Troubleshoot the Data explorer

By default, only 10 series per metric are shown.

To increase the series number

1. In the query editor, select the + button.
2. Select Limit.
3. Set Limit to 1, 10, 20, or 100.

The root cause of this issue is often the same as for the question above. The metric series are limited to a certain number.

Let's assume you query builtin:host.cpu.usage and builtin:host.cpu.idle split by dt.entity.host. For both metrics, the top 10 hosts are requested per default. But the top 10 of the CPU usage % metric probably diverges from the top 10 of the CPU idle metric, leading to empty cells in the table.

Note that even if you increase the limit to 100, the issue may occur if there are more than 100 dimensions for the queried timeframe.

When the Data explorer charts single values, it may happen that the value is too high for particular metrics. For example, the metric Host availability % is affected.

The workaround in such cases is to go to the Code tab of the Data explorer and remove the :avg:auto:sort(value(avg,descending)):limit(100) from the query string. For the Host availability %, the query string is then just builtin:host.availability:splitBy().

The root cause of this problem is that the Data explorer always performs a value extraction (by calling :avg in the example above) before the values are aggregated. Due to the value extraction, the semantics of the metric are lost and the values are summed up instead of being averaged.

The metrics builtin:host.availability and builtin:pgi.availability are based on timeseries data provided by the OneAgents, whereas the availability values shown on the Host and Process pages are calculated by dedicated events. Thus, the values may slightly diverge.

In the future, the availability shown in the Host and Process pages will be based on the availability metrics as well.