Still on Drupal 7? Security support for Drupal 7 ended on 5 January 2025. Please visit our Drupal 7 End of Life resources page to review all of your options.
Does anybody using Nagios for monitoring drupal deployments? Nagios already has check_http and check_curl plugins but I can only validate index.php. I was thinking about test tool called WebInject. It has a Nagios plugin http://www.webinject.org/plugin.html . You can create custom xml config files for different deployments and check if specific function of web application is working. Here my problem comes. How to login to drupal using WebInject and howto navigate on the website.
I started exploring Drupal and I must say it is great! But I'm not sure to invest time and build a Drupal site because of speed and scalability. Recently I found an article "The Drupal Database Design is a Comedy of Errors"
Rather than go through a bunch of queries or rather hundreds of queries, I will just say that the database design has no relational qualities that would optimize and speed up the system what so ever.
Our organization is serving many web applications from a 'cluster' of LAMP servers. They are sharing storage on a RAID with GFS and Red Hat Cluster Suite is providing high-availability failover. All three machines are running apache. A pair of load-balancing directors are forwarding requests to them using LVS (ipvsadm and such). The idea is to make the three machines in the cluster behave as one, transparent to the end user.
Two of these machines are configured with Master-Master MySQL replication. This means that each is the other's slave, so that any write operation to either database server is [almost] immediately duplicated to the other. Each of our sites are pointed to localhost, assuming that data will be consistant across servers. For all of our in-house web applications this works fine, providing both load-balancing and high-availability.
Drupal, however, is able to 'break' this replication. Every once in a while, replication encounters errors and stops. They're always duplicate key errors for INSERT statements, and they look something like this one: