Virtualmin admin panel timeout after installation

k4emic · July 1, 2009, 1:59pm

I’ve seemingly installed virtualmin on my server but I can’t log in to the post-installation panel through https://some.ip.he.re:10000 I simply get a timeout error for some reason or another. (I can access the standard http “welcoming” just fine though)

The installation didn’t prompt me for any information and didn’t really tell me whether the installation was finished - The only hint I got was that the terminal was at my disposal again. At first I though it might be a corrupted sh file or just a coincidence of some sort. But after four installs on four neaty, newly installed ubuntu boxes, I’m ready to give up!

Any pointers will be greatly appreciated.

– Mads

Eric · July 1, 2009, 2:57pm

Howdy,

Are you able to SSH into your server?

If so, do you see anything when you run this command:

ps auw | grep miniserv

Also, are you running your server behind a NAT router? Is it possible that either your ISP is blocking port 10000, or that you need to setup port forwarding?

-Eric

k4emic · July 1, 2009, 4:28pm

SSH works. No problems there.
the ps auw | grep miniserv command outputs the following:
1000 9919 0.0 0.0 3004 756 pts/0 R+ 18:21 0:00 grep miniserv

The server shouldn’t have any firewall issues since its a dedicated server with a static WAN IP. I’ll ask the provider anyway, just in case.

– Mads

Eric · July 1, 2009, 4:34pm

Okay, it looks like Webmin isn’t running.

Try this command:

/etc/init.d/webmin start

k4emic · July 1, 2009, 4:57pm

Still getting a timeout.
root 11136 0.0 0.0 3004 764 pts/1 S+ 18:56 0:00 grep miniserv

ronald · July 1, 2009, 5:47pm

are there any leads in the virtualmin-install.log ? its in the /dir from where you installed virtualmin, per haps /root

Eric · July 1, 2009, 6:22pm

Okay, looks like Webmin isn’t starting.

So Ronald’s suggestion of looking in the install log is a great idea.

In addition, I might also look in the webmin log, over in /var/webmin/miniserv.error, to see if that’s showing any errors.

-Eric

Joe · July 1, 2009, 7:42pm

Was there any error when you tried to start the webmin service? Also look in /var/webmin/miniserv.error

k4emic · July 2, 2009, 9:15am

miniserv.error tells me that port 10000 is already in use even though I haven’t installed anything besides openSSH.


[02/Jul/2009:10:21:19 +0200] miniserv.pl started
[02/Jul/2009:10:21:19 +0200] PAM authentication enabled
Failed to bind to port 10000 : Address already in use
Could not listen on any ports at /usr/share/webmin/miniserv.pl line 453.`

I ran cat /etc/services, looked for port 10000 and found this.
webmin 10000/tcp

I found the following line which indicates that I might have done something terribly wrong.
Webmin install complete. You can now login to https://pdhttp1:10000/

Install log attached

ronald · July 2, 2009, 3:26pm

okay you did something terribly wrong

fully qualified as pdhttp1.example.com

but there is also good news. Likely this is easily solved. You need a domainname you control, i.e registered in your name. Then you want to use that and point it to your server from the registrars control panel.

then log into your server over ssh and type: hostname ns1.yourdomainname.com

where yourdomainname.com is the domain name you have pointed to your server. So that’s a good start I think to get this box working

ronald · July 2, 2009, 3:35pm

also localhost and postfix need to be fixed, i.e comment out myorigin and set localhost.localdomain

setting myorigin
setting destinations: pdhttp1.example.com, localhost.example.com, , localhost

k4emic · July 2, 2009, 4:21pm

type: hostname ns1.yourdomainname.com Shouldn't I change the hostname through the /etc/hostname file instead?

Before doing this, I get pdhttp1.example.com when typing hostname --fqdn. This domain is set up on a DNS elsewhere (and works on the regular HTTP protocol).

cat /etc/hosts
127.0.0.1 localhost.localdomain 89.xx.xx.68 pdhttp1.example.com pdhttp1

The following lines are desirable for IPv6 capable hosts

::1 ip6-localhost ip6-loopback
fe00::0 ip6-localnet
ff00::0 ip6-mcastprefix
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
ff02::3 ip6-allhosts

also localhost and postfix need to be fixed, i.e comment out myorigin and set localhost.localdomain If you're referring to the main.cf file in /etc/postfix, myorigin was already commented out. I set it to localhost.localdomain.

I’m terribly sorry if I’m making some of you shake your head, but I’ve never dealt with anything but regular LAMP servers maintained through SSH. Postfix and DNS management is new to me.

– Mads

ronald · July 2, 2009, 4:22pm

did you edit the virtualmin-install.log ?

As I believed you have set the hostname using example.com! Using pdhttp1 is fine of course. The question lies in the domainname. This needs to be a valid one. example.com wouldn’t be valid for your server hehe

127.0.0.1 localhost.localdomain

should read

127.0.0.1 localhost.localdomain localhost

k4emic · July 2, 2009, 5:50pm

did you edit the virtualmin-install.log ?
I replaced the real domain name with example.com out of security concerns.

k4emic · July 4, 2009, 3:34pm

I’ve performed the suggested changes and checked up on the firewall issue. No change on the issue.

ronald · July 4, 2009, 4:20pm

Edit my post.

can you reach the site at https ? just port 10000 isn’t working? Can you reach usermin on port 20000 ? http and/or https?
Do you mind publishing the domain your server can be reached on to check per haps via dns tools?

k4emic · July 6, 2009, 11:04am

Just tried https without specifying the port.

Firefox gave me this.
SSL received a record that exceeded the maximum permissible length.

(Error code: ssl_error_rx_record_too_long)

You can run some diagnostics on http1.pagedesign.dk and see if that turns up anything.

– Mads