Page 1 of 1

Load Balanced NLS broken since 1.4.2 upgrade

Posted: Tue Jul 26, 2016 10:29 am
by roddergreg
We are using a VIP address to load balance the NLS instances. This was working previously before upgrade to 1.4.2

We are using rewrite condition to redirect http to https in the ssl.conf file /etc/httpd/conf.d/ssl.conf

Code: Select all

RewriteEngine on
RewriteCond $1 !^(index\.php|scripts|media|app|js|css|img|font|vendor|config.js)
RewriteCond %{REQUEST_FILENAME} !-f
RewriteCond %{REQUEST_FILENAME} !-d
RewriteRule nagioslogserver/(.*)$ /var/www/html/nagioslogserver/www/index.php/$1 [L,QSA]
</VirtualHost>

Did something change due to the upgrade?

Re: Load Balanced NLS broken since 1.4.2 upgrade

Posted: Tue Jul 26, 2016 10:36 am
by hsmith
We're going to need some information about what's broken in order to proceed. Most of the known issues at the moment can be resolved by clearing your browser cache.

Re: Load Balanced NLS broken since 1.4.2 upgrade

Posted: Tue Jul 26, 2016 10:45 am
by roddergreg
Clearing browser cache doesn't help i'ved tried firefox, chrom and IE.

I have a VIP address called prhsyslog. It round robins the http traffic between 3 servers. apache then redirects to https. Once I login i click one of the menus its asks me to login again. every time i try accessing a different link in web interface.

Re: Load Balanced NLS broken since 1.4.2 upgrade

Posted: Wed Jul 27, 2016 9:42 am
by hsmith
Are these the same instructions you followed to configure SSL on this system? https://assets.nagios.com/downloads/nag ... Server.pdf

Re: Load Balanced NLS broken since 1.4.2 upgrade

Posted: Wed Jul 27, 2016 1:50 pm
by roddergreg
I followed the instructions just as the document states. The only difference is we are using our own certificate that worked perfectly before upgrading to NLS 1.4.2. Instead of a self signed cert as the document states.

Re: Load Balanced NLS broken since 1.4.2 upgrade

Posted: Wed Jul 27, 2016 2:00 pm
by rkennedy
I have a VIP address called prhsyslog.
Could you please explain this? Usually, the setup we see on our end is people put a load balancer in front of their cluster.

Also, please post your /var/log/httpd/error_log and /var/log/httpd/ssl_error_log files for us to look at.

Re: Load Balanced NLS broken since 1.4.2 upgrade

Posted: Wed Jul 27, 2016 3:51 pm
by roddergreg
I've enabled persistency on our load balancer by source address. Which pins a users web interface to specific instance of the nagioslogserver. This fixes the issue.

Did something change from NLS 1.4.1 to NLS 1.4.2? to make this change necessary.

Re: Load Balanced NLS broken since 1.4.2 upgrade

Posted: Wed Jul 27, 2016 4:21 pm
by rkennedy
We had a few security vulnerabilities that were fixed in 1.4.2, and one of them was related to sessions. This was fixed, and probably why you experienced the issue that forced you to enable persistence.

Re: Load Balanced NLS broken since 1.4.2 upgrade

Posted: Thu Jul 28, 2016 8:27 am
by roddergreg
Thanks for the information. You can close this issue now if you like.

Greg

Re: Load Balanced NLS broken since 1.4.2 upgrade

Posted: Thu Jul 28, 2016 9:41 am
by mcapra
Closing this up