Virtual Hosting on ecsesql box
NETWORK PROBLEMS on ecsesql box
- All networks stop talking - Solution is REBOOT
- Disk issues, segmentation fault, fsck and kernel panic

Virtual Hosting on ecsesql box

DNS entries

ECSESQL2a = CTIEWARE = www.ds == entires in the 'special DNS entry' list that is maintained by ITS.

To add new virtual hostnames you have to submit an ITS helpdesk job to have them update it.

You can't just add to addhost.

Create user account and add to correct group

useradd -c "pimce geoff taylor" -m pimce

Give it a password

 passwd pimce

edit /etc/group and add pimce to the group www - may not need this!!

Configure apache

The apache config file is in /usr/local/apache

edit the httpd.conf file in conf and add

<VirtualHost *>
    ServerAdmin root@ecsesql2a.eng.monash.edu.au
    DocumentRoot /home/www/pimce/
    ServerName www.pimce.edu.au
    ErrorLog logs/pimce/error_log
    CustomLog logs/pimce/access_log common
</VirtualHost>

create the virtual host log directory

create the pimce directory in the /usr/local/apache/logs directory.

create the web tree home directory with correct permissions

cd /home/www
mkdir pimce
chown pimce:www pimce
chmod g+s pimce

Next restart apache


/usr/local/apache/bin/apachectl configtest             Verify it is ok first

/usr/local/apache/bin/apachectl graceful

NETWORK PROBLEMS on ecsesql box

All networks stop talking - Solution is REBOOT

Pretty simple really - do not attempt service network restart as it doesn't work!

Instead REBOOT

For some reason ecsesql2a=ctieware=www.ds loses the ability to talk thru it's 2 network cards - both internal and monash networks. ifconfig still shows them as being active and working, but there is no traffic in or out. pulling out leads pops up the ethernet cable disconnected and reconnected messages on the console - but nothing will cause the tcp/ip stack to send packets in or out of the interfaces. All you can do is reboot.

Just in case it related to something I spotted with Ahmet logging in from an IPv6 enabled box, I enabled IPv6, but haven't put in any DNS entries for it.

-- DanGrimm - 29 Nov 2005

Disk issues, segmentation fault, fsck and kernel panic

It died on me!

Not happy!

Had to power down to reboot, give root password and run fsck /home manually to fix it (basically say yes to all proposed fixes, including deleting dud file that contained deleted bad inode).

1.4Gb file transfer via sftp and/or scp killed it [transfer from mammoth via monash and internal ip numbers]

FSCK takes about 5 minutes with no reporting of faults - so you just have to be patient or use -v for verbose.

New box soon!!!!!!! -- DanGrimm - 21 Feb 2008

Topic VirtualHostSql . { Edit | Attach | Ref-By | Printable | Diffs | r1.3 | > | r1.2 | > | r1.1 | More }

Revision r1.3 - 21 Feb 2008 - 03:13 GMT - DanGrimm Parents: WebHome	Copyright © 1999-2003 by the contributing authors. All material on this collaboration platform is the property of the contributing authors. Ideas, requests, problems regarding TWiki? Send feedback.

	TWiki > ECSEIT > VirtualHostSql	TWiki webs: Main \| TWiki \| Know \| Sandbox
	ECSEIT . { Changes \| Index \| Search \| Go }