Conversation
rita-aga
force-pushed
the
fix/ci-timeout-60m
branch
from
013b8c3 to
0a1437a
Compare
rita-aga
added a commit
that referenced
this pull request
Mar 15, 2026
rita-aga
added a commit
that referenced
this pull request
Mar 15, 2026
rita-aga
added a commit
that referenced
this pull request
Mar 15, 2026
rita-aga
added a commit
that referenced
this pull request
Mar 15, 2026
rita-aga
added a commit
that referenced
this pull request
Mar 15, 2026
* ci: bump test timeout to 60m (45m too tight for cold cache) (#31) * feat: instrument trajectory, evolution, and unmet intents for Logfire Add tracing spans to all trajectory entry creation sites (5 sites), evolution engine operations, sentinel checks, and persistence layer. Unmet intents now emit tracing::warn! with structured fields. 13 files changed, 687 insertions. Co-authored-by: Codex <codex@openai.com> * ci: update readability baseline * style: cargo fmt * ci: retrigger after staging worktree fix * fix: collapse nested if to satisfy clippy --------- Co-authored-by: Codex <codex@openai.com>
rita-aga
added a commit
that referenced
this pull request
Mar 16, 2026
* ci: bump test timeout to 60m (45m too tight for cold cache) (#31) * ci: run CI on staging PRs too (#29) * fix: OS app installation persistence across restarts (#25) * fix: OS app installation persistence across restarts Co-Authored-By: Codex <noreply@openai.com> * fix: clippy useless_vec + update readability baseline * ci: retrigger * ci: retrigger with target cache * fix: drop entity_index write lock before metrics call to prevent deadlock populate_index_from_store held entity_index.write() while calling record_server_state_metrics, which tries to acquire entity_index.read(). std::sync::RwLock is not reentrant, so this self-deadlocked, causing cargo test to hang indefinitely in CI. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Codex <noreply@openai.com> Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> * fix: persist bootstrap verification cache to prevent OOM on Railway (#35) * fix: persist bootstrap verification cache to prevent OOM on Railway Built-in specs (system + agent) ran the Z3/Stateright/proptest verification cascade on every boot because the hash-gated cache had a read path but no write path — results were never persisted to Turso. On Railway's 512MB containers this caused OOM. Wire the write side: after bootstrap verifies specs, upsert them into the specs table with their content hash and mark verified=1. Subsequent boots hit the cache and skip the cascade entirely. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * chore: cargo fmt Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: update readability baseline after verification cache fix Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * ci: retrigger --------- Co-authored-by: rita-aga <rita.mirai@gmail.com> Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> * fix: OS app install merges specs instead of replacing + handle 409 on cloud provision Bug 1: provision_cloud_database now recovers from 409 Conflict by fetching existing DB hostname and creating a fresh auth token. Makes ensure_tenant truly idempotent. Bug 2: install_os_app now merges incoming app CSDL with existing tenant CSDL and passes merge=true to bootstrap_tenant_specs. Installing a second OS app into a tenant no longer nukes the first app's entity types. Tests: ensure_tenant idempotency, 409 cloud recovery (wiremock), multi-app install merge + idempotency. --------- Co-authored-by: Codex <noreply@openai.com> Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> Co-authored-by: sesh nalla <39490039+nerdsane@users.noreply.github.com>
rita-aga
added a commit
that referenced
this pull request
Mar 16, 2026
…re (#47) * ci: bump test timeout to 60m (45m too tight for cold cache) (#31) * ci: run CI on staging PRs too (#29) * fix: OS app installation persistence across restarts (#25) * fix: OS app installation persistence across restarts Co-Authored-By: Codex <noreply@openai.com> * fix: clippy useless_vec + update readability baseline * ci: retrigger * ci: retrigger with target cache * fix: drop entity_index write lock before metrics call to prevent deadlock populate_index_from_store held entity_index.write() while calling record_server_state_metrics, which tries to acquire entity_index.read(). std::sync::RwLock is not reentrant, so this self-deadlocked, causing cargo test to hang indefinitely in CI. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Codex <noreply@openai.com> Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> * fix: persist bootstrap verification cache to prevent OOM on Railway (#35) * fix: persist bootstrap verification cache to prevent OOM on Railway Built-in specs (system + agent) ran the Z3/Stateright/proptest verification cascade on every boot because the hash-gated cache had a read path but no write path — results were never persisted to Turso. On Railway's 512MB containers this caused OOM. Wire the write side: after bootstrap verifies specs, upsert them into the specs table with their content hash and mark verified=1. Subsequent boots hit the cache and skip the cascade entirely. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * chore: cargo fmt Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: update readability baseline after verification cache fix Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * ci: retrigger --------- Co-authored-by: rita-aga <rita.mirai@gmail.com> Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> * docs(skills): add toolsmith vision, approval UX, and meta-loop architecture - temper-agent: Add 'You Build Your Own Tools' section — specs are hot-reloaded, not PRs - temper-agent: Add 'Walk the states, don't skip them' — no shortcut actions - temper-agent: Add meta-loop agent concept — improving the platform itself - temper-agent: Add approval UX guidance — chat buttons > Observe UI > CLI - temper-openclaw: Add Discord/Telegram/Slack approval bridge pattern - temper-openclaw: Add hot-reload emphasis — spec changes are NOT PRs - temper-openclaw: Add meta-loop agent architecture section - temper-openclaw: Document the distinction: spec changes (hot-reload) vs platform changes (PRs) --------- Co-authored-by: Codex <noreply@openai.com> Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> Co-authored-by: sesh nalla <39490039+nerdsane@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
45m was too tight — tests pass but doctests push past 45 min on cold cache.