docs: describe continuous-mode memory bounds #125

2026-08-20T06:25:33Z

Lochnair commented

2026-08-20 06:25:33 +00:00

(Migrated from github.com)

Summary

Document bounded continuous-mode client and statistics retention
Explain the finite-mode exact-median memory tradeoff
Note bounded/lossy managed events and TUI history limits

## Summary - Document bounded continuous-mode client and statistics retention - Explain the finite-mode exact-median memory tradeoff - Note bounded/lossy managed events and TUI history limits

👍 1

chatgpt-codex-connector[bot] (Migrated from github.com) reviewed

2026-08-20 06:37:13 +00:00

chatgpt-codex-connector[bot] (Migrated from github.com) left a comment

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: e5f83ff592

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

### 💡 Codex Review Here are some automated review suggestions for this pull request. **Reviewed commit:** `e5f83ff592` <details> <summary>ℹ️ About Codex in GitHub</summary> [Your team has set up Codex to review pull requests in this repo](https://chatgpt.com/codex/cloud/settings/general). Reviews are triggered when you - Open a pull request for review - Mark a draft as ready - Comment "@codex review". If Codex has suggestions, it will comment; otherwise it will react with 👍. Codex can also answer questions or update the PR. Try commenting "@codex address that feedback". </details>

README.md

				`@@ -72,6 +72,39 @@ irtt-client netperf-eu.bufferbloat.net:2112 --duration 0`

chatgpt-codex-connector[bot] (Migrated from github.com) commented

2026-08-20 06:37:13 +00:00

Document pending-capacity exhaustion as terminal

For the CLI/TUI managed path, when a high-rate continuous target accumulates 4,096 still-pending probes before the first timeout, PendingMap::preflight_insert returns PendingLimitExceeded, and ManagedClientTask::poll_one_send routes that error to begin_open_session_failure rather than retrying after space is freed. The target is therefore drained and terminated as resource-exhausted (and a single-target run fails), so this sentence incorrectly describes the limit as transient backpressure; document the terminal behavior instead.

AGENTS.md reference: AGENTS.md:L78-L78

Useful? React with 👍 / 👎.

**![P2 Badge](https://img.shields.io/badge/P2-yellow?style=flat) Document pending-capacity exhaustion as terminal** For the CLI/TUI managed path, when a high-rate continuous target accumulates 4,096 still-pending probes before the first timeout, `PendingMap::preflight_insert` returns `PendingLimitExceeded`, and `ManagedClientTask::poll_one_send` routes that error to `begin_open_session_failure` rather than retrying after space is freed. The target is therefore drained and terminated as resource-exhausted (and a single-target run fails), so this sentence incorrectly describes the limit as transient backpressure; document the terminal behavior instead. AGENTS.md reference: [AGENTS.md:L78-L78](https://github.com/Lochnair/irtt-rs/blob/e5f83ff5925c652ea996f65dfa6d5238d497f1f5/AGENTS.md#L78-L78) Useful? React with 👍 / 👎.

chatgpt-codex-connector[bot] (Migrated from github.com) commented

2026-08-20 06:37:13 +00:00

Qualify continuous memory growth as bounded

Continuous-mode retained state does grow with probes until its bounds are reached: successful replies populate the IPDV tracker up to 4,096 sequences, classification stores fill up to their limits, and the TUI graph grows to 100,000 samples per target. The later paragraphs acknowledge these limits, so saying the state “does not grow” is contradictory and can understate warm-up memory consumption; state that it does not grow without bound instead.

AGENTS.md reference: AGENTS.md:L78-L78

Useful? React with 👍 / 👎.

**![P2 Badge](https://img.shields.io/badge/P2-yellow?style=flat) Qualify continuous memory growth as bounded** Continuous-mode retained state does grow with probes until its bounds are reached: successful replies populate the IPDV tracker up to 4,096 sequences, classification stores fill up to their limits, and the TUI graph grows to 100,000 samples per target. The later paragraphs acknowledge these limits, so saying the state “does not grow” is contradictory and can understate warm-up memory consumption; state that it does not grow *without bound* instead. AGENTS.md reference: [AGENTS.md:L78-L78](https://github.com/Lochnair/irtt-rs/blob/e5f83ff5925c652ea996f65dfa6d5238d497f1f5/AGENTS.md#L78-L78) Useful? React with 👍 / 👎.

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: Lochnair-Labs/irtt-rs#125