Is there a way to take down roundcube and notify the state to end users? I'd like to have a 'service down' page during backups for the early birds when they come in.
Chris skrev den 2015-08-25 17:55:
Is there a way to take down roundcube and notify the state to end users? I'd like to have a 'service down' page during backups for the early birds when they come in.
sure in apache:
ifdefined RCU_UP endif
ifdefined RCU_DOWN endif
put there virtualhost roundcube in RCU_UP
and the maintaince in RCU_DOWN
now set apache conf to -D RCU_DOWN on maintaince, and back to -D RCU_UP when done
offcourse not both at the same time :=)
On 08/25/2015 11:55 AM, Chris wrote:
Is there a way to take down roundcube and notify the state to end users? I'd like to have a 'service down' page during backups for the early birds when they come in.
Replace index.php with something of higher-priority, like index.html, saying the service is down. The priority depends on your web server configuration -- for example "DirectoryIndex" in Apache.
When Roundcube comes back up, delete index.html.
So nothing in Roundcube then. :>
I ask the silly question this way because I'm hosted on OS X Server which has a parent tool to manage httpd.conf - which does not include virtual host flexibility. I'll configure it and see if the GUI admin breaks my VH's.
Thanks!
On Tue, Aug 25, 2015 at 12:17 PM, Benny Pedersen me@junc.eu wrote:
Chris skrev den 2015-08-25 17:55:
Is there a way to take down roundcube and notify the state to end users? I'd like to have a 'service down' page during backups for the early birds when they come in.
sure in apache:
ifdefined RCU_UP endif
ifdefined RCU_DOWN endif
put there virtualhost roundcube in RCU_UP
and the maintaince in RCU_DOWN
now set apache conf to -D RCU_DOWN on maintaince, and back to -D RCU_UP when done
offcourse not both at the same time :=) _______________________________________________ Roundcube Users mailing list users@lists.roundcube.net http://lists.roundcube.net/mailman/listinfo/users
That works.
On Tue, Aug 25, 2015 at 4:46 PM, Michael Orlitzky michael@orlitzky.com wrote:
On 08/25/2015 11:55 AM, Chris wrote:
Is there a way to take down roundcube and notify the state to end users? I'd like to have a 'service down' page during backups for the early birds when they come in.
Replace index.php with something of higher-priority, like index.html, saying the service is down. The priority depends on your web server configuration -- for example "DirectoryIndex" in Apache.
When Roundcube comes back up, delete index.html.
Roundcube Users mailing list users@lists.roundcube.net http://lists.roundcube.net/mailman/listinfo/users
It's elegant, too. However, renaming or moving the "out of service" page seems wise. You'll need it in the future.
Michael Orlitzky skrev den 2015-08-25 22:46:
When Roundcube comes back up, delete index.html.
thats more simple yes, but http://www.example.net/index.php will still work
apache need another documentroot for being down
Chris skrev den 2015-08-25 22:47:
So nothing in Roundcube then. :>
correct
I ask the silly question this way because I'm hosted on OS X Server which has a parent tool to manage httpd.conf - which does not include virtual host flexibility. I'll configure it and see if the GUI admin breaks my VH's.
there is no silly questions, only silly answers sometimes
if it breaks its sure not gentoo :=)
Chris ecwh73@gmail.com wrote:
Is there a way to take down roundcube and notify the state to end users? I'd like to have a 'service down' page during backups for the early birds when they come in.
I use a different vhost-config which I enable while I do my maintenance.
It looks like this:
,----
| <IfModule mod_ssl.c>
| <VirtualHost *:443>
| ServerAdmin webmaster@example.com
| ServerName webmail.example.com
|
| DocumentRoot /srv/www/webmail.example.com-maint/
| <Directory />
| Options FollowSymLinks
| AllowOverride None
| </Directory>
| <Directory /srv/www/webmail.example.com/>
| Options Indexes FollowSymLinks MultiViews
| AllowOverride All
| Order allow,deny
| allow from all
| </Directory>
|
| ErrorLog ${APACHE_LOG_DIR}/webmail.example.com-maint-error.log
|
| LogLevel warn
|
| CustomLog ${APACHE_LOG_DIR}/webmail.example.com-maint-access.log combined
|
| ErrorDocument 404 /error.html
|
| SSLEngine on
|
| SSLCertificateFile /etc/ssl/certs/webmail.example.com.pem
| SSLCertificateKeyFile /etc/ssl/private/webmail.example.com.key
| SSLCertificateChainFile /etc/ssl/certs/webmail.example.com.pem
|
| <FilesMatch ".(cgi|shtml|phtml|php)$">
| SSLOptions +StdEnvVars
| </FilesMatch>
|
| BrowserMatch "MSIE [2-6]"
| nokeepalive ssl-unclean-shutdown
| downgrade-1.0 force-response-1.0
| BrowserMatch "MSIE [17-9]" ssl-unclean-shutdown
|
| </VirtualHost>
| </IfModule>
`----
The documentroot /srv/www/webmail.example.com-maint/ only contains the error.html and the needed CSS and JPG files. The ErrorDocument option redirects every access attempt to that error document.
It is important to do it this way and _not_ using any redirects, because the latter will poison the cache of the browser with "fake" HTTP-200 responses.
Grüße, Sven.
On 08/25/2015 07:35 PM, Benny Pedersen wrote:
Michael Orlitzky skrev den 2015-08-25 22:46:
When Roundcube comes back up, delete index.html.
thats more simple yes, but http://www.example.net/index.php will still work
apache need another documentroot for being down
If this solution prompts the evolution of a better fool, then you can always rename index.php to index.nopegoaway.
symlink!
On Tue, Aug 25, 2015 at 8:40 PM, Michael Orlitzky michael@orlitzky.com wrote:
On 08/25/2015 07:35 PM, Benny Pedersen wrote:
Michael Orlitzky skrev den 2015-08-25 22:46:
When Roundcube comes back up, delete index.html.
thats more simple yes, but http://www.example.net/index.php will still work
apache need another documentroot for being down
If this solution prompts the evolution of a better fool, then you can always rename index.php to index.nopegoaway.
Roundcube Users mailing list users@lists.roundcube.net http://lists.roundcube.net/mailman/listinfo/users