openclaw - 💡(How to fix) Fix Not a duplicate -- FIX NEEDED [1 comments, 2 participants]

I'm not able to re-open the issue, but it is not a duplicate of 52577

Thanks for the context here. I swept through the related work, and this is now duplicate or superseded.

Close as duplicate/superseded: the real remaining symptom is Discord guild inbound messages not being received after restart on v2026.5.7, which is already tracked by the open canonical Discord READY/guild-inbound issue; the missing session row by itself is documented as expected for a quiet channel.

So I’m closing this here and keeping the remaining discussion on the canonical linked item.

<details> <summary>Review details</summary>

Best possible solution:

Keep the canonical v2026.5.7 Discord guild-inbound failure on https://github.com/openclaw/openclaw/issues/79794, and route reconnect-gap replay/catch-up work through https://github.com/openclaw/openclaw/issues/52577.

Do we have a high-confidence way to reproduce the issue?

No. The reporter gives concrete restart steps and logs, but this review did not run a live Discord restart; source inspection only confirms that the missing session row is not a valid liveness signal and that the remaining no-response symptom matches the open canonical tracker.

Is this the best way to solve the issue?

Yes. Closing this as a duplicate is the best cleanup because pre-creating session rows would contradict the documented sessions contract, while the real Discord guild-inbound regression already has an open canonical issue.

Security review:

Security review needs attention: The issue attachment appears to expose a Discord bot-token-looking prefix and should be redacted after token rotation.

• [medium] Rotate and redact the Discord bot token attachment The linked ISSUE_REPORT attachment includes a bot-token-looking prefix in the sample config. Even partial credential disclosure should be treated as sensitive; the reporter should rotate that Discord bot token and replace the attachment with a fully redacted version. Confidence: 0.88

What I checked:

• live issue context: The issue and attached report describe OpenClaw v2026.5.7, Discord channel resolution after restart, no new session row, and no guild message events after tagging the bot.
• canonical duplicate: The open report https://github.com/openclaw/openclaw/issues/79794 tracks the same central v2026.5.7 Discord symptom: bot online/initialized but guild channel messages are not received after startup or restart.
• related remaining reconnect tracker: The open report https://github.com/openclaw/openclaw/issues/52577 is the existing tracker for Discord reconnect/READY-window inbound loss and replay/catch-up behavior, so any reconnect-gap fix should preserve this issue's context there rather than split a new tracker.
• session row is not liveness: Current docs say openclaw sessions, Gateway sessions.list, and agent sessions_list report stored conversation rows, not channel socket health; a quiet Discord account can be healthy after restart without a Discord session row until an inbound or outbound event. Public docs: docs/cli/channels.md. (docs/cli/channels.md:46, 98e65198789e)
• channel config hypothesis checked: Current startup resolves Discord guild/channel allowlists, assigns the resolved guildEntries, and passes those same entries into the Discord message handler; source inspection does not support the attachment's hypothesis that resolved channel config is dropped after restart. (extensions/discord/src/monitor/provider.ts:278, 98e65198789e)
• channel override matching checked: Current channel config lookup uses channel ID/name/slug plus parent fallback and resolveDiscordShouldRequireMention prefers channelConfig.requireMention over guild defaults, so an ID-matched requireMention: false channel should not be forced back to the guild mention requirement. (extensions/discord/src/monitor/allow-list.ts:447, 98e65198789e)

Likely related people:

• steipete: Local blame and GitHub commit history show recent work across the Discord provider startup, allowlist/channel config, gateway lifecycle, and channel-health docs used to triage this report. (role: recent area contributor; confidence: high; commits: 92c84bbee463, 0362b7582430, 827b0de0ce74; files: extensions/discord/src/monitor/provider.ts, extensions/discord/src/monitor/allow-list.ts, extensions/discord/src/monitor/provider.lifecycle.ts)
• Satoshi: GitHub history shows a recent Discord startup-readiness hardening commit in the lifecycle area related to gateway restart and reply behavior. (role: adjacent readiness contributor; confidence: medium; commits: e259938e9643; files: extensions/discord/src/monitor/provider.lifecycle.ts)
• IVY-AI-gif: GitHub history attributes the earlier Discord IDENTIFY race fix to this contributor, which is adjacent to the canonical READY/guild-inbound regression linked from this cleanup. (role: prior related fix contributor; confidence: medium; commits: 5adf9d261944; files: extensions/discord/src/monitor/gateway-plugin.ts, extensions/discord/src/monitor/provider.proxy.test.ts)

Codex review notes: model gpt-5.5, reasoning high; reviewed against 98e65198789e.

</details> <!-- clawsweeper-review item=81262 -->

Originally posted by @clawsweeper in #81262