Firewall/Mailman/Routing Help!

[Sp00n]

Ok, the situation is as follows, I have a mailman server internally on our network which has been working fine up until saturday when we had a powercut, since then I cannot communicate to it externally on any port.

I've checked the firewall rules for 25 and also setup port 80 to test the apache on there but neither of these work. I've checked selinux which is disabled and I've rebooted again to no avail.

However, I can ssh/telnet/http to it on the same subnet on my local lan and it works fine but not from outside of the firewall, but it's not the firewall (as far as i can tell) because I have RDP rules set up amoungst others which are working fine, same reason why I don't think it's the routing.

I'm a linux noob and once again I've been thrown in at the deep end, does anyone have any pointers?

/edit firewall is debian running shoreline, mailman is also a debian box.

 

[Sp00n]

I can't remember if shoreline uses iptables. Type iptables -L -v and see what it comes up with.

As above, can you ping the firewall? Can the firewall ping the debian box?

What are the logs saying? cat /var/log/*logfile* should show you something.

Both firewall and mailman can see each other, I'm looking through the logs now and I can't see anything particularly obvious.

Netstat shows me:

tcp 0 0 *:smtp *:* LISTEN

Shoreline supercedes iptables afaik, but I checked and iptables is not enabled on the firewall anyway.

Grrr, I can't think of anything else, the routes seem fine, the firewall seems fine so I don't know what to do!!

Checked the gateway, that's fine and I can ping google via it's ip address no problem, wtf!!

 

[Sp00n]

Is the firewall and mail server on the same subnet ... you haven't lost a default gateway entry somewhere due to the restart have you, (you'd still be able to get on from within the same subnet though).

I've seen that happen before now.

I could kiss you, that looks to have sorted it

Why didn't I bloomin think of that one!

er... its gone again, wtf!

And now its back again, wtf!

How can something work one minute and the not the next?

/edit again: possible cause was 2 default gw's, one of which was my local computer, god knows how that managed to get in to the routing table!