fix(gateway): keep Telegram topic bindings aligned with compression children#34409
Merged
Conversation
…hildren Telegram DM topic bindings persist (chat_id, thread_id) -> session_id in SQLite so reopening a topic resumes the right Hermes session. When compression rotated session_entry.session_id mid-turn, the binding row stayed pointed at the pre-compression parent. On the next inbound message in that topic the gateway reloaded the oversized parent transcript, retriggering preflight compression — sometimes in a loop. Two-pronged fix: 1. `_sync_telegram_topic_binding(source, entry, *, reason)` helper called immediately after each of the three session_id rotation sites in _handle_message_with_agent (hygiene compression, agent-result compression rotation, /compress command). Keeps future bindings fresh. 2. Read-path self-heal: when resolving an existing topic binding, walk SessionDB.get_compression_tip() forward and switch_session to the descendant instead of the stored parent. Rewrites the binding row to the tip so subsequent messages skip the walk. Heals existing stale state on the next user message without requiring a gateway restart. Skipped from competing PRs as not load-bearing for the bug: - advance_session_after_compression SessionStore primitive (#26204/ #28870/#33416) — preserves end_reason='compression' analytics nicety but doesn't affect routing correctness. - Cached-agent eviction on session_id mismatch — _compress_context() already mutates tmp_agent.session_id on the cached object so the in-memory agent self-corrects. - Startup repair pass (#33416) — redundant once the read path heals on the next message; one-line CLI follow-up can address bindings for topics users never reopen. Closes #20470, #29712, #33414. Acknowledges work in #23195 (@litvinovvo), #26204 (@bizyumov), #28870 (@donrhmexe), #29713 (@hehehe0803), #29945 (@eugeneb1ack), #33416 (@bizyumov).
Contributor
🔎 Lint report:
|
This was referenced May 29, 2026
KKT-OPT
pushed a commit
to KKT-OPT/hermes-agent
that referenced
this pull request
May 31, 2026
…hildren (NousResearch#34409) Telegram DM topic bindings persist (chat_id, thread_id) -> session_id in SQLite so reopening a topic resumes the right Hermes session. When compression rotated session_entry.session_id mid-turn, the binding row stayed pointed at the pre-compression parent. On the next inbound message in that topic the gateway reloaded the oversized parent transcript, retriggering preflight compression — sometimes in a loop. Two-pronged fix: 1. `_sync_telegram_topic_binding(source, entry, *, reason)` helper called immediately after each of the three session_id rotation sites in _handle_message_with_agent (hygiene compression, agent-result compression rotation, /compress command). Keeps future bindings fresh. 2. Read-path self-heal: when resolving an existing topic binding, walk SessionDB.get_compression_tip() forward and switch_session to the descendant instead of the stored parent. Rewrites the binding row to the tip so subsequent messages skip the walk. Heals existing stale state on the next user message without requiring a gateway restart. Skipped from competing PRs as not load-bearing for the bug: - advance_session_after_compression SessionStore primitive (NousResearch#26204/ NousResearch#28870/NousResearch#33416) — preserves end_reason='compression' analytics nicety but doesn't affect routing correctness. - Cached-agent eviction on session_id mismatch — _compress_context() already mutates tmp_agent.session_id on the cached object so the in-memory agent self-corrects. - Startup repair pass (NousResearch#33416) — redundant once the read path heals on the next message; one-line CLI follow-up can address bindings for topics users never reopen. Closes NousResearch#20470, NousResearch#29712, NousResearch#33414. Acknowledges work in NousResearch#23195 (@litvinovvo), NousResearch#26204 (@bizyumov), NousResearch#28870 (@donrhmexe), NousResearch#29713 (@hehehe0803), NousResearch#29945 (@eugeneb1ack), NousResearch#33416 (@bizyumov).
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
2 participants
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.
Summary
Telegram DM topic conversations no longer get stuck in compression loops or reload pre-compression parent transcripts.
When context compression rotates
session_entry.session_idmid-turn, the (chat_id, thread_id) → session_id row in SQLite stayed pointed at the parent. The next inbound message in that topic loaded the oversized parent transcript and re-ran preflight compression — sometimes in a loop.Closes #20470, #29712, #33414.
Changes
gateway/run.py: new_sync_telegram_topic_binding(source, entry, *, reason)helper called after each of the threesession_entry.session_id = ...rotation sites (hygiene compression, agent-result compression rotation,/compresscommand).gateway/run.py: read-path self-heal — when an existing binding is found, walkSessionDB.get_compression_tip()forward and switch_session to the descendant. Rewrites the binding row to the tip so subsequent messages skip the walk. Heals already-stale state on next user message; no restart required.tests/gateway/test_telegram_topic_mode.py: regression test exercises the full inbound flow with a real SessionDB + parent→child compression chain; verifies the route advances to the child and the binding row is rewritten.Skipped from competitor PRs (intentionally)
advance_session_after_compressionSessionStore primitive (fix(gateway): publish compression route advances #26204/fix: harden Telegram topic routing across session splits #28870/fix(gateway): repair Telegram topic compression routes #33416) — preservesend_reason='compression'over a manualswitch_sessionevent. Analytics nicety, not routing correctness._compress_context()already mutatestmp_agent.session_idon the cached object, so the in-memory agent self-corrects without explicit eviction.Validation
switch_sessioncalled withparent-session)child-session, binding rewritten)tests/gateway/test_telegram_topic_mode.pytests/gateway/(5,974 tests)Synthesis credit
Reviewed all six competing PRs to settle on the minimal load-bearing fix; closing each with credit:
Infographic