Skip to content

Releases: BuoyantIO/bb

v0.0.7

21 May 22:00
971aeae
Compare
Choose a tag to compare
  • Update go dependencies (#20)
  • Update repo to more modern tooling (#19)
  • Add go.mod and go.sum files (#18)
  • Fix ci due to httpbin url (#16)
  • Update README.md (#15)
  • Fix readme reference to examples (#14)
  • Upgrade dep script to 0.5.0 (#13)
  • Better handling of contexts/deadlines/timeouts (#12)

v0.0.5

21 Sep 21:43
c578036
Compare
Choose a tag to compare
  • Fix gRPC clients to honor downstream-timeout.
  • Remove grpc-downstream-authority flag in favor of grpc-proxy flag.
    If grpc-proxy is set, the target URL will be set by grpc-proxy and the
    :authority header will be set by grpc-downstream-server. If grpc-proxy
    is not set, the target URL will be set by grpc-downstream-server.
  • Change log-level default from debug to info.
  • Modify gRPC Client IDs to include grpc-proxy and grpc-downstream-server.
  • Additional debug logging around broadcast requests.
  • Bump go-grpc to 1.15.0, golang/protobuf to v1.2.0, logrus to
    v1.0.6.
  • Bump Docker build Golang version to 1.11.0.

v0.0.4

10 Sep 17:54
9b82cf0
Compare
Choose a tag to compare
  • Introduce grpc-downstream-authority flag, to enable setting authority separately from grpc-downstream-server.

v0.0.3

22 May 18:04
Compare
Choose a tag to compare
  • Update docs and example files to latest release.

v0.0.2

22 May 18:01
ab87349
Compare
Choose a tag to compare
  • Introduce terminate-after flag, which instructs the process to shutdown
    after a specified number of requests.
  • Introduce graceful shutdown. Upon receiving a shutdown message via SIGTERM, or
    via terminate-after, call shutdown on each server, allowing requests to
    drain.

v0.0.1

22 Mar 21:06
b876676
Compare
Choose a tag to compare

bb 0.0.1 is the first public release of bb

  • This release supports HTTP 1.1 and gRPC.
  • Available strategies are: broadcast channel, point-to-point channel, terminus, and HTTP egress
  • Allows users to define a percentage of requests that should fail and a duration to wait for
    before processing requests.
  • This release has been tested locally on Mac OS and on both Google Kubernetes Engine and
    Minikube.