wiki audit 2026-10-03: cliprox second job confirmed (2/3), work-mailbox incident page, streak recount, skills-index 341->344

This commit is contained in:
Tony0410
2026-10-03 00:18:51 +08:00
parent 0c4ed3350b
commit b9f69d3f75
10 changed files with 296 additions and 51 deletions
@@ -0,0 +1,82 @@
---
title: Work Outlook Mailbox — Basic Auth Disabled (Himalaya)
type: incident
status: open
created: 2026-10-03
updated: 2026-10-03
verified_on: 2026-10-03
confidence: high
tags: [email, himalaya, office365, oauth2, auth, incident]
sources: [live-himalaya-debug, live-tcp-probe, cron-session-20260826_180700_a94098]
---
# Work Outlook Mailbox — Basic Auth Disabled (Himalaya)
**Window:** first detected 2026-08-26; re-verified 2026-10-03 (still open)
**Status:** Open — blocked pending an OAuth2 path
**Impact:** The agent cannot read or send from the `work` mailbox. Every work-email
request is unanswerable until this is fixed.
## Symptoms
```
$ himalaya envelope list -a work -s 1 --debug
email::imap: connecting to IMAP server outlook.office365.com:993
email::imap: supported auth mechanisms mechanisms=[XOAuth2]
Error: cannot build IMAP client
Caused by:
login not supported
```
Network is *not* the problem: the TCP/TLS connection to `outlook.office365.com:993`
succeeds and the server completes its capability exchange. The failure is purely at
the authentication step.
## Root Cause
The account is configured for **password** auth, and the tenant no longer accepts it.
- Account config (`~/.config/himalaya/config.toml`, account `work`) sets
`backend.type = "imap"`, `backend.host = "outlook.office365.com"`,
`backend.login = "<work address>"`, `backend.auth.type = "password"` with a helper
script at `~/.config/himalaya/get_work_outlook_password.sh`. The SMTP half
(`smtp.office365.com:587`, start-tls) is configured the same way.
- The server advertises **only `XOAuth2`**. Basic authentication for IMAP/SMTP is
disabled at the Microsoft 365 tenant level — a WA Government tenant policy, not a
wrong password. Retrying the helper script can never succeed.
This is the same wall hit on 2026-08-26: the account was set up that day and every
attempt to list folders returned a login rejection. The mechanism list has since
*narrowed* — on 2026-08-26 the tenant rejected plain password login despite still
advertising `AUTH=PLAIN`; by 2026-10-03 only `XOAuth2` is offered.
The personal Gmail account is unaffected (Google still allows app passwords), so
this is specific to the work tenant.
## Fix Options
1. **OAuth2 for the `work` account — preferred, needs Anthony.** Requires an app
registration in the tenant with the IMAP/SMTP delegated scopes, a device-code or
interactive sign-in to mint a refresh token, and `backend.auth.type = "oauth2"`
plus token storage in the himalaya config. Anthony's normal login is
*number-matching approval in Microsoft Authenticator* — that is exactly the
device-code / interactive OAuth flow, so the human step is one sign-in. Tenant
admin consent may still be required.
2. **Tenant exception** — an admin re-enables IMAP/SMTP basic auth or issues an app
password. Unlikely on a government tenant; not worth relying on.
3. **Alternative access path** — browser/OWA session or Microsoft Graph. Slower and
more brittle than native IMAP, but needs no admin action.
4. **Do not** keep re-running the password helper. The credential is not the fault.
## Consequences for tooling
- The `email/mail-lookup` skill resolves the account by address before searching; when
the requested mail lives in the `work` mailbox it will reach this failure and must
report the account as unreachable rather than as "no results found".
- No cron job currently polls the work mailbox, so this is an on-demand capability
gap, not a background failure.
## Related
- [[systems/messaging-integrations]]
- [[systems/scheduled-tasks]]
@@ -3,8 +3,8 @@ title: OmniRoute cliprox/terra Connection Auth Expiry
type: incident
status: open
created: 2026-10-02
updated: 2026-10-02
verified_on: 2026-10-02
updated: 2026-10-03
verified_on: 2026-10-03
confidence: high
tags: [cron, omniroute, provider, auth, incident]
sources: [live-jobs-json, cron-executions-db, cron-session-cron_06a0a1c8704f_20261001_100040]
@@ -14,7 +14,7 @@ sources: [live-jobs-json, cron-executions-db, cron-session-cron_06a0a1c8704f_202
**Window:** first failure 2026-10-01 10:00 AWST (job schedule `0 10 * * 4`); still open
**Status:** Open — not self-recovering
**Impact:** 1 job failed outright; 2 more enabled jobs share the same model pin and are expected to fail on their next runs
**Impact:** 2 of the 3 exposed jobs have now failed on this pin; 1 remains untested
## Symptoms
@@ -49,9 +49,32 @@ Three enabled jobs run `cliprox/terra`:
| `1b23c042e17d` | Enhanced Weekend Adventure Planner | `0 8 * * 5` | `telegram:-1004321904721:102` |
| `79f9bb41ecf5` | Weekly Review | `0 18 * * 0` | `origin` |
Only `06a0a1c8704f` had run by 2026-10-02; the other two are exposed on their next
Only `06a0a1c8704f` had run by 2026-10-02; the other two were exposed on their next
scheduled runs.
### Second job confirmed — 2026-10-02 08:00
`1b23c042e17d` (**Enhanced Weekend Adventure Planner**, Fri 08:00) failed on schedule
with the *identical* error:
```
HTTP 401: [openai-compatible-chat-2ea20f63-…] All 1 connection(s) authentication expired
— please reconnect in the dashboard
```
That is now **2 of 3** exposed jobs confirmed broken on this pin, which retires any
doubt about a one-job fluke. Its failure notice was delivered to topic `102`, so the
run is recorded `delivery_outcome: delivered` even though no plan was produced.
`79f9bb41ecf5` (**Weekly Review**, Sun 18:00) is the last untested one — its only
recorded run was 2026-09-27 18:00, four days *before* the connection lapsed, and it
succeeded. Its next scheduled run is **2026-10-04 18:00**, which will confirm whether
the expiry is tenant-wide for this model or per-connection.
Corollary worth recording: a job can be recorded as **successfully delivered while
having failed**, because the failure notice itself is a deliverable. Read the
execution `status`, not the delivery outcome, when judging whether a job worked.
## Fix
Re-authenticate the OmniRoute connection for `cliprox/terra`. Do **not** repin these jobs