Some bugs on Debian 6 64bit

blustic

blustic

Verified User
Joined
May 26, 2006
Messages
80
Location
France
Hello,

After a fresh install on Debian 6 64bit i have some bugs :

Exim must be reinstalled with source because i have a segmaunt fault and email repeat bug with 5 servers with a new install, and after install with source, all works fine.

named conf file must be modified /etc/bind/named.conf.options :
...
...
auth-nxdomain no; # conform to RFC1035
listen-on-v6 { ::1; };
// listen-on { 127.0.0.1; };
allow-recursion { 127.0.0.1; };
...
...
Click to expand...
by defaut named listen only on 127.0.0.1.. we must comment line before.


Another bug on Admin Level -> Services Monitor when restart named, i have error message : An error has occurred

And when i try to restart it in ssh it works fine :
root@srv5:~# /etc/init.d/named restart
Stopping domain name service...: bind9 waiting for pid 2943 to die.
Starting domain name service...: bind9.
root@srv5:~#
Click to expand...

Sorry for my english, i'm french !

Thank's.
 
Thank you for posting this and the work around, it was driving me nuts.
 
What problem? Someone posted the workaround.

Are you a real person, or a spambot? If you're a real person you should be able to ask a question we can answer. If you're a spambot and eventually figure out a way to spam us, we'll remove all your posts and close your account, since we read every post.

Jeff
 
How to bind by trying to reset the panel have the same message:
An error has occurred
 
The only way you're going to solve the problem is to shell in to your server as root, try restarting BIND from your command line as root, and reading the error message, which may show up either on the command line or in your /var/log/messages file.

How do you restart BIND? That depends on your OS. On both Red Hat Enterprise Linux and CentOS you do it with the following command:
Code: 
# service named restart
(You don't type the #; it's there to tell you that you must run this command as root.)

Jeff
 
Fixed; thanks for bringing it to my attention; the p character is next to the open square bracket key; my finger must have slipped.

Jeff
 
blustic said:
Debian 6 64bit i have bug :
on Admin Level -> Services Monitor when restart named, i have error message : An error has occurred

And when i try to restart it in ssh it works fine :

root@srv5:~# /etc/init.d/named restart
Stopping domain name service...: bind9 waiting for pid 2943 to die.
Starting domain name service...: bind9.
root@srv5:~#

Thank's.
Click to expand...
I have this problem too. New server, new license. Just installed.
Thanx!
 
Possibly you're using a DirectAdmin binary created for some other OS Distribution.

Try tailing the /var/log/messages file while restarting named from DirectAdmin.

Jeff
 
When restarting from DA:
Code: 
17050 execve("/bin/sh", ["sh", "-c", "/etc/init.d/named restart 2>&1"], [/* 17 vars */]) = 0

From bash:
Code: 
execve("/bin/sh", ["sh", "/etc/init.d/named", "restart"], [/* 16 vars */]) = 0

using -c is causing problems with argument passing:
Code: 
root@peter:~# cat ./testargs
#!/bin/bash
echo \$0 is $0
echo \$1 is $1
echo \$2 is $2
# sh ./testargs arg arg arg
$0 is ./testargs
$1 is arg
$2 is arg
# sh -c ./testargs arg arg arg
$0 is ./testargs
$1 is
$2 is

due to:
-c string If the -c option is present, then commands are read from string. If there are arguments after the string, they are assigned to the positional parameters, starting with $0.
Click to expand...

'/etc/init.d/named restart' is being passed as a single argument to sh(1)
 
Not sure why you are talking about bash because everything is executed in sh
 
Hello,

Just an FYI, once recent box, we needed to use this guide:
http://help.directadmin.com/item.php?id=40

because the provided named boot scripts were not working.
We also needed to update the PID file, as it changed for that case.
Code: 
PID_FILE=/var/run/named/named.pid

For the impatient:
Code: 
cd /etc/init.d/
chmod 0 bind bind9
rm -f named
wget -O named http://www.directadmin.com/named.debian
chmod 755 named
perl -pi -e 's#/var/run/bind/run/named.pid#/var/run/named/named.pid#'
mkdir -p /var/run/named
chown named:named /var/run/named
killlall -9 named

Check to see where the named.pid file shows up prior to running the above code, as often, the binary dictates where the PID file ends up, so we need to cater to it.

John
 
Hello !

Thanks for reply. It's work good :

Code: 
cd /etc/init.d/
chmod 0 bind bind9
mv named named.backup
wget -O named http://www.directadmin.com/named.debian
chmod 755 named
perl -pi -e 's#/var/run/bind/run/named.pid#/var/run/named/named.pid#' named
mkdir -p /var/run/named
chown named:named /var/run/named
killlall -9 named
 
Last edited:
DirectAdmin Support said:
Code: 
perl -pi -e 's#/var/run/bind/run/named.pid#/var/run/named/named.pid#'
Click to expand...
Should be:
Code: 
perl -pi -e 's#/var/run/bind/run/named.pid#/var/run/named/named.pid#' named
 
You must log in or register to reply here.
Back
Top