The [Jolokia](http://jolokia.org) _agent_ and _proxy_ input plugins collect JMX metrics from an HTTP endpoint using Jolokia's [JSON-over-HTTP protocol](https://jolokia.org/reference/html/protocol.html).
The `jolokia2_proxy` input plugin reads JMX metrics from one or more _targets_ by interacting with a [Jolokia proxy](https://jolokia.org/features/proxy.html) REST endpoint.
Each `metric` declaration generates a Jolokia request to fetch telemetry from a JMX MBean.
| Key | Required | Description |
|----------------|----------|-------------|
| `mbean` | yes | The object name of a JMX MBean. MBean property-key values can contain a wildcard `*`, allowing you to fetch multiple MBeans with one declaration. |
| `paths` | no | A list of MBean attributes to read. |
| `tag_keys` | no | A list of MBean property-key names to convert into tags. The property-key name becomes the tag name, while the property-key value becomes the tag value. |
| `tag_prefix` | no | A string to prepend to the tag names produced by this `metric` declaration. |
| `field_name` | no | A string to set as the name of the field produced by this metric; can contain substitutions. |
| `field_prefix` | no | A string to prepend to the field names produced by this `metric` declaration; can contain substitutions. |
Use `*` wildcards against `mbean` property-key values to create distinct series by capturing values into `tag_keys`.
```toml
[[inputs.jolokia2_agent.metric]]
name = "jvm_garbage_collector"
mbean = "java.lang:name=*,type=GarbageCollector"
paths = ["CollectionTime", "CollectionCount"]
tag_keys = ["name"]
```
Since `name=*` matches both `G1 Old Generation` and `G1 Young Generation`, and `name` is used as a tag, the preceeding `jvm_garbage_collector``metric` declaration produces two metrics.
Use `tag_prefix` along with `tag_keys` to add detail to tag names.
```toml
[[inputs.jolokia2_agent.metric]]
name = "jvm_memory_pool"
mbean = "java.lang:name=*,type=MemoryPool"
paths = ["Usage", "PeakUsage", "CollectionUsage"]
tag_keys = ["name"]
tag_prefix = "pool_"
```
The preceeding `jvm_memory_pool``metric` declaration produces six metrics, each with a distinct `pool_name` tag.
```
jvm_memory_pool,pool_name=Compressed\ Class\ Space PeakUsage.max=1073741824,PeakUsage.committed=3145728,PeakUsage.init=0,Usage.committed=3145728,Usage.init=0,PeakUsage.used=3017976,Usage.max=1073741824,Usage.used=3017976 1503764025000000000
jvm_memory_pool,pool_name=G1\ Eden\ Space CollectionUsage.max=-1,PeakUsage.committed=56623104,PeakUsage.init=56623104,PeakUsage.used=53477376,Usage.max=-1,Usage.committed=49283072,Usage.used=19922944,CollectionUsage.committed=49283072,CollectionUsage.init=56623104,CollectionUsage.used=0,PeakUsage.max=-1,Usage.init=56623104 1503764025000000000
jvm_memory_pool,pool_name=G1\ Old\ Gen CollectionUsage.max=1073741824,CollectionUsage.committed=0,PeakUsage.max=1073741824,PeakUsage.committed=1017118720,PeakUsage.init=1017118720,PeakUsage.used=137032208,Usage.max=1073741824,CollectionUsage.init=1017118720,Usage.committed=1017118720,Usage.init=1017118720,Usage.used=134708752,CollectionUsage.used=0 1503764025000000000
jvm_memory_pool,pool_name=G1\ Survivor\ Space Usage.max=-1,Usage.init=0,CollectionUsage.max=-1,CollectionUsage.committed=7340032,CollectionUsage.used=7340032,PeakUsage.committed=7340032,Usage.committed=7340032,Usage.used=7340032,CollectionUsage.init=0,PeakUsage.max=-1,PeakUsage.init=0,PeakUsage.used=7340032 1503764025000000000
Use substitutions to create fields and field prefixes with MBean property-keys captured by wildcards. In the following example, `$1` represents the value of the property-key `name`, and `$2` represents the value of the property-key `topic`.
The preceeding `kafka_topic``metric` declaration produces a metric per Kafka topic. The `name` Mbean property-key is used as a field prefix to aid in gathering fields together into the single metric.
