-   Debian (
-   -   Server refusing all connections (

Brodi816 04-18-2012 02:55 AM

Server refusing all connections
My server is refusing to establish any connection from any port including ssh. The problem started Saturday, I had the server rebooted but the problem has come up again. I currently have a ticket submitted to get the server rebooted again. I'm using debian 6 and I'm wondering where I should start looking for the problem once my server comes back online.

pan64 04-18-2012 03:10 AM

do you have outgoing traffic, or it has no network at all?

Brodi816 04-18-2012 04:13 AM

There's no traffic at all, any connection made to the server is immediately closed.

pan64 04-18-2012 04:17 AM

in /var/log/messages you may find error messages
you need to check the network configuration, but as you told they were not changed at all. it looks like a hw failure.

Brodi816 04-18-2012 03:57 PM

So the server was rebooted and now it's not accepting connections and they're just timing out.

1  DD-WRT (  0.376 ms  0.556 ms  0.877 ms
 2  * * *
 3  te-0-1-0-3-sur01.**** (**.**.**.***)  51.093 ms  56.714 ms  62.324 ms
 4 (  67.971 ms  74.273 ms  80.052 ms
 5 (  86.266 ms  91.897 ms  97.565 ms
 6 (  103.401 ms  105.129 ms  110.891 ms
 7 (  116.758 ms  75.296 ms  75.295 ms
 8 (  76.225 ms  76.224 ms  75.828 ms
 9 (  75.762 ms  75.197 ms  75.218 ms
10  * * *

Brodi816 04-18-2012 06:27 PM

I'm connected through kvm right now and the only thing I noticed in the syslog file was:
Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2) QMYSQL: Unable to connect

Could this indicate a socket problem for the server? I'm still going through some of the logs though.

iptraf also shows that there is 0 packets going in/out through any of the different types.

I was going through the kern.log and saw that ip_tables took 1586 seconds to initiate. nf_conntrack also took 256 seconds to initiate.

Disabled iptables and the kernal boots in 7 seconds now, but the timeout issue is still occuring

When I shut down the service apache2, I get the error apr_sockaddr_info_get().

jschiwal 04-18-2012 07:01 PM

The MySQL message can be what you get if the MySQL service isn't running. It seems that other services aren't running as well. I'd look closer at the boot log again. Start with the health of the partitions.Are they all mounted? Are any 100% full? A missing /lib/ would cause modules not to load. Then the NICs won't work. Services write to /var/run/ even if they just create a PID file. If there is a problem with the filesystem, it will be mounted ro possible causing services like MySQL and SSH to fail.

Brodi816 04-18-2012 07:10 PM

My server has one hardrive and it's only 5% full. Although it may have been earlier as I forgot to delete a huge amount of backups.
I backed up all my important data and reformatted the server.

All times are GMT -5. The time now is 03:14 PM.