There are many items that could be potentially used as a "ping" item. For example, having DB monitoring with the "SELECT 1" query allows to "ping" database. This item will return "1" in cases when DB is up and running and credentials are valid. The downside of such a check is the fact any error during the data collection state will cause an item to become not supported. And there is really no way to get some "0" in case of an error. This is the limiting factor of using some of the items as "ping" monitoring.

Another example is the HTTP monitoring (HTTP agent) that will become unsupported if the Web server responds with the 500 or any other HTTP error.

Some of the "ping" metrics are introduced in agent2, but this limited to agent2 and specific metrics.

It would be great if items would have an extra flag (or a whole new preprocessing step) like "Override errors". The idea behind this flag (or preprocessing step) is to allow bypassing the errors during data collection and setting item value to some X. This would allow using any of the items as "ping" items.