| View previous topic :: View next topic |
| Author |
Message |
Joseph_sys
Advocate
Joined: 08 Jun 2004
Posts: 2716
Location: Edmonton, AB
|
 Posted: Sat Sep 23, 2006 5:45 am Post subject: [SOLVED] could not bind to address [::]:443 |
 |
|
After today's upgrade I'm getting an error with apache:
| Code: | Starting apache2 ...
(98)Address already in use: make_sock: could not bind to address [::]:443
no listening sockets available, shutting down |
| Code: | # netstat -natp | grep LISTEN
tcp 0 0 127.0.0.1:3306 0.0.0.0:* LISTEN 6483/mysqld
tcp 0 0 0.0.0.0:139 0.0.0.0:* LISTEN 10200/smbd
tcp 0 0 127.0.0.1:631 0.0.0.0:* LISTEN 6844/cupsd
tcp 0 0 127.0.0.1:5432 0.0.0.0:* LISTEN 6565/postmaster
tcp 0 0 0.0.0.0:25 0.0.0.0:* LISTEN 6970/master
tcp 0 0 0.0.0.0:445 0.0.0.0:* LISTEN 10200/smbd
tcp 0 0 :::80 :::* LISTEN 9003/apache2
tcp 0 0 :::443 :::* LISTEN 9003/apache2 |
In addition when I try to start sshd I get:
/etc/init.d/sshd start
/etc/init.d/sshd: line 17: 7470 Segmentation fault ${SSHD_BINARY} -t ${myopts}
This only happened on my AMD64 box, both x86 boxes upgraded just fine.
Last edited by Joseph_sys on Sun Sep 24, 2006 5:38 am; edited 1 time in total |
|
| Back to top |
|
 |
phorn
Tux's lil' helper
Joined: 01 Jul 2006
Posts: 109
|
| Posted: Sat Sep 23, 2006 6:04 am Post subject: |
|
|
But in your listing of "netstat" it shows apache2 as running. That would explain why it can't bind port 443.
Try killing apache2 (the PID 9003 one) first...
Otherwise, it seems like apache is already running. Can you connect to the local server on HTTP or HTTPS?
The sshd error is in fact strange. You may want to run it in debug mode (figure out what myopts and SSHD_BINARY are, and run it manually on the command line with a debug option).
It may be that the OpenSSL library is bad, or that you should re-emerge openssh...
You might want to try a revdep-rebuild on ssh, to make sure it is using the correct libraries. |
|
| Back to top |
|
|
Joseph_sys
Advocate
Joined: 08 Jun 2004
Posts: 2716
Location: Edmonton, AB
|
| Posted: Sat Sep 23, 2006 6:12 am Post subject: |
|
|
| phorn wrote: | But in your listing of "netstat" it shows apache2 as running. That would explain why it can't bind port 443.
Try killing apache2 (the PID 9003 one) first...
Otherwise, it seems like apache is already running. Can you connect to the local server on HTTP or HTTPS?
The sshd error is in fact strange. You may want to run it in debug mode (figure out what myopts and SSHD_BINARY are, and run it manually on the command line with a debug option).
It may be that the OpenSSL library is bad, or that you should re-emerge openssh...
You might want to try a revdep-rebuild on ssh, to make sure it is using the correct libraries. |
I tried killing apache PID, did not help, when I start it, it listen to: :::80 and :::443
nor can I connect to localhost.
I've re-emerge openssh already and did revdep-rebuild
Noting helps :-/ |
|
| Back to top |
|
|
Joseph_sys
Advocate
Joined: 08 Jun 2004
Posts: 2716
Location: Edmonton, AB
|
| Posted: Sat Sep 23, 2006 6:22 am Post subject: |
|
|
Though doing: netstat -natp I get:
| Code: | Active Internet connections (servers and established)
Proto Recv-Q Send-Q Local Address Foreign Address State PID/Program name
tcp 0 0 127.0.0.1:3306 0.0.0.0:* LISTEN 6483/mysqld
tcp 0 0 0.0.0.0:139 0.0.0.0:* LISTEN 10200/smbd
tcp 0 0 127.0.0.1:631 0.0.0.0:* LISTEN 6844/cupsd
tcp 0 0 127.0.0.1:5432 0.0.0.0:* LISTEN 6565/postmaster
tcp 0 0 0.0.0.0:25 0.0.0.0:* LISTEN 6970/master
tcp 0 0 0.0.0.0:445 0.0.0.0:* LISTEN 10200/smbd
tcp 0 0 10.0.0.104:48268 64.233.167.99:80 ESTABLISHED 7348/firefox-bin
tcp 0 0 :::80 :::* LISTEN 10470/apache2
tcp 0 0 :::443 :::* LISTEN 10470/apache2
tcp 396 0 ::ffff:127.0.0.1:80 ::ffff:127.0.0.1:58804 CLOSE_WAIT -
tcp 442 0 ::ffff:10.0.0.104:80 ::ffff:10.0.0.104:38080 CLOSE_WAIT - |
The last two lines are interesting, and it could be causing this problem.
How to shut it down, there is no PID associated with it?
update:
After killing process ID 10470 the last two lines dissapeard but after restarting apache still get :::80 :::443
What else should I try? |
|
| Back to top |
|
|
phorn
Tux's lil' helper
Joined: 01 Jul 2006
Posts: 109
|
| Posted: Sat Sep 23, 2006 11:51 pm Post subject: |
|
|
So what's the actual error?
Does it hang: you connect to it on a web browser, and it then it sits there saying connecting... or waiting...?
Or does it give a 500 error when loading?
Or does it spit out an error message when you start it?
What happens if you run that netstat -natp when apache is not running?
It may be a module causing it. I have had a problem with mod_rewrite or mod_alias causing a hang on start, and commenting out that LoadModule line in httpd.conf made it start normally.
If it's just the "address in use" error, what happens if you reboot the machine? |
|
| Back to top |
|
|
Joseph_sys
Advocate
Joined: 08 Jun 2004
Posts: 2716
Location: Edmonton, AB
|
| Posted: Sun Sep 24, 2006 12:16 am Post subject: |
|
|
| phorn wrote: | So what's the actual error?
Does it hang: you connect to it on a web browser, and it then it sits there saying connecting... or waiting...?
Or does it give a 500 error when loading?
Or does it spit out an error message when you start it?
What happens if you run that netstat -natp when apache is not running?
It may be a module causing it. I have had a problem with mod_rewrite or mod_alias causing a hang on start, and commenting out that LoadModule line in httpd.conf made it start normally.
If it's just the "address in use" error, what happens if you reboot the machine? |
When I try to start it: /etc/init.d/apache2 start
* Starting apache2 ...
(98)Address already in use: make_sock: could not bind to address [::]:443
no listening sockets available, shutting down
Unable to open logs
When I kill that starnge process ID :::80 and run netstat
| Code: | netstat -natp
Active Internet connections (servers and established)
Proto Recv-Q Send-Q Local Address Foreign Address State PID/Program name
tcp 0 0 127.0.0.1:3306 0.0.0.0:* LISTEN 6483/mysqld
tcp 0 0 0.0.0.0:139 0.0.0.0:* LISTEN 7037/smbd
tcp 0 0 127.0.0.1:631 0.0.0.0:* LISTEN 6845/cupsd
tcp 0 0 127.0.0.1:5432 0.0.0.0:* LISTEN 6566/postmaster
tcp 0 0 0.0.0.0:25 0.0.0.0:* LISTEN 6971/master
tcp 0 0 0.0.0.0:445 0.0.0.0:* LISTEN 7037/smbd |
Base on the address it looks to me as if apache was trying to bind to IPv6 instead of IPv4 but I don't know why?
tcp 0 0 :::80 :::* LISTEN 10470/apache2
tcp 0 0 :::443 :::* LISTEN 10470/apache2
It should be:
tcp 0 0 0.0.0:80 :::* LISTEN 10470/apache2
tcp 0 0 0.0.0:443 :::* LISTEN 10470/apache
Restarting the machine doesn't help.
mod_rewrite or mod_alias don't cause any problem commenting them out doesn't help it still try to bind to IPv6
#Joseph |
|
| Back to top |
|
|
Joseph_sys
Advocate
Joined: 08 Jun 2004
Posts: 2716
Location: Edmonton, AB
|
| Posted: Sun Sep 24, 2006 1:32 am Post subject: |
|
|
If I change in http.conf
Listen 80
to
Listen 0.0.0.0:80
It binds to IPv4 and apache is working OK but WHY???
with Listen 80 it is tryig to bind to IPv6 |
|
| Back to top |
|
|
Joseph_sys
Advocate
Joined: 08 Jun 2004
Posts: 2716
Location: Edmonton, AB
|
| Posted: Sun Sep 24, 2006 5:37 am Post subject: |
|
|
SOLVED, it took me all day. So if somebody has the same problem here is are the solutions (options).
The problem was caused by the fact that I had a IPv6 protocol enabled in the Kernel so after today's upgrade to apache-2.0.58-r2. It decided that IPv6 should take precedence over IPv4 and apache binded to :::80 instead of 0.0.0.0:80
The only question I have why all of a sudden and which package cause it?
a.)
Quick way around it is to modify correctly you /etc/hosts file:
...
# IPV6 versions of localhost and co
::1 ip6-localhost ip6-loopback syscon3.your_domain_name syscon3
...
restart apache.
Though, make sure your connection gets through your firewall to IPv6
b.)
Another solution is to tell your apache to listen to IPv4 that is in your httpd.conf
change from:
Listen 80
to
Listen 0.0.0.0:80
c.)
Disable IPv6 in your kernel
If I missed anything please add. |
|
| Back to top |
|
|
|
Networking & Security
