Commit Graph

2 Commits

Author SHA1 Message Date
Reed Allman
51ff7caeb2 Bye bye openapi (#1081)
* add DateTime sans mgo

* change all uses of strfmt.DateTime to common.DateTime, remove test strfmt usage

* remove api tests, system-test dep on api test

multiple reasons to remove the api tests:

* awkward dependency with fn_go meant generating bindings on a branched fn to
vendor those to test new stuff. this is at a minimum not at all intuitive,
worth it, nor a fun way to spend the finite amount of time we have to live.
* api tests only tested a subset of functionality that the server/ api tests
already test, and we risk having tests where one tests some thing and the
other doesn't. let's not. we have too many test suites as it is, and these
pretty much only test that we updated the fn_go bindings, which is actually a
hassle as noted above and the cli will pretty quickly figure out anyway.
* fn_go relies on openapi, which relies on mgo, which is deprecated and we'd
like to remove as a dependency. openapi is a _huge_ dep built in a NIH
fashion, that cannot simply remove the mgo dep as users may be using it.
we've now stolen their date time and otherwise killed usage of it in fn core,
for fn_go it still exists but that's less of a problem.

* update deps

removals:

* easyjson
* mgo
* go-openapi
* mapstructure
* fn_go
* purell
* go-validator

also, had to lock docker. we shouldn't use docker on master anyway, they
strongly advise against that. had no luck with latest version rev, so i locked
it to what we were using before. until next time.

the rest is just playing dep roulette, those end up removing a ton tho

* fix exec test to work

* account for john le cache
2018-06-21 11:09:16 -07:00
Reed Allman
9eaf824398 add jaeger support, link hot container & req span (#840)
* add jaeger support, link hot container & req span

* adds jaeger support now with FN_JAEGER_URL, there's a simple tutorial in the
operating/metrics.md file now and it's pretty easy to get up and running.
* links a hot request span to a hot container span. when we change this to
sample at a lower ratio we'll need to finagle the hot container span to always
sample or something, otherwise we'll hide that info. at least, since we're
sampling at 100% for now if this is flipped on, can see freeze/unfreeze etc.
if they hit. this is useful for debugging. note that zipkin's exporter does
not follow the link at all, hence jaeger... and they're backed by the Cloud
Empire now (CNCF) so we'll probably use it anyway.

* vendor: add thrift for jaeger
2018-03-13 15:57:12 -07:00