Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Only populate secondary endpoints if the config is set #7

Open
wants to merge 2 commits into
base: main
Choose a base branch
from

Conversation

KenGordon
Copy link

@KenGordon KenGordon commented Jan 11, 2024

First change to support Azure port. It came about as a result of the Azure KMS implementation being different - there is only ever one KMS and so there is never a SECONDARY_COORDINATOR_PRIVATE_KEY_ENDPOINT.

This change does two main things:

  1. Modifies TrustedServersConfigClient::HasParameter() to return false when parameter is present in config_entries_map but has a default empty value.

    • This is because all flags passed into the constructor of TrustedServersConfigClient are added to the map with the empty value
    • So if SECONDARY_COORDINATOR_PRIVATE_KEY_ENDPOINT isn't specified, TrustedServersConfigClient::HasParameter(SECONDARY_COORDINATOR_PRIVATE_KEY_ENDPOINT) incorrectly returns true
  2. When initialising PrivateKeyFetcherFactory, reorganises the logic so that the secondary endpoint is only included when SECONDARY_COORDINATOR_PRIVATE_KEY_ENDPOINT is set

These two changes mean that when combining primary and secondary endpoints into a single vector in the data plane, we don't get a secondary endpoint with a blank URL

Copy link

google-cla bot commented Jan 11, 2024

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

@DomAyre DomAyre force-pushed the fix-secondary-endpoints branch 6 times, most recently from bdae18f to 6a2d1a3 Compare January 12, 2024 11:00
jasarora-google pushed a commit that referenced this pull request Oct 12, 2024
### ⚠ BREAKING CHANGES

* All sellers and buyers must update the js_num_workers parameter to udf_num_workers in terraform
* Sellers must wait for all the integrated buyers to update to 4.1.0 before updating to accomodate serialization changes in the GetBidsRequest.

### Features

* [BYOB] Add buyer code fetcher for executables
* [BYOB] Add dispatch client for generateBid
* [BYOB] Add GCP build setup for Bidding service
* [BYOB] Add gRPC reactor for generateBid binary
* [BYOB] Add info about runtime flags to demo/README.md
* [BYOB] Add interface for clients that load code
* [BYOB] Add to Bidding service critical path
* Add a generic model validator interface for Tensorflow and PyTorch
* Add PrivateAggregation protos to generateBid spec
* add project setup scripts
* add proto support for updateIfOlderThanMs
* BFE parses perInterestGroupData, returns updateIfOlderThanMs data to SFE
* Compress SFE <> BFE comms
* Create new inference dashboard for GCP
* Enable chaffing on SFE prod mode builds
* Enable connecting to TKV via an External LB
* Enable dynamic partition for inference metrics
* Enable model validation for PyTorch
* Enable TCMalloc for B&A inference
* Export all available models as metric
* Freeze a TensorFlow model ([bcc798c]( )), closes [#7]( )
* Implement a PyTorch model validator
* Implement a TensorFlow model validator
* Rename JS_NUM_WORKERS to UDF_NUM_WORKERS
* SFE sends updateIfOlderThanMs data to client
* update start_bidding with minimal defaults and local file flag

### Bug Fixes

* BYOB generateBid should only parse first bid from binary response
* Ensure gRPC calls don't hang on error
* Fix js_execution.error.count metric
* Fix memory leak in BFE client
* Fix readonly declarations in shell script
* Fix version bump in release script
* Load request files in the init function
* Reduce absl::GetFlag() call in GetBidsReactor
* Renames prev_wins_ms to prev_wins
* Set inference sidecar GRPC keepalive timeout
* update tls cert commands for GCP setup
* Use std::hash instead of absl::hash

Bug: N/A
Change-Id: Ia7d37c5a7e5be0e4c91c26c8fce4cc15683946a3
GitOrigin-RevId: 671dc23f8e04b711f7efe95907b66002e5a3b019
jasarora-google added a commit that referenced this pull request Oct 14, 2024
### ⚠ BREAKING CHANGES

* All sellers and buyers must update the js_num_workers parameter to udf_num_workers in terraform
* Sellers must wait for all the integrated buyers to update to 4.1.0 before updating to accomodate serialization changes in the GetBidsRequest.

### Features

* [BYOB] Add buyer code fetcher for executables
* [BYOB] Add dispatch client for generateBid
* [BYOB] Add GCP build setup for Bidding service
* [BYOB] Add gRPC reactor for generateBid binary
* [BYOB] Add info about runtime flags to demo/README.md
* [BYOB] Add interface for clients that load code
* [BYOB] Add to Bidding service critical path
* Add a generic model validator interface for Tensorflow and PyTorch
* Add PrivateAggregation protos to generateBid spec
* add project setup scripts
* add proto support for updateIfOlderThanMs
* BFE parses perInterestGroupData, returns updateIfOlderThanMs data to SFE
* Compress SFE <> BFE comms
* Create new inference dashboard for GCP
* Enable chaffing on SFE prod mode builds
* Enable connecting to TKV via an External LB
* Enable dynamic partition for inference metrics
* Enable model validation for PyTorch
* Enable TCMalloc for B&A inference
* Export all available models as metric
* Freeze a TensorFlow model ([bcc798c]( )), closes [#7]( )
* Implement a PyTorch model validator
* Implement a TensorFlow model validator
* Rename JS_NUM_WORKERS to UDF_NUM_WORKERS
* SFE sends updateIfOlderThanMs data to client
* update start_bidding with minimal defaults and local file flag

### Bug Fixes

* BYOB generateBid should only parse first bid from binary response
* Ensure gRPC calls don't hang on error
* Fix js_execution.error.count metric
* Fix memory leak in BFE client
* Fix readonly declarations in shell script
* Fix version bump in release script
* Load request files in the init function
* Reduce absl::GetFlag() call in GetBidsReactor
* Renames prev_wins_ms to prev_wins
* Set inference sidecar GRPC keepalive timeout
* update tls cert commands for GCP setup
* Use std::hash instead of absl::hash

Bug: N/A
GitOrigin-RevId: 671dc23f8e04b711f7efe95907b66002e5a3b019
Change-Id: I16a2f5227cfdc86469fffcbd494c68b2b029a5ef
@mihnjong-l mihnjong-l requested review from jasarora-google and removed request for jasarora-google December 11, 2024 18:37
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants