I run into a problem when I tried applying my new firewall module
based config on RedHat version 5 systems... the problem is that th
iptables-save behaves differently than the one on RedHat 6:

# /sbin/iptables --version
iptables v1.3.5
# /sbin/iptables-save | grep SSH
-A INPUT -s 10.128.0.0/255.128.0.0 -p tcp -m multiport --dports 22 -m
comment --comment "010 allow SSH traffic from subnet 10.128.0.0/9" -m
state --state NEW -j ACCEPT


I added this issue to : https://projects.puppetlabs.com/issues/10723 ,
but am not sure it is their bug to post to.

Basically iptables-save does not report the IP addresses in the format
expected by Puppet. On newer RedHat 6, it works:

# iptables --version
iptables v1.4.7

# iptables-save |grep SSH
-A INPUT -s 10.128.0.0/9 -p tcp -m multiport --dports 22 -m comment
--comment "010 allow SSH traffic from subnet 10.128.0.0/9" -m state
--state NEW -j ACCEPT

Am wondering if there is a workaround to this issue?

Thanks a lot,
Mohamed.

On Wed, Nov 30, 2011 at 1:10 PM, Ken Barber <k...@puppetlabs.com> wrote:
> Agreed.
>
> On Wed, Nov 30, 2011 at 5:31 PM, Mohamed Lrhazi <lrh...@gmail.com> wrote:
>> Right. hostnames are attractive.... ideal use case:
>>
>> - Allow all from nagios.my.dom.ain
>>
>> If someone changes the IP address of the monitor server, puppet would
>> find out and fix the rule on all the monitored nodes!
>>
>> Maybe such a simplistic and ideal case is rare in the real world, but
>> it is attractive.
>>
>> Regarding DNS being reconfigured, I think that as long as we
>> understand exactly how puppet would be impacted, it should be OK. DNS
>> is not supposed to be misconfigured and puppet failure could be seen
>> as a healthy event.
>>
>> Thanks,
>> Mohamed.
>>
>> On Wed, Nov 30, 2011 at 11:30 AM, Ken Barber <k...@puppetlabs.com> wrote:
>>> Yeah - I'm glad you spotted that I was just about to point it out.
>>>
>>> Its a hairy problem - Jonathan has submitted a patch for it but its
>>> stil in review. Generally the feeling is that having iptables use
>>> hostnames can be flakey at times - especially if you try to add a rule
>>> for a host where DNS is misconfigured or broken for some reason. Maybe
>>> the hostname doesn't exist yet for example. However - I realise some
>>> people want to do this regardless of the risks so we are looking to
>>> fix it.
>>>
>>> Anyway - the solution Jonathan has provided is looking reasonably good
>>> and should be merged in soon.
>>>
>>> ken.
>>>
>>> On Wed, Nov 30, 2011 at 3:34 PM, Mohamed Lrhazi <lrh...@gmail.com> wrote:
>>>> Just saw this bug report: http://projects.puppetlabs.com/issues/10723
>>>>
>>>> Sorry. Thanks.
>>>>
>>>> On Tue, Nov 29, 2011 at 8:50 PM, Mohamed Lrhazi <lrh...@gmail.com> wrote:
>>>>> Hello,
>>>>>
>>>>> The source and destination parameters accept both IP address or a
>>>>> hostname. If using a hostname, the firewall module thinks the rule
>>>>> changed each time it runs reporting:
>>>>>
>>>>>
>>>>> notice: /Firewall[300 allow netbackup traffic from
>>>>> nbmaster2-63.example.com]/source: current_value 192.168.63.42/32,
>>>>> should be nbmaster2-63.example.com (noop)
>>>>>
>>>>>
>>>>> Is there an easy workaround to this? other than not using hostnames?
>>>>>
>>>>> A similar issue is also seen with the value of debug-level. From some
>>>>> reason it always thinks it needs to be reset:
>>>>>
>>>>> notice: /Firewall[998 drop noisy local traffic]/log_level:
>>>>> current_value , should be warning (noop)
>>>>>
>>>>>
>>>>> # Log everything else, then reject it with the default deny rule
>>>>> firewall { '998 drop noisy local traffic':
>>>>>    state       => 'NEW',
>>>>>    log_level   => warning,
>>>>>    jump        => 'LOG',
>>>>> }
>>>>>
>>>>> iptables -nL shows this rule as:
>>>>>
>>>>> LOG        tcp  --  0.0.0.0/0            0.0.0.0/0           /* 998
>>>>> drop noisy local traffic */ state NEW LOG flags 0 level 4
>>>>>
>>>>> I tried setting "log_level" to 4, instead of "warning" and got :
>>>>>
>>>>> notice: /Firewall[998 drop noisy local traffic]/log_level:
>>>>> current_value , should be 4 (noop)
>>>>>
>>>>>
>>>>> Thanks a lot.
>>>>> Mohamed.
>>>>>
>>>>
>>>> --
>>>> You received this message because you are subscribed to the Google Groups 
>>>> "Puppet Users" group.
>>>> To post to this group, send email to puppet-users@googlegroups.com.
>>>> To unsubscribe from this group, send email to 
>>>> puppet-users+unsubscr...@googlegroups.com.
>>>> For more options, visit this group at 
>>>> http://groups.google.com/group/puppet-users?hl=en.
>>>>
>>>
>>> --
>>> You received this message because you are subscribed to the Google Groups 
>>> "Puppet Users" group.
>>> To post to this group, send email to puppet-users@googlegroups.com.
>>> To unsubscribe from this group, send email to 
>>> puppet-users+unsubscr...@googlegroups.com.
>>> For more options, visit this group at 
>>> http://groups.google.com/group/puppet-users?hl=en.
>>>
>>>
>>
>> --
>> You received this message because you are subscribed to the Google Groups 
>> "Puppet Users" group.
>> To post to this group, send email to puppet-users@googlegroups.com.
>> To unsubscribe from this group, send email to 
>> puppet-users+unsubscr...@googlegroups.com.
>> For more options, visit this group at 
>> http://groups.google.com/group/puppet-users?hl=en.
>>
>
> --
> You received this message because you are subscribed to the Google Groups 
> "Puppet Users" group.
> To post to this group, send email to puppet-users@googlegroups.com.
> To unsubscribe from this group, send email to 
> puppet-users+unsubscr...@googlegroups.com.
> For more options, visit this group at 
> http://groups.google.com/group/puppet-users?hl=en.
>

-- 
You received this message because you are subscribed to the Google Groups 
"Puppet Users" group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to 
puppet-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/puppet-users?hl=en.

Reply via email to