3:5" # FW_SERVICES_TRUSTED_TCP="ssh" # Common: ssh FW_SERVICES_TRUSTED_UDP="syslog time ntp" # Common: syslog time ntp # # 11.) # How is access allowed to high (unpriviliged [above 1023]) ports? # # You may either allow everyone from anyport access to your highports ("yes"), # disallow anyone ("no"), anyone who comes from a defined port (portnumber or # known portname) [note that this is easy to circumvent!], or just your # defined nameservers ("dns"). # Note that if you want to use normal (active) ftp, you have to set the TCP # option to ftp-data. If you use passive ftp, you don't need that. # Note that you can't use rpc requests (e.g. rpcinfo, showmount) as root # from a firewall using this script. # # Choice: "yes", "no", "dns", portnumber or known portname, defaults to "no" # FW_ALLOW_INCOMING_HIGHPORTS_TCP="ftp-data" # Common: "ftp-data" (sadly!) FW_ALLOW_INCOMING_HIGHPORTS_UDP="dns" # Common: "dns" # # 12.) # Are you running some of the services below? # They need special attention - otherwise they won´t work! # # Set services you are running to "yes", all others to "no", defaults to "no" # FW_SERVICE_DNS="no" # if yes, FW_SERVICES_*_TCP needs to have port 53 # (or "domain") set to allow incoming queries. # also FW_ALLOW_INCOMING_HIGHPORTS_UDP needs to be "yes" FW_SERVICE_DHCLIENT="yes" # if you use dhclient to get an ip address # you have to set this to "yes" ! FW_SERVICE_DHCPD="no" # set to "yes" if this server is a DHCP server FW_SERVICE_SAMBA="yes" # set to "yes" if this server uses samba as client # or server. As a server, you still have to set # FW_SERVICES_{WORLD,DMZ,INT}_TCP="139" # Everyone may send you udp 137/138 packets if set # to yes! # (note: samba on the firewall is not a good idea!) # # 13.) # Which services accessed from the internet should be allowed to the # dmz (or internal network - if it is not masqueraded)? # REQUIRES: FW_ROUTE # # With this option you may allow access to e.g. your mailserver. The # machines must have valid, non-private, IP addresses which were assigned to # you by your ISP. This opens a direct link to your network, so only use # this option for access to your dmz!!!! # # Choice: leave empty (good choice!) or use the following explained syntax # of forwarding rules, seperated each by a space. # A forwarding rule consists of 1) source IP/net, 2) destination IP (dmz/intern) # and 3) destination port, seperated by a comma (","), e.g. # "4.0.0.0/8,1.1.1.1,22", # "4.4.4.4/12,20.20.20.20,22 12.12.12.12/12,20.20.20.20,22" # FW_FORWARD_TCP="" # Beware to use this! FW_FORWARD_UDP="" # Beware to use this! # # 14.) # Which services accessed from the internet should be allowed to masqueraded # servers (on the internal network or dmz)? # REQUIRES: FW_ROUTE, FW_MASQUERADE # # With this option you may allow access to e.g. your mailserver. The # machines must be in a masqueraded segment and may not have public IP addesses! # # Please note that this should *not* be used for security reasons! You are # opening a hole to your precious internal network. If e.g. the webserver there # is compromised - your full internal network is compromised!! # # Choice: leave empty (good choice!) or use the following explained syntax # of forward masquerade rules, seperated each by a space. # A forward masquerade rule consists of 1) source IP/net, 2) destination IP # (dmz/intern) and 3) destination port, seperated by a comma (","), e.g. # "4.0.0.0/8,1.1.1.1,22", # "4.4.4.4/12,20.20.20.20,22 12.12.12.12/12,20.20.20.20,22" # FW_FORWARD_MASQ_TCP="" # Beware to use this! FW_FORWARD_MASQ_UDP="" # Beware to use this! # # 15.) # Which accesses to services should be redirected to a localport on the # firewall machine? # # This can be used to force all internal users to surf via your squid proxy, # or transparently redirect incoming webtraffic to a secure webserver. # # Choice: leave empty or use the following explained syntax of redirecting # rules, seperated by a space. # A redirecting rule consists of 1) source IP/net, 2) destination IP/net, # 3) original destination port and 4) local port to redirect the traffic to, # seperated by a colon. e.g. "10.0.0.0/8,0/0,80,3128 0/0,172.20.1.1,80,8080" # FW_REDIRECT_TCP="" FW_REDIRECT_UDP="" # # 16.) # Which logging level should be enforced? # You can define to log packets which were accepted or denied. # You can also the set log level, the critical stuff or everything. # Note that logging *_ALL is only for debugging purpose ... # # Choice: "yes" or "no", FW_LOG_*_CRIT defaults to "yes", # FW_LOG_*_ALL defaults to "no" # FW_LOG_DENY_CRIT="yes" FW_LOG_DENY_ALL="no" FW_LOG_ACCEPT_CRIT="yes" FW_LOG_ACCEPT_ALL="no" # # 17.) # Do you want to enable additional kernel TCP/IP security features? # If set to yes, some obscure kernel options are set. # (icmp_ignore_bogus_error_responses, icmp_echoreply_rate, # icmp_destunreach_rate, icmp_paramprob_rate, icmp_timeexeed_rate, # ip_local_port_range, log_martians, mc_forwarding, mc_forwarding, # rp_filter, routing flush) # Tip: Set this to "no" until you have verified that you have got a # configuration which works for you. Then set this to "yes" and keep it # if everything still works. (It should!) ;-) # # Choice: "yes" or "no", defaults to "yes" # FW_KERNEL_SECURITY="yes" # # 18.) # Keep the routing set on, if the firewall rules are unloaded? # REQUIRES: FW_ROUTE # # If you are using diald, or automatic dialing via ISDN, if packets need # to be sent to the internet, you need to turn this on. The script will then # not turn off routing and masquerading when stopped. # You *might* also need this if you have got a DMZ. # Please note that this is *insecure*! If you unload the rules, but are still # connected, you might your internal network open to attacks! # The better solution is to remove "/sbin/SuSEfirewall stop" or # "/sbin/init.d/firewall stop" from the ip-down script! # # # Choices "yes" or "no", defaults to "no" # FW_STOP_KEEP_ROUTING_STATE="no" # # 19.) # Allow (or don't) ICMP echo pings on either the firewall or the dmz from # the internet? # REQUIRES: FW_ROUTE for FW_ALLOW_PING_DMZ # # Choice: "yes" or "no", defaults to "no" # FW_ALLOW_PING_FW="yes" FW_ALLOW_PING_DMZ="no" ## # END of rc.firewall ## # # #-------------------------------------------------------------------------# # # # EXPERT OPTIONS - all others please don't change these! # # # #-------------------------------------------------------------------------# # # # # 20.) # Allow (or don't) ICMP time-to-live-exceeded to be send from your firewall. # This is used for traceroutes to your firewall (or traceroute like tools). # # Please note that the unix traceroute only works if you say "yes" to # FW_ALLOW_INCOMING_HIGHPORTS_UDP, and windows traceroutes only if you say # "yes" to FW_ALLOW_PING_FW # # Choice: "yes" or "no", defaults to "no" # FW_ALLOW_FW_TRACEROUTE="no" # # 21.) # Allow ICMP sourcequench from your ISP? # If set to yes, the firewall will notice when connection is choking, however # this opens yourself to a denial of service attack. Choose your poison. # # Choice: "yes" or "no", defaults to "yes" # FW_ALLOW_FW_SOURCEQUENCH="yes" # # 22.) # Which masquerading modules should be loaded? # REQUIRES: FW_ROUTE, FW_MASQUERADE # # (omit the path or "ip_masq_" prefix as well as the ".o" suffix!) # FW_MASQ_MODULES="autofw cuseeme ftp irc mfw portfw quake raudio user vdolive" Thank you for your help and I appologise for the long message length. -- Ralph F. De Witt ----------------------------------------------------------------------------- PROUD USER OF SuSE LINUX 7.0 It said uses Windows 98 or better so I installed SuSE 7.0 Register Linux User 168814 ICQ #49993234 "Beer is proof that God loves us and wants us to be happy" -Benjamin Franklin -----------------------------------------------------------------------------

3 2

DNS and DHCP connection
by suse＠diplan.de 19 Nov '00

19 Nov '00

Hi, Sorry for the beginners question, but how does my DNS server get to know about leased DHCP IP <-> hostname information? If the IP address changes from lease to lease the DNS has to be informed by that. I looked into some manpages and howtos but did not find the information. Thank you very much for a hint and best regards, Rainer

5 4

Re: [suse-security] name resolution by ipchains
by Volker Kuhlmann 19 Nov '00

19 Nov '00

> is it possible to have ipchains resolving internet names? > If yes, how? If you mean that you'd like to be able to specify a nme instead of an IP, yes ipchains will resolve the name. However, unless the name can be resolved with the local /etc/hosts file, the DNS will be queried, and if that happen to be blocked just at that time (either because it's not yet enabled, or you never allow it through), ipchains will wait for a rather long time... Volker

1 0

name resolution by ipchains
by Philipp Snizek 18 Nov '00

18 Nov '00

Hi all is it possible to have ipchains resolving internet names? If yes, how? Thank you Philipp

2 1