LIVEDUMP ---------------- livedump is a small utility that can dump configuration and status information from a running Nagios core in order to transport that data to another core - effectively replicating the data. This can be used as a very efficient replacement for NSCA. Advantages: - Much more performant - Creates your Nagios configuration for the passive services - No obsess_over.. configuration neccessary on the source host Disadvantages: - Introduces some latency How to setup livedump --------------------- Copy the file livedump to a convenient place. Make sure that ../livestatus/api/python/livestatus.py is either in the same directory as livedump or somewhere in the Python path. Using livedump -------------- The first step is to extract the configuration from the source system using. This step is needed whenever your configuration of hosts or services changes. NSCA users will now this. This is easily done by: ./livedump -TC > some_file.cfg and then copying that file to your objects or conf.d directory on your target nagios. Note: The option -C will add some templates to the configuration that are used by the created host and service definitions. If you import data from more than on source those template will be duplicate. Use -T in that case in order to avoid duplication. Now restart your target system and new hosts and services will appear and be in pending state. Now create a cronjob that does the following every 1 or 5 minutes: 1. ./livedump > ca1b2c3 2. copy that file to the target system into the checkresults directory (in OMD this is ~/tmp/nagios/checkresults) 3. After a correct copy touch the file ca1b2c3.ok in that directory Nagios will now read in that file (which is containing all current host and service states). The filename is arbitrary, but has to be 7 characters in length, starting with a "c". Filtering --------- In each operation mode livedump allows to filter the number of data to be dumped. This is done by adding options -H and -S for host and service livestatus headers. The following example will only dump hosts and services of the host group "foo". Please make sure that all hosts that are needed by the dumped services are also dumped: ./livedump -H "Filter: host_groups > foo" Note: The -H headers will also be added to the service queries. Make sure that all columns are prefixed with host_ in these. If you are using this is in conjunction with NSCA transport and do not wish to transfer the templated configuration, you can use the filter to only export services that have "obsess_over_service = 1" set.
