I want to do distributed monitoring of remote networks where the connection back to the central server is very bandwidth limited.
Documentation says that no item related information (like last check and last value of items) is sent across nodes, but history and events are sent every 10 seconds.
What exactly is contained in the history that gets sent back to the master?
Assuming events are infrequent, how do you calculate the bandwidth requirement for history updates every 10 seconds from, for example, one remote node monitoring 10 servers with 100 items each?
PS. Apparently NodeNoHistory and NodeNoEvent parameters aren't supported yet, so no way of turning off the history updates (https://support.zabbix.com/browse/ZBX-2802).
thanks!
Documentation says that no item related information (like last check and last value of items) is sent across nodes, but history and events are sent every 10 seconds.
What exactly is contained in the history that gets sent back to the master?
Assuming events are infrequent, how do you calculate the bandwidth requirement for history updates every 10 seconds from, for example, one remote node monitoring 10 servers with 100 items each?
PS. Apparently NodeNoHistory and NodeNoEvent parameters aren't supported yet, so no way of turning off the history updates (https://support.zabbix.com/browse/ZBX-2802).
thanks!
Comment