PDA

View Full Version : yum update caused weird network problems


derzok
19th November 2007, 08:40 PM
So I installed Fedora Core 6 a while ago on my server. I have a decently static IP so I have a domain name for it. I recently did a

yum update


It took about an hour to download all the packages (a few hundred of them, actually). They all installed fine with no error messages. The only things to note were

Updating: pam [17/493]warning: /etc/pam.d/system-auth created as /etc/pam.d/system-auth.rpmnew
Updating: postfix [ 56/493]warning: /etc/postfix/main.cf created as /etc/postfix/main.cf.rpmnew
Updating: postfix [ 56/493]warning: /usr/lib/sasl2/smtpd.conf created as /usr/lib/sasl2/smtpd.conf.rpmnew
Updating: sendmail [116/493]warning: /etc/mail/sendmail.cf created as /etc/mail/sendmail.cf.rpmnew


None of those looked too crucial.

So now when I access my website via the url http://zoklet.net it won't load.

I assumed SELinux decided to **** with me again so I did a

setenforce 0

just to make sure. No luck. nmap says that all the ports are filtered. However when I run nmap on the box's lan IP it says all the necessary ports are open.

The server's local IP has not changed, my outside IP has not changed, and all the ports are still forwarded properly. I am thinking that it's got something to do with the upgrade, but I can't figure out what.

Bottom line:
The problem is that I can not access my server from inside OR outside the network using my IP address OR the domain name. My server address is http://69.135.212.19/ and nobody can see it. The ports are forwarded properly on all of my routers. Something is blocking my server. When I access the server with it's local IP: http://192.168.0.12/ it works fine.

Edit: I didn't look in the rules, but if there's a rule against posting IP addresses, I'll gladly remove mine.

stevea
19th November 2007, 09:13 PM
So what does the network config look like at the moment. Look at "ip addr" and "ip route" and "cat /etc/resolv.conf"

derzok
19th November 2007, 10:37 PM

[root@zoklet zok]# /sbin/ifconfig
eth1 Link encap:Ethernet HWaddr 00:E0:18:A2:6A:41
inet addr:192.168.0.12 Bcast:192.168.0.255 Mask:255.255.255.0
inet6 addr: fe80::2e0:18ff:fea2:6a41/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:4817 errors:0 dropped:0 overruns:0 frame:0
TX packets:4001 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:524101 (511.8 KiB) TX bytes:878481 (857.8 KiB)
Interrupt:20 Base address:0x2000

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:2338 errors:0 dropped:0 overruns:0 frame:0
TX packets:2338 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:2856472 (2.7 MiB) TX bytes:2856472 (2.7 MiB)

[root@zoklet zok]# cat /etc/resolv.conf
; generated by /sbin/dhclient-script
nameserver 65.24.7.3
nameserver 65.24.7.6

stevea
19th November 2007, 11:47 PM
Not exactly what I asked for; ifconfig is deprecated, please learn "ip".

(repeat) "ip route" ?


Can you browse out ?
What does "iptables -L" report ?
What does "netstat -lntp" report ?
What does "nmap 127.0.0.1" report ?
--

FWIW I've got an FC6 server and I've had no recent problems. I assume you've tried re-booting.

derzok
20th November 2007, 12:49 AM
The server is able to make outbound connections. I'm using it to run irssi right now. I also am able to browse the web with lynx. I assume other protocols work as well.


[root@zoklet ~]# ip route
192.168.0.0/24 dev eth1 proto kernel scope link src 192.168.0.12
169.254.0.0/16 dev eth1 scope link
default via 192.168.0.1 dev eth1

[root@zoklet ~]# iptables -L
Chain INPUT (policy ACCEPT)
target prot opt source destination

Chain FORWARD (policy ACCEPT)
target prot opt source destination

Chain OUTPUT (policy ACCEPT)
target prot opt source destination

[root@zoklet ~]# netstat -lntp
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address Foreign Address State PID/Program name
tcp 0 0 127.0.0.1:2208 0.0.0.0:* LISTEN 2511/hpiod
tcp 0 0 192.168.0.12:2401 0.0.0.0:* LISTEN 2554/xinetd
tcp 0 0 0.0.0.0:3306 0.0.0.0:* LISTEN 2666/mysqld
tcp 0 0 0.0.0.0:139 0.0.0.0:* LISTEN 2789/smbd
tcp 0 0 0.0.0.0:111 0.0.0.0:* LISTEN 2292/portmap
tcp 0 0 0.0.0.0:21 0.0.0.0:* LISTEN 2567/vsftpd
tcp 0 0 127.0.0.1:631 0.0.0.0:* LISTEN 2528/cupsd
tcp 0 0 0.0.0.0:445 0.0.0.0:* LISTEN 2789/smbd
tcp 0 0 127.0.0.1:2207 0.0.0.0:* LISTEN 2516/python
tcp 0 0 0.0.0.0:799 0.0.0.0:* LISTEN 2313/rpc.statd
tcp 0 0 :::993 :::* LISTEN 2696/dovecot
tcp 0 0 :::995 :::* LISTEN 2696/dovecot
tcp 0 0 :::110 :::* LISTEN 2696/dovecot
tcp 0 0 :::143 :::* LISTEN 2696/dovecot
tcp 0 0 :::80 :::* LISTEN 2726/httpd
tcp 0 0 :::22 :::* LISTEN 2542/sshd
tcp 0 0 :::443 :::* LISTEN 2726/httpd

[root@zoklet ~]# nmap 127.0.0.1

Starting Nmap 4.11 ( http://www.insecure.org/nmap/ ) at 2007-11-19 19:49 EST
Interesting ports on localhost.localdomain (127.0.0.1):
Not shown: 1666 closed ports
PORT STATE SERVICE
21/tcp open ftp
22/tcp open ssh
80/tcp open http
110/tcp open pop3
111/tcp open rpcbind
139/tcp open netbios-ssn
143/tcp open imap
443/tcp open https
445/tcp open microsoft-ds
631/tcp open ipp
799/tcp open controlit
993/tcp open imaps
995/tcp open pop3s
3306/tcp open mysql

Nmap finished: 1 IP address (1 host up) scanned in 0.114 seconds


Also tried rebooting the server, router, and modem - still no luck.

stevea
20th November 2007, 08:23 AM
tcp 0 0 :::80 :::* LISTEN 2726/httpd

Well that looks good. Do you get a connection when you browse to http://127.0.0.1 or http://192.168.0.12 ?
If these both work then it points the finger at your router (or whatever sits between your system and the public IP).

You could try running wireshark while making an external connection - to see what packets appear.

best wishes

hiberphoptik
20th November 2007, 08:29 AM
seems obvious but i didnt see you mention it above... check the linux's firewall config and make sure the firewall is off, then if it works you know it was the firewall, turn it back on and open the proper ports in the firewall :)

the router cant forward to the server if the server's software firewall (IP Tables) is blocking

might not be the firewall but i find that 9 times out of 10 its the firewall and 9 times out of 10 people say "the firewall is disabled" but then when they finally double check it... its enabled and blocking :)

derzok
20th November 2007, 01:22 PM
What is the linux firewall executable called and how can I disable it from command line? The only info I found online told me how to access it through KDE's menus, but that doesn't help if you don't have a monitor, mouse, or keyboard.

Edit: I just ran "service --status-all" and found:

hpssd (pid 2516) is running...
httpd (pid 2780 2776 2775 2774 2773 2772 2771 2770 2726) is running...
Firewall is stopped.
Table: filter
Chain INPUT (policy ACCEPT)
num target prot opt source destination

Chain FORWARD (policy ACCEPT)
num target prot opt source destination

Chain OUTPUT (policy ACCEPT)
num target prot opt source destination


Particularly, the line that says "firewall is stopped" - still no luck. I'll be able to set up another server on the network in just a few hours - I will see what happens when I set that machine as the dmz. If it works, then we can eliminate the router/modem. I don't know why I didn't try that immediately.

derzok
20th November 2007, 02:04 PM
Update: I just installed apache on a windows machine and set it as my router's DMZ. The apache test page showed when I went to both my IP and to my domain name. It's definitely a problem with the server, not the router or modem.

Edit: Oddly enough, when I switched DMZ back to my linux machine, it worked fine. I can now access my server from both the IP and domain name - though I'm not sure why... Thanks for all your help you guys! I wish we knew what happened :P