OK. Several things to consider here. Comments follow each of your steps below.

At 04:48 PM 12/6/02 -0500, Haines Brown wrote:
Well, I'm afraid I didn't get very far in straightening out my
iptables so that I could move tcpip packets.

I should mention that I'm running RedHat 8.0. I use pop3 to get email
and sendmail to send mail. I don't care to be pinged, and I'll do
without telnet. No one needs to access my standalone machine for FTP
etc.

1. First, I verify that INPUT chain's policy is set to ACCEPT:

Chain INPUT (policy ACCEPT 0 packets, 0 bytes)
pkts bytes target prot opt in out source destination
150 10348 RH-Lokkit-0-50-INPUT all -- * * 0.0.0.0/0 0.0.0.0/0

This lokkit is a Firewall Configuration utility. It starts by saying
that "high security blocks all incoming accesses." Is that my
problem?
Maybe. That's what we're trying to test out. But the fetchmail tests are not "incomming accesses"; they involve replies to outgoing connections. Same for the ping tests.

2. I flushed all rules for the INPUT chain:

   # iptables -F INPUT
Right.

3. This had no feedback, and so I looked at my iptables:

  # iptables -nvL
  Chain INPUT (policy ACCEPT 0 packets, 0 bytes)
    pkts bytes target prot opt in     out     source  destination

  Chain FORWARD (policy ACCEPT 0 packets, 0 bytes)
    pkts bytes target prot opt in     out     source  destination

  Chain OUTPUT (policy ACCEPT 150 packets, 10348 bytes)
    pkts bytes target prot opt in     out     source   destination

  Chain RH-Lokkit-0-50-INPUT (0 references)
   pkts bytes target  prot opt in  out source          destination
      0     0 ACCEPT  udp  --  *    *   206.141.193.55  0.0.0.0/0
                udp spt:53 dpts:1025:65535
      0     0 ACCEPT  udp  --  *    *   206.73.20.40    0.0.0.0/0
                udp spt:53 dpts:1025:65535
      0     0 ACCEPT  udp  --  eth0 *   0.0.0.0/0       0.0.0.0/0
                udp spts:67:68 dpts:67:68
      0     0 ACCEPT  udp  --  eth1 *   0.0.0.0/0       0.0.0.0/0
                udp spts:67:68 dpts:67:68
    150 10348 ACCEPT  all  --  lo   *   0.0.0.0/0       0.0.0.0/0
      0     0 REJECT  tcp  --  *    *   0.0.0.0/0       0.0.0.0/0
                tcp flags:0x16/0x02 reject-with icmp-port-unreachable
      0     0 REJECT  udp  --  *    *   0.0.0.0/0       0.0.0.0/0
                udp reject-with icmp-port-unreachable

  Am I right to infer that since under the INPUT CHAIN, nothing is
  listed, that the flush succeeded? I guess so, for Lokkit is gone.
Yes.

4. Next, I connected with my DSL provider using rp-pppoe:

        # adsl-start
        . . . .Connect

5. To check the result, I ran ifconfig -a, and I see that eth0, lo, and
   ppp0 are all up, and ppp0 has been assigned an address.

  # ifconfig -a
  ppp0      Link encap:Point-to-Point Protocol
    inet addr:64.252.164.246  P-t-P:64.252.160.1  Mask:255.255.255.255
At this point, it would be wise to check "iptables -nvL" again. Just to see if bringing ppp0 up caused execution of a script that changed your firewall settings.

6. I try # fetchmail. No go. Same behavior as before. Fetchmail
   accesses the mail server, reports the number of messages waiting,
   but hangs when it comes to downloading the first.
OK. This *may* still be a firewall problem (see item #5). But if it is not, it may also be a fetchmail problem. Check your logs for any messages from fetchmail. And describe "hangs" in more detail.

7. I tried some pings. Same as before:
   a. I ping my ppp0 address (64.252.164.246). Works fine.
   b. Then my DNS: # ping 206.73.20.40. While no returns, the server
   might block pings.
   c. I ping hardware.redhat.com.  Nothing comes back back.
   d. I ping aol.com. Every five seconds a return. About 30 ms round
   trip, which I guess is normal.
The problem with ping tests is that many servers are set not to respond to pings. I get the same result from hardware.redhat.com that you do. (This is why I always suggest you ping a known-good address, and I often even suggest using mine, since I know it is set to answer pings.) But since the other three work, I'd say you are OK with respect to ping.

In sum, unless I somehow failed to flush rules for the INPUT chain,
while my iptables were messed up, that is not the basic problem.
Agreed, subject to your doing the check of item #5. In fact, do this *after* you have a fetchmail failure, since the report includes a listing of how many packets each rule or policy processed.

I'm using the same box, and simply swapping hard disks, and so the
only hardware that changes is the hard disk. I have a high software
firewall, and have run Bastille so set up certain rules, but not
affecting iptables.
I don't know what this ("certain rules")  means.

I also recall that my initial installation of RH8.0 resulted in a
broken rp-pppoe, which I replaced/upgraded with happier results. So I
wonder if something else might be broken. I'll try running
/usr/sbin/lokkit with either No Firewall or Medium security to see if
that helps. Medium security is what I have on my present machine, that
can receive tcp mail packets.
By all means try. But you can't set a firewall lower than "ACCEPT everything", so (in a one-host setting, where NAT is not an issue) changing the lokkit settings is unlikely to help (always assuming the test in #5 doesn't turn up something). I'd check to see how the two fetchmail setups differ.


--
-------------------------------------------"Never tell me the odds!"--------
Ray Olszewski -- Han Solo
Palo Alto, California, USA [EMAIL PROTECTED]
-------------------------------------------------------------------------------

-
To unsubscribe from this list: send the line "unsubscribe linux-newbie" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.linux-learn.org/faqs

Reply via email to