Annotate the context with feature flags when handling flux queries in influxdb.
Taking advantage of this in flux end-to-end tests. Using a custom flagger that
can set overrides based on the test case that is about to be run, allowing us
to enable features in the end-to-end tests.
This removes the spec and updates the lang package usage to make use of
passing in the runtime as a parameter.
It removes all direct dependendencies on the flux runtime from the http
package.
The tasks subsystem will now use the flux language service to parse and
evaluate flux instead of directly interacting with the parser or
runtime. This helps break the dependency on the libflux parser for the
base influxdb package.
This includes the task notification packages which were changed at the
same time.
Prior to this change influxql requests were sent to the same back end as Flux queries.
This MAY not always be the case. Now InfluxQL queries are specifically routed to the InfluxQLService.
In the case of this OSS build the FluxService and InfluxQLService are the same.
* chore: Remove several instances of WithLogger
* chore: unexport Logger fields
* chore: unexport some more Logger fields
* chore: go fmt
chore: fix test
chore: s/logger/log
chore: fix test
chore: revert http.Handler.Handler constructor initialization
* refactor: integrate review feedback, fix all test nop loggers
* refactor: capitalize all log messages
* refactor: rename two logger to log
* feat(tracing): dont trace spans with full URL path names in ExtractFromHTTPRequest
* chore(multiple): replace all occurrences of julienschmidt/httprouter with influxdata/httprouter
If client requests with Accept-Encoding: gzip, then
compress response and return with Content-Encoding: gzip.
This increases the server-side load by about 12%, but saves network bandwidth.
benchmark old ns/op new ns/op delta
Benchmark_Query_no_gzip-4 123609 137885 +11.55%
benchmark old allocs new allocs delta
Benchmark_Query_no_gzip-4 149 150 +0.67%
benchmark old bytes new bytes delta
Benchmark_Query_no_gzip-4 14297 15205 +6.35%
If we handle the flux errors in the query controller, it makes it so we
are handling the errors in the location where the happen rather than at
a layer further up the stack.
This should simplify it so the errors are handled in this single
location instead.
feat(http): add prometheus counters for tracking write/query usage
feat(http/metric): add metric recoder for recording http metrics
feat(prometheus): implement metric.Recorder for prometheus metrics
fix(prometheus): remove erroneous fmt.Printlns
feat(http): add prometheus registry to API backend
This was done as exposing prometheus metrics to a higher level was quite
difficult. It was much simple to simply pass the registry down to
anything that needs it.
feat(cmd/influxd/launcher): pass prom registry in on api backend
feat(http): collect metrics for write and query endpoints
This was much messier than I would have preferred. Future work is
outlined in TODOs.
review(influxdb): rename metric.Metric to metric.Event
Co-Authored-By: Kelvin Wang <sherkrainwang@gmail.com>
Co-Authored-By: Michael Desa <mjdesa@gmail.com>
Co-Authored-By: Christopher M. Wolff <chris.wolff@influxdata.com>
Previously the APIBackend understood only a ProxyQueryService,
but it needs to understand that there are two implementations of the
ProxyQueryService one for handling InfluxQL queries and one for handling
Flux queries. The names of the fields have been updated to make this
clear. As well as the FluxBackend is now initialized using the
FluxService explicitly.
Adrian Thurston
jlapacik
Brett Buddin
Jonathan A. Sternberg
Nathaniel Cook
Lorenzo Affetti
Johnny Steenbergen
Gavin Cabbage
George
Jacob Marble
docmerlin (j. Emrys Landivar)
Jakub Bednář
Chris Goller
Kelvin Wang
kelwang
Christopher M. Wolff
Adam
Mark Rushakoff
Michael Desa
Todd Persen