There was an update on 01 February for ClearOS on one of my clients and it has broken the user profile section, they're no certificates available to download and the web page is extremely messed up example of what it looks like if I manually type in app/user_profile
I can't do the upload on the forum haveu ploaded to external site:
http://s27.postimg.org/5kxr0vg4j/clearos2.png
I'm unsure why this is happening all other clearos clients seem to be fine apart from this one I'm going to do some testing tonight and see if I delete certificate manager and remove openvpn if this will resolve it.
I can't do the upload on the forum haveu ploaded to external site:
http://s27.postimg.org/5kxr0vg4j/clearos2.png
I'm unsure why this is happening all other clearos clients seem to be fine apart from this one I'm going to do some testing tonight and see if I delete certificate manager and remove openvpn if this will resolve it.
Share this post:
Responses (10)
-
Accepted Answer
Unfortunately this wouldn't solve the issue as port 81 would still be used via the firewall it would need to be a different port. I think the easiest solution is just expose 81 when a client needs to re-download their certificates to that particular server and then disable it at all other times.
Thanks for your helping solving this issue Ben. -
Accepted Answer
Instead of hacking source code (../apps/base/libraries/Access_Control.php is what's tripping you up), you may want to consider setting port 81 not to use HTTPS. This would be ill-advised if you were accessing these over a public network, but if it's just for LAN-LAN or you always access Webconfig from outside over a VPN tunnel, I suppose it's a valid work around:
In
/usr/clearos/sandbox/etc/httpd/conf.d/framework.conf
In the default virtual host using port 81 to bind to, change:
SSLEngine on
SSLProtocol all -SSLv2
SSLCipherSuite ALL:!ADH:!EXPORT:!SSLv2:RC4+RSA:+HIGH:+MEDIUM:+LOW
SSLCertificateFile /usr/clearos/sandbox/etc/httpd/conf/server.crt
SSLCertificateKeyFile /usr/clearos/sandbox/etc/httpd/conf/server.key
SetEnvIf User-Agent ".*MSIE.*" nokeepalive ssl-unclean-shutdown downgrade-1.0 force-response-1.0
to
#SSLEngine on
#SSLProtocol all -SSLv2
#SSLCipherSuite ALL:!ADH:!EXPORT:!SSLv2:RC4+RSA:+HIGH:+MEDIUM:+LOW
#SSLCertificateFile /usr/clearos/sandbox/etc/httpd/conf/server.crt
#SSLCertificateKeyFile /usr/clearos/sandbox/etc/httpd/conf/server.key
#SetEnvIf User-Agent ".*MSIE.*" nokeepalive ssl-unclean-shutdown downgrade-1.0 force-response-1.0
and restart Webconfig:
service webconfig restart
Then access Webconfig like follows:
http://x.x.x.x:81/app/user_profile -
Accepted Answer
Ben,
That seems to work. However that can be a slight problem. We have several virtual machines running on a data centre and sometimes they need to share IP addresses.
The servers port to access it remotely is 810 but I have kept it's original port as 81 via iptables using port forwarding on the data centre server. This seemed to work and was fine for a very long time until the 6.5 update, is this how it needs to be from now on or is there a way to change clearos's port to not use port 81 as a base? -
Accepted Answer
You can get to this state if you are accessing Webconfig as user without HTTPS on port 82. Try HTTPS protocol on port 81.
Eg:
https://x.x.x.x:81/app/user_profile
B. -
Accepted Answer
Thought I could try and just un-install theme-default and re-install clearos seems to want to remove everything to do so. I've moved the client over to GDrive as they have GApps however I am concerned about other clients we have with ClearOS that this might happen to them some being high profile clients... I'm going to keep the VM open for a while if you can maybe suggest a few things that might work to resolve it? -
Accepted Answer
-
Accepted Answer
-
Accepted Answer
-
Accepted Answer
Please login to post a reply
You will need to be logged in to be able to post a reply. Login using the form on the right or register an account if you are new here.
Register Here »