Koozali.org: home of the SME Server

Obsolete Releases => SME Server 7.x => Topic started by: spook on February 17, 2007, 01:39:20 PM

Title: Httpd doesn't start suddenly? - Fixed!
Post by: spook on February 17, 2007, 01:39:20 PM

Without changing any config files, installing any updates, og changing anything, suddenly Apache won't start up. At boot, the server shows that httpd starts ok, and if I type "httpd" at the console, I get no errors. However, If I do "apachectl -k graceful" to restart it, I get "httpd not running, trying to start". But it doesnt start.

I'm totally lost, and my mother-in-law is busting my chops 'cause she can't log in to my website to see pictures of her grandson.

Oh, can't access server-manager off course, but network shares are up and running. Imap mail access seems down too, tho...

Update:
The tricks from the other posts fixed HTTPD and IMAP(S), but SMTP is still down. Webmail works fine, tho, but users can't send mail from their mail applications (ie. Thunderbird, Outlook etc.). I just don't get it...

Title: Httpd doesn't start suddenly? - Fixed!
Post by: mmccarn on February 17, 2007, 03:07:15 PM

Are there any interesting entries in /var/log/httpd/error_log or /var/log/httpd/admin_error_log?

Title: Httpd doesn't start suddenly? - Fixed!
Post by: pfloor on February 17, 2007, 06:04:02 PM

Quote

If I do "apachectl -k graceful" to restart it

SME doesn't use that command, you are trying to start an additional web server.  There should be 2 instances of apache running.  Try:

/etc/rc7.d/S86httpd-e-smith status
/etc/rc7.d/S86httpd-admin status

Title: Httpd doesn't start suddenly? - Fixed!
Post by: spook on February 17, 2007, 06:50:40 PM

Results:

Code: [Select]

[root@server ~]# /etc/rc7.d/S86httpd-e-smith status
/service/httpd-e-smith: down 1 seconds, want up
and

Code: [Select]


[root@server ~]# /etc/rc7.d/S86httpd-admin status
/service/httpd-admin: run (pid 3501) 19656 seconds, normally down

Starting to feel the panic here... any and all help appreciated!

Title: Httpd doesn't start suddenly? - Fixed!
Post by: spook on February 17, 2007, 09:45:26 PM

Quote from: "mmccarn"

Are there any interesting entries in /var/log/httpd/error_log or /var/log/httpd/admin_error_log?

/var/log/httpd/error_log is filled with this, repeating over and over, once every second:

Code: [Select]

[Sat Feb 17 18:35:59 2007] [crit] (28)No space left on device: mod_rewrite: could not create rewrite_log_lock
Configuration Failed
and /var/log/httpd/admin_error_log is only these 4 lines:

Code: [Select]

[Sat Feb 17 13:19:53 2007] [warn] Init: Session Cache is not configured [hint: SSLSessionCache]
[Sat Feb 17 13:19:53 2007] [notice] Digest: generating secret for digest authentication ...
[Sat Feb 17 13:19:53 2007] [notice] Digest: done
[Sat Feb 17 13:19:54 2007] [notice] Apache configured -- resuming normal operations

Title: Httpd doesn't start suddenly? - Fixed!
Post by: pfloor on February 17, 2007, 09:56:05 PM

This is a known issue, search is your friend.

http://forums.contribs.org/index.php?topic=35534.0

and

http://forums.contribs.org/index.php?topic=35298.0

And many more.

Also this bug will give you some background:

http://bugs.contribs.org/show_bug.cgi?id=2035

Title: Httpd doesn't start suddenly? - Fixed!
Post by: spook on February 18, 2007, 10:51:31 AM

Thanks a ton! HTTP services are up & running again (www.gutaarn.dk back in business :)) but SMTP is dead :(

Whenever I try to send mail from Thunderbird I get error 12226 - Connection refused from SMTP.

I tried the imap trick (which worked for imap) of killing the SMTP services, and then deleting the certificates and running the script. But still no go on the SMTP.

Title: Httpd doesn't start suddenly? - Fixed!
Post by: mmccarn on February 18, 2007, 03:59:09 PM

Are your Thunderbird clients configured to use the SME as their SMTP server, or are they configured to use another off-site server for outgoing email?

If they're configured to use the SME, do you see anything interesting in /var/log/qpsmtpd/current when your clients try to send mail?

If they're configured to use an off-site SMTP server, do you have SMTP proxying turned on or off?  (server-manager/ Security->Proxy settings)?  (I've never seen SMTP proxying work, but I haven't tried very hard, either...)

Title: Httpd doesn't start suddenly? - Fixed!
Post by: spook on February 18, 2007, 07:57:23 PM

They're using the SME for SMTP, but its refusing connections - just like IMAP and HTTPD did...

Not much seems to be happening in /var/log/qpsmtpd/current, it doesn't change one bit (literally) when I try to log in to SMTP.

SMTP is running tho - if smtp-auth-proxy IS the right service?
[root@server ~]# /etc/init.d/smtp-auth-proxy status
/service/smtp-auth-proxy: run (pid 2219) 33459 seconds

Title: This worked for me
Post by: bugme on February 19, 2007, 02:08:05 AM

Try running this:

 /sbin/e-smith/expand-template /var/service/qpsmtpd/ssl/cert.pem

Title: Re: Httpd doesn't start suddenly? - Fixed, but SMTP still do
Post by: CharlieBrady on February 19, 2007, 03:05:27 AM

Quote from: "spook"

Without changing any config files, installing any updates, og changing anything, suddenly Apache won't start up.

Please report all malfunctions via (only) the Bug Tracker. Thanks.

Title: Re: This worked for me
Post by: spook on February 19, 2007, 09:45:27 PM

Quote from: "bugme"

Try running this:

 /sbin/e-smith/expand-template /var/service/qpsmtpd/ssl/cert.pem

Awesome, worked like a charm :)