That sounds very interesting. I have a customer using AstLinux as Proxy as well.
Maybe we could document this in the Wiki? 

> Am 27.09.2021 um 02:54 schrieb Lonnie Abelbeck <li...@lonnie.abelbeck.com>:
> 
> Michael,
> 
> The /mnt/kd/arno-iptables-firewall/custom-rules is a basic shell script, so 
> parsing sip.conf using 'sed' or such should be reasonably straightforward.
> 
> BTW, for extra credit, if you combined all the allowed SIP IPs into an ipset 
> (ex. udp_sip_hosts), you can very efficiently match all of them with only one 
> rule:
> --
> iptables -A EXT_INPUT_CHAIN -m set --match-set udp_sip_hosts src -p udp 
> --dport 5060 -j ACCEPT
> --
> That would allow you to rebuild only the "udp_sip_hosts" ipset when the 
> sip.conf got changed, without rebuilding the firewall.  Though requires some 
> 'ipset' command knowledge, though not complex at all.
> 
> Example 'ipset' usage in AstLinux:
> https://github.com/astlinux-project/astlinux/blob/d95ba9c3914b135da4440cb95f32af61a41d4650/package/arnofw/aif/bin/arno-iptables-firewall#L4275
> 
> If you only use IPv4 a lot of the example can be simplified.
> 
> Lonnie
> 
> 
> 
>> On Sep 26, 2021, at 7:17 PM, Michael Knill 
>> <michael.kn...@ipcsolutions.com.au> wrote:
>> 
>> Thanks Lonnie.
>> 
>> Actually now that I think about it, is there any reason why the custom rule 
>> could not parse sip.conf for host=<IP Address> and open up all Public IP's?
>> It would mean that you would need to restart the firewall every time you 
>> modified sip.conf but I'm sure we could build this into our portal very 
>> simply.
>> 
>> Regards
>> Michael Knill
>> 
>> On 27/9/21, 9:47 am, "Lonnie Abelbeck" <li...@lonnie.abelbeck.com> wrote:
>> 
>>   Hi Michael,
>> 
>>   With 300 rules and the same across all your boxes, I would use 
>> /mnt/kd/arno-iptables-firewall/custom-rules to define these.
>> 
>>   Very similar to the deny_ext_local() example I posted recently, but the 
>> reverse ... pass_ext_local() using -j ACCEPT
>> 
>>   Without testing, something like ...
>>   --
>>   pass_ext_local()
>>   {
>>     local proto="$1" host="$2" port="$3"
>> 
>>     echo "[CUSTOM RULE] Pass EXT->Local for Proto: $proto, Host: $host, 
>> Port: $port"
>>     iptables -A EXT_INPUT_CHAIN -s $host -p $proto --dport $port -j ACCEPT
>>   }
>>   ## uncomment to enable ##
>>   #pass_ext_local udp 1.2.3.4 5060
>>   #pass_ext_local tcp 1.2.3.0/24 5061
>>   --
>> 
>>   If you only use udp/5060, you could simplify things, maybe only one "echo" 
>> statement and a variable defining all 300 IPs.  Generic shell scripting.
>> 
>>   Again untested ...
>>   --
>>   pass_ext_local_udp_sip()
>>   {
>>     local host proto="udp" port="5060" IFS
>>     local sip_hosts="1.2.3.4 1.22.33.40 1.22.33.41 1.22.33.42 1.22.33.43 
>> 1.22.33.44 1.22.33.45 1.22.33.46 1.22.33.47 1.22.33.48"
>> 
>>     echo "[CUSTOM RULE] Pass EXT->Local for UDP/5060 SIP Hosts"
>>     unset IFS
>>     for host in $sip_hosts; do
>>       iptables -A EXT_INPUT_CHAIN -s $host -p $proto --dport $port -j ACCEPT
>>     done
>>   }
>>   pass_ext_local_udp_sip
>>   --
>> 
>>   Alternatively, you could define the sip_hosts variable with a file if 
>> desired.
>> 
>>   Lonnie
>> 
>> 
>> 
>> 
>> 
>>> On Sep 26, 2021, at 5:32 PM, Michael Knill 
>>> <michael.kn...@ipcsolutions.com.au> wrote:
>>> 
>>> Hi Group
>>> 
>>> I'm looking to have a large number of firewall entries in Astlinux e.g. 
>>> 300. They would be all the same e.g. I want to open port 5060 from multiple 
>>> sites.
>>> Is there an easier/neater way to do this other than lots of firewall 
>>> entries in the Firewall Tab?
>>> 
>>> Regards
>>> 
>>> Michael Knill
>>> Managing Director
>>> 
>>> D: +61 2 6189 1360
>>> P: +61 2 6140 4656
>>> E: michael.kn...@ipcsolutions.com.au
>>> W: ipcsolutions.com.au


Michael

http://www.mksolutions.info





_______________________________________________
Astlinux-users mailing list
Astlinux-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/astlinux-users

Donations to support AstLinux are graciously accepted via PayPal to 
pay...@krisk.org.

Reply via email to