fix(pb-envoy): reduce round trips for ws connection

[MasterPtato]

Description

Please include a summary of the changes and the related issue. Please also include relevant motivation and context.

Type of change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

How Has This Been Tested?

Please describe the tests that you ran to verify your changes.

Checklist:

• My code follows the style guidelines of this project
• I have performed a self-review of my code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes

[MasterPtato]

• fix(epoxy): use vbare #4616
• fix: dont upsert for ensure if missing #4615
• fix(envoy): require runner config for ws conn #4612
• fix(pb): backport serverless fix #4611
• fix(pb): optimize actor key allocation #4610
• fix: move runner configs to epoxy #4605
• chore(epoxy): clean up kv get #4604
• fix(pb-envoy): reduce round trips for ws connection #4603 👈 (View in Graphite)
• fix(envoy-client): fix handling actor event ack #4599
• fix(envoy-client): add tls #4595 : 1 other dependent PR (#4597 )
• chore: fix driver test suite #4591
• chore: tunnel auth #4588
• chore: remove udb as a dependency of envoy-client #4587
• chore: revert actor v2 hack #4586
• fix: fix pool metadata protocol version refresh, rivetkit native ws #4585
• feat: US-001 - Define SqliteKv trait in rivetkit-sqlite-native #4584
• feat(rust/engine-runner): minimal rust engine runner #4289
• chore: delete kv channel #4583
• feat(engine): rust runner #4582
• chore: remove non-engine drivers #4574 : 1 other dependent PR (#4581 )
• feat: dynamic actors #4397
• fix: driver test suite & sqlite reliability #4559
• fix: remove serverless token #4572
• fix: misc token fixes #4569
• fix(pb): clean up actor stop decision handling #4566
• fix(envoy): use global instance, add signal handlers #4565
• chore: misc fixes, add pb snapshot test #4561
• fix(cache): make in memory cache global #4560
• main

This stack of pull requests is managed by Graphite. Learn more about stacking.

[claude]

PR Review: fix(pb-envoy): reduce round trips for ws connection

This is a well-structured optimization that eliminates a blocking round trip during WebSocket connection establishment by moving version and envoy_key from an initial protocol message to URL query parameters. The concurrent tokio::try_join! is clean. A few issues are worth addressing.

BUG 1: is_serverless hardcoded to false in Conn
File: engine/packages/pegboard-envoy/src/conn.rs

The computed is_serverless value is correctly used to call report_success, but never propagated to the Conn struct (is_serverless: false is hardcoded). This means the guard in ws_to_tunnel_task.rs that calls pegboard::ops::envoy::expire only for serverful envoys on ToRivetStopping will incorrectly fire for serverless envoys too, since conn.is_serverless is always false. Fix: use is_serverless (the computed variable) in the struct literal.

BUG 2: Double util::timestamp::now() call for create_ts
File: engine/packages/pegboard-envoy/src/conn.rs

In the new-envoy branch of the DB transaction, create_ts is assigned from one call to timestamp::now(), but tx.write calls timestamp::now() again for the stored value. These two timestamps will differ. Since create_ts is later used to construct ActiveEnvoyKey, ActiveEnvoyByNameKey, and EnvoyLoadBalancerIdxKey, those keys will not match what was written to create_ts_key.
Fix: tx.write(&create_ts_key, create_ts)?

MINOR 1: Misleading metric name
File: engine/packages/pegboard-envoy/src/conn.rs

RECEIVE_INIT_PACKET_DURATION now measures namespace resolution time rather than the time to receive an init packet (no longer a blocking step). The name should be updated (e.g. NAMESPACE_RESOLVE_DURATION) to keep dashboards and alerts accurate.

MINOR 2: prepopulate_actor_names always Some in client
File: engine/sdks/rust/envoy-client/src/connection.rs

prepopulate_actor_names: Some(prepopulate_map) is sent even when the map is empty, triggering the write loop in handle_metadata unconditionally. Sending None when the map is empty avoids the unnecessary DB transaction.

NOTE: envoy_key now in URL query string

Moving envoy_key from a binary protocol message to a URL query parameter means it may appear in access logs on proxies or load balancers. Since these are server-to-server connections over TLS this is a low-risk trade-off, but worth auditing log retention and scrubbing policies to ensure keys are not persisted in plain-text logs.

Summary: The core optimization is sound and the protocol rename (ToRivetInit to ToRivetMetadata) is cleaner semantically. BUG 1 (is_serverless hardcode) and BUG 2 (double timestamp::now()) should be followed up on.

[MasterPtato]

Merge activity

• Apr 13, 12:21 AM UTC: A user started a stack merge that includes this pull request via Graphite.
• Apr 13, 12:29 AM UTC: Graphite rebased this pull request as part of a merge.
• Apr 13, 12:30 AM UTC: @MasterPtato merged this pull request with Graphite.