Home / server / Questions / 61337
Accepted
neezer
Asked: 2009-09-02 14:02:19 +0800 CST

Changed hostname on system; Postfix won't start

  • 772

Title says it all. I changed my hostname on my system to a FQDN and now Postfix won't start, saying:

Sep  1 21:57:06 mailer postfix/master[4759]: fatal: bind 127.0.0.1 port 10029: Cannot assign requested address

Interestingly enough, it still has the old name "mailer" in the error log. I've updated my configuration so that it should have the new hostname:

myhostname = mailer.xxx.com
mydomain = mailer.xxx.com

But no joy.

Please help! Let me know if there's any other info you need...

Postconf -n:

alias_database = hash:/etc/aliases
alias_maps = hash:/etc/aliases
append_dot_mydomain = no
biff = no
config_directory = /etc/postfix
inet_interfaces = all
mailbox_size_limit = 0
mydestination = mailer.xxx.com
mydomain = mailer.xxx.com
myhostname = mailer.xxx.com
mynetworks = xxx.xxx.xxx.0/24, 127.0.0.0/8
myorigin = /etc/mailname
notify_classes = bounce, 2bounce, delay, policy protocol, resource, software
readme_directory = no
recipient_delimiter = +
relayhost = 
smtp_tls_session_cache_database = btree:${data_directory}/smtp_scache
smtpd_banner = $myhostname ESMTP $mail_name (Ubuntu)
smtpd_client_restrictions = permit_mynetworks, reject
smtpd_recipient_restrictions = permit_mynetworks, reject_unauth_destination
smtpd_tls_cert_file = /etc/ssl/certs/ssl-cert-snakeoil.pem
smtpd_tls_key_file = /etc/ssl/private/ssl-cert-snakeoil.key
smtpd_tls_session_cache_database = btree:${data_directory}/smtpd_scache
smtpd_use_tls = yes

ip addr:

1: lo: <LOOPBACK> mtu 16436 qdisc noop state DOWN 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UNKNOWN qlen 1000
    link/ether 40:40:4b:1c:8b:d1 brd ff:ff:ff:ff:ff:ff
    inet 173.45.237.59/24 brd 173.45.237.255 scope global eth0
    inet6 fe80::4240:4bff:fe1c:8bd1/64 scope link 
       valid_lft forever preferred_lft forever

/etc/hosts:

127.0.0.1     localhost localhost.localdomain
173.45.237.59     mailer.xxx.com

/etc/network/interfaces

# Used by ifup(8) and ifdown(8). See the interfaces(5) manpage or
# /usr/share/doc/ifupdown/examples for more information.
# The loopback network interface
auto lo
iface lo inet loopback
pre-up iptables-restore < iptables.up.rules

# The primary network interface
# Uncomment this and configure after the system has booted for the first time
auto eth0
iface eth0 inet static
    address 173.45.237.59
    netmask 255.255.255.0
    gateway 173.45.237.1
    dns-nameservers 173.45.224.4 173.45.224.5
networking

5 Answers

  1. Best Answer

    Adding as a new answer, since the topic of the question has changed from Postfix to a general networking problem.

    I suspect it's the pre-up clause on the lo interface in /etc/network/interfaces. If that script (iptables-restore) fails, ifup aborts processing the interface. Try commenting it out and running ifup lo.

    • 3 
  2. 1: lo: <LOOPBACK> mtu 16436 qdisc noop state DOWN 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00

    Why isn't 127.0.0.1 on there? On a server I run (that works), I get:

    1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever

    As a test, try this:

    > nc -l -s 127.0.0.1 -v -p 12345
  listening on [127.0.0.1] 12345 ...

    If you don't get that, it's not postfix's problem :)

    • 2

  3. Please paste the output of postconf -n, ip addr and the content of /etc/hosts.

    Generally the error you're describing indicates that it's trying to bind to a non-local IP address. The output of the above commands should help us figure out why.

    • 0

  4. You have no loopback address. It's got nothing to do with your Postfix config, something (or someone) has monkeyed with your /etc/network/interfaces file. Make sure it contains:

    auto lo
iface lo inet loopback

    At the top, then run ifup lo and all will be well.

    • 0

  5. I just ran into the exact same problem. Postfix and my dkim filter wouldn't start and it turned out to be linked to the loopback interface being down. I tried an ifup lo but got ifup: interface lo already configured. So, I just did an ifdown lo and then an ifup lo. I didn't get an error message so I tried starting dkim/postfix and everything started right up.

    Not sure why the loopback interface didn't start properly after a reboot. Still need to look into that.

    • 0