Sites stopped working

This morning all my test sites have stopped working. They have been working well since Monday.

The dns records seem to have resolved ok with wooodsidekennels.com using my ns1.badbison.co.uk

my own dns records haven't changed and point to 93.97.188.1.

If I type the ip address directly into a browser it displays the default apache page.

The only change I've noticed this morning is the mail server has now resolved and delivering all mail. Also there was a warning message on ispconfig advising of an update required. To deal with this from the command line I typed:
apt-get update
apt-get upgrade
and it performed the commands.
I then rebooted the server from command line.

The warning message is now gone from ispconfig but it's stil not serving the sites up.

Can anyone suggest what might have happened? Do I need to restart any services or something?

The server was setup using the karmic perfect guide.

 

Are there any errors in Apache's error log?

 

[Wed Dec 16 20:41:11 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Wed Dec 16 20:42:26 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Wed Dec 16 20:45:32 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Wed Dec 16 20:50:50 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Thu Dec 17 09:32:01 2009] [notice] caught SIGTERM, shutting down
[Thu Dec 17 09:33:02 2009] [notice] suEXEC mechanism enabled (wrapper: /usr/lib/apache2/suexec)
[Thu Dec 17 09:33:04 2009] [notice] Apache/2.2.12 (Ubuntu) PHP/5.2.10-2ubuntu6.3 with Suhosin-Patch mod_ssl/2.2.12 OpenSSL/0.9.8g configured -- resuming normal operations
[Thu Dec 17 09:46:37 2009] [notice] caught SIGTERM, shutting down
[Thu Dec 17 09:47:36 2009] [notice] suEXEC mechanism enabled (wrapper: /usr/lib/apache2/suexec)
[Thu Dec 17 09:47:38 2009] [notice] Apache/2.2.12 (Ubuntu) PHP/5.2.10-2ubuntu6.3 with Suhosin-Patch mod_ssl/2.2.12 OpenSSL/0.9.8g configured -- resuming normal operations
[Thu Dec 17 10:44:34 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Thu Dec 17 10:58:50 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Thu Dec 17 12:51:58 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Thu Dec 17 13:53:41 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Thu Dec 17 14:50:54 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Thu Dec 17 15:27:25 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Thu Dec 17 15:55:12 2009] [notice] caught SIGTERM, shutting down
[Thu Dec 17 15:56:12 2009] [notice] suEXEC mechanism enabled (wrapper: /usr/lib/apache2/suexec)
Use of uninitialized value $DBI_DUMP in alarm at /usr/local/ispconfig/server/scripts/vlogger line 531, <CONF> line 1.
202,1
[Thu Dec 17 15:56:12 2009] [notice] suEXEC mechanism enabled (wrapper: /usr/lib/apache2/suexec)
Use of uninitialized value $DBI_DUMP in alarm at /usr/local/ispconfig/server/scripts/vlogger line 531, <CONF> line 1.
[Thu Dec 17 15:56:13 2009] [notice] Apache/2.2.12 (Ubuntu) PHP/5.2.10-2ubuntu6.3 with Suhosin-Patch mod_ssl/2.2.12 OpenSSL/0.9.8g configured -- resuming normal operations
[Thu Dec 17 16:05:32 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Thu Dec 17 16:36:56 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Thu Dec 17 18:26:08 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Thu Dec 17 19:56:01 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Thu Dec 17 20:27:16 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Thu Dec 17 22:52:36 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Fri Dec 18 01:46:48 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Fri Dec 18 02:37:41 2009] [error] [client 221.195.73.68] script '/usr/local/ispconfig/interface/web/prx2.php' not found or unable to stat
[Fri Dec 18 08:25:00 2009] [notice] caught SIGTERM, shutting down
[Fri Dec 18 08:25:59 2009] [notice] suEXEC mechanism enabled (wrapper: /usr/lib/apache2/suexec)
[Fri Dec 18 08:26:01 2009] [notice] Apache/2.2.12 (Ubuntu) PHP/5.2.10-2ubuntu6.3 with Suhosin-Patch mod_ssl/2.2.12 OpenSSL/0.9.8g configured -- resuming normal operations
[Fri Dec 18 15:44:38 2009] [notice] caught SIGTERM, shutting down
[Fri Dec 18 15:45:38 2009] [notice] suEXEC mechanism enabled (wrapper: /usr/lib/apache2/suexec)
Use of uninitialized value $DBI_DUMP in alarm at /usr/local/ispconfig/server/scripts/vlogger line 531, <CONF> line 1.
[Fri Dec 18 15:45:41 2009] [notice] Apache/2.2.12 (Ubuntu) PHP/5.2.10-2ubuntu6.3 with Suhosin-Patch mod_ssl/2.2.12 OpenSSL/0.9.8g configured -- resuming normal operations
[Fri Dec 18 20:38:16 2009] [notice] caught SIGTERM, shutting down
[Fri Dec 18 20:39:16 2009] [notice] suEXEC mechanism enabled (wrapper: /usr/lib/apache2/suexec)

I got a bad feeling about all those logs!
Help

 

Can you post one of the vhost configurations that isn't working?
What's the output of

Code:

ifconfig

?

 

Thanks for the help. I took a different approach Friday - started again on debian lenny, transferred a domain to schlundtech to use as a glue record and no2 ns server left it over the weekend to propogate and this morning it's all started working fine!

All test websites can be accessed from different Internet areas. Hooray.

I've not tried it yet but i'm confident the mail services work too. What I would like to do is set up pop or imap mail services so any poiinters there would be good. I'm guessing it'll be just a case of setting up dns records?

If anyone knows of any howtos to cover that it would be appreciated.

Thanks

 

Did you follow this tutorial? http://www.howtoforge.com/perfect-server-debian-lenny-ispconfig3
Then POP3 and IMAP are already running. All you have to do is use a hostname/domain or IP address in your email client that points to the server.

 

yes I did and oh wow! It does work! I just used the IP address.

What should I use as the outgoing server details though? The IP address doesn't seem to work and neither does mail.mydomain.com

Also I'm having a little trouble with incoming mails. Some get through fine and others just never seem to appear.

 

The same address that you use for POP3/IMAP. Make sure that port 25 is accessible from the outside.

Any errors in your mail log?

 

Thanks. The ip address as the outgoing host worked perfect.

I'll put this all into a beginners tutorial for others so you don't keep getting the same questions rephrased by people.

There were no errors in the mail error log but there were warnings in that log that I'll paste at the end.

There is something I haven't managed to get my head around regarding secondary name servers. I appreciate that I've told the schundtech registry to use my ip glued to a domain as the primary server. And it also knows the ns10.schlundtech.de is the secondary. And I understand that in my dns records I've added the ip of ns10 to allow export of the records.

Does this mean that my ns will contact the schlundtech one and give it the details, or does ns10 somehow know to contact my ns to get the records?

I'm nearly done With beginners questions (i think)

 

The secondary DNS will contact the primary one and get the records.

It seems as if there's something wrong with your name resolution. What's in /etc/resolv.conf? Are these valid nameservers?

 

I thought the name servers are valid. My dns shows a record for ns1.badbison.co.uk pointing to my public ip address which the router points all ports to the same ports on the lan ip of the server.

As the second name server I've used ns10.schlundtech.de as that is used in the howto. To get the ip address of this server I used the dig command which returned in the answer section 62.116.163.100. I have put this address in the ip address to export zone records section of my dns records.

That sort of refers back to my earlier question of how does ns10.schlundtech.de know it has to act as my secondary name server and go get the records.

I can't display the contents of resolv.conf as I'm on a mobile device right now and the result get lost out of the screen, although you're welcome to ssh it yourself, though I appreciate you can't help that personally to everyone.

The fault does seem intermittent. On my local network I can view web pages held fine. It also works from my phone fine using the telephone connection. It worked fine from work yesterday, but today it does not. The template sites I've put up are badbison.com and prizepalace.co.uk if you'd like to view.

There is something playing on my mind though, I have a faint memory of Reading somewhere not to use the same domain name as your LAN is. My router does have the same LAN domain as my name servers domain - badbison.co.uk. Have I imagined this or should I change the routers domain setting?

 

Maybe the nameservers in your resolv.conf are not allowed to do recursive lookups.
Can you try

Code:

nameserver 145.253.2.75

as the first line in /etc/resolv.conf?

 

ok, back on a terminal today... the resolve.conf file looks like this:

Code:

domain badbison.co.uk
search badbison.co.uk
nameserver 87.194.255.154
nameserver 87.194.255.155
nameserver 192.168.0.1

which I've now adjusted to look like this:

Code:

nameserver 145.253.2.75
domain badbison.co.uk
search badbison.co.uk
nameserver 87.194.255.154
nameserver 87.194.255.155
nameserver 192.168.0.1

I'm not sure where it's got those names servers from though, the router is the 0.1 and I have a feeling the .154 and .155 are the names servers that the router dishes out when leasing dhcp. However, the server that this is running on is on a static IP if 0.100 as per the lenny perfect tutorial.

I've just rebooted it. Should I remove the other lines or anything?

 

Looks good. Do you still get so many Name or service not known errors in your mail log?

 

There's no mail error logs but there are mail warnings as below

Code:

Dec 25 22:53:26 badbuntu postfix/smtpd[7303]: warning: 190.245.168.244: hostname 244-168-245-190.fibertel.com.ar verification failed: Name or service not known
Dec 25 23:59:57 badbuntu postfix/smtpd[25627]: warning: 89.42.187.16: address not listed for hostname 16-187-42-89-cable.canals.ro
Dec 26 00:54:41 badbuntu postfix/smtpd[26425]: warning: 123.19.24.244: address not listed for hostname localhost
Dec 26 03:54:12 badbuntu postfix/smtpd[29026]: warning: 98.145.80.232: hostname cpe-98-145-80-232.natnow.res.rr.com verification failed: Name or service not known
Dec 26 08:18:54 badbuntu postfix/smtpd[14102]: warning: 113.170.214.251: address not listed for hostname localhost. 

it's odd, when I send an email from another address i get a rejection reply straight away, although often the email does come through but not for about an hour later. A sample mail box is my [email protected] my name is David!

Should I delete those other name servers from resolv.conf. ? Also my file doesn't end with the question mark like the one you mentioned. I presume this doesn't matter?

Thanks for all your help With this by the way.

 

I'd leave them where they are.

What question mark?

I guess that you still get these Name or service not known warnings simply means that the hostnames really do not exist.

 

Leave even the one that relates to the router ip? I'll trust you on that.

What does the resolv.conf file actually do?
Ref the ? , a question mark was included a couple of times at the end of resolv.conf. Just wanted to make sure I was editting the right file.

Still not having much luck with those emails. If I send on from an mailbox on the same server it gets through fine, if it comes from other servers outside the domain then we run into problems. Please see the return message below. I'd love to get this working properly.

Code:

.This message was created automatically by mail delivery software.

A message that you sent could not be delivered to one or more of its
recipients. This is a permanent error. The following address(es) failed:

*[email protected]
***retry timeout exceeded

------ This is a copy of the message, including all the headers. ------

Return-path: <[email protected]>
Received: from [82.132.139.201] (helo=[10.28.192.186])
* *by bluestar5.ukhost4u.com with esmtpa (Exim 4.69)
* *(envelope-from <[email protected]>)
* *id 1NP0se-0003z8-Hj
* *for [email protected]; Sun, 27 Dec 2009 21:42:29 +0000
Message-Id: <[email protected]>
From: "[email protected]" <[email protected]>
To: [email protected]
Content-Type: text/plain;
* *charset=us-ascii;
* *format=flowed
Content-Transfer-Encoding: 7bit
X-Mailer: iPhone Mail (7D11)
Mime-Version: 1.0 (iPhone Mail 7D11)
Date: Sun, 27 Dec 2009 21:42:33 +0000

Hhhh
 

 

I tried from a different computer on a different network today to access my two test sites. One, prizepalace.co.uk came up first time. The other, badbison.com failed the first two times quoting dns error, unable to find host.. clicked retry a third time and it came up no problems and has done ever since.

Email is still behaving very odd.. someone sent me an email from a yahoo email account last night. It never arrived and returned the following message to the sender:

and some more...

However, this morning someone sent me an email from a yahoo account and it came through fine with no reported errors.

I get the feeling one of the name servers are not working properly and it's purely chance which name server is used for the look up. Is there anyway to tell which is working properly? According to ISP config3 I have MX and A records setup for the account.

It looks like this:

Is there a way I could test to see if my name server gives out the correct details and then a way to test to see if NS10.schlundtech.de holds the correct details?

 

You can test your DNS records for badbison.com on www.intodns.com

 

That site did prove useful.

I wasn't sure if my previous slave name server was doing it's job so I've used one supplied by zoneedit which does seem to be picking up the records from the primary. The serials are currently out of sync but I think that'll cure in time.

I seem to have a problem with glue records. The results are as below

web pages are still appearing intermittently and emails take six hours to get through from some domains.
I'd really appreciate some guidance as what to do next please.