Instead of adding active prpfiles for every one we encounter, we need to
build up a complete "default" Environment and then ask it what the active
profiles are. Implemented in ConfigFileApplicationListener.
Fixes gh-198
Main user-facing interface is still Counter/GaugeService but the
back end behind that has more options. The Default*Services write
metrics to a MetricWriter and there are some variants of that, and
also variants of MetricReader (basic read-only actions).
MetricRepository is now a combination of MetricReader, MetricWriter
and some more methods that make it a bit more repository like.
There is also a MultiMetricReader and a MultiMetricRepository for
the common case where metrics are stored in related (often open
ended) groups. Examples would be complex metrics like histograms
and "rich" metrics with averages and statistics attached (which
are both closed) and "field counters" which count the occurrences
of values of a particular named field or slot in an incoming message
(e.g. counting Twitter hastags, open ended).
In memory and redis implementations are provided for the repositories.
Generally speaking the in memory repository should be used as a
local buffer and then scheduled "exports" can be executed to copy
metric values accross to a remote repository for aggregation.
There is an Exporter interface to support this and a few implementations
dealing with different strategies for storing the results (singly or
grouped).
Codahale metrics are also supported through the MetricWriter interface.
Currently implemented through a naming convention (since Codahale has
a fixed object model this makes sense): metrics beginning with "histogram"
are Histograms, "timer" for Timers, "meter" for Meters etc.
Support for message driven metric consumption and production are provided
through a MetricWriterMessageHandler and a MessageChannelMetricWriter.
No support yet for pagination in the repositories, or for HATEOAS style
HTTP endpoints.
We get more control over the handling and in particular the registration
of the endpoint this way. It was practically impossible to disable the
AgentServlet bean when in a parent context of the management server
because of lifecyce issues - you don't know that the user wants a
separate management server until too late.
This approach also makes it possible to test with spring-test MVC
support.
You can contribute additional HttpMessageConverters
by simply adding beans of that type in a Spring Boot
context. If a bean you add is of a type that would have been included
by default anyway (like MappingJackson2HttpMessageConverter for JSON
conversions) then it will replace the default value. A convenience
bean is provided of type MessageConverters (always available if you
use the default MVC configuration) which has some useful methods to
access the default and user-enhanced message converters (useful, for
example if you want to manually inject them into a custom
RestTemplate).
There are also some convenient configuration shortcuts for Jackson2.
The smallest change that might work is to just add beans of type
Module to your context. They will be registered with the default
ObjectMapper and then injected into the default message
converter. In addition, if your context contains any beans of type
ObjectMapper then all of the Module beans will be registered with
all of the mappers.
Previously if a user happened to provide an @EnableWebSecurity bean
the SecurityProperties would not be created, which is fine until you
add the Actuator (which needs them). Fixed by adding an explicit
SecurityProperties @Bean if not already present.
When management endpoints are on a different port the HandlerMappings
are restricted to a single EndpointHandlerMapping, so the error
controller (which is a normal @Controller with @RequestMappings) does
not get mapped.
Fixed by addinga shim Endpoint on "/error" that delegates to the
ErrorController (which interface picks up an extra method).
ServerProperties formerly had an @OnMissingBeanCondition
that didn't restrict the hierarchy. It also asserts that
the current context (not including parents) contains such
a bean. This led to an inevitable failure when there was
an existing instance in the parent context.
Fixed by a) searching only the current context, b) not
adding a ServerProperties bean if the context is not a
web app.