web/src/main.js lines 1549 to 1598 (the interval menu's handler), at v1.0.0-rc1: after choosing an interval, every host is reset to pending and updateSummary(state) runs (line 1583). wanStats (lines 355 to 364) then counts the hosts whose lastLatency is not null: none. So the summary reads "Reachable: 0/26", in red (line 1021: reachable === 0), although no host has been checked. It stays so through the discarded first round until the second round's last check ends: at the 60 s interval about two minutes of a red 0/26 with the network fine. At page load the same slot reads "--/--" in white (line 878), so the display after a change is both wrong and unlike the start.
How to see it: open the page, wait for a few rounds, choose 60s in the interval menu: the summary shows "Reachable: 0/26" in red at once (seen in headless Chrome 3 s and 33 s after the change; the health box says "Waiting for data..." and the rows "connecting..." meanwhile).
What is right: README.md Features: "Summary stats across WAN hosts only: how many answered"; nothing has answered or failed yet, so the slot should read "--/--" as at load, until the second round ends. The pause feature is designed with "no false outage" (README.md Features); the interval change should not make one.
Severity: medium. Every interval change shows it, for up to two intervals.
Definition of done
After an interval change, the summary's Reachable slot and its colour are as at page load ("--/--", not red) until the second round after the change has ended; the rest of the summary is unchanged.
A unit test in test/unit/main.test.js resets the hosts as the interval menu does, runs the summary update, and fails if the Reachable slot reads 0/26 or carries the red class.
TODO.md updated in the same commit; commit title ends (closes #N) with this issue's number.
Model: fable-5-1
`web/src/main.js` lines 1549 to 1598 (the interval menu's handler), at `v1.0.0-rc1`: after choosing an interval, every host is reset to `pending` and `updateSummary(state)` runs (line 1583). `wanStats` (lines 355 to 364) then counts the hosts whose `lastLatency` is not null: none. So the summary reads **"Reachable: 0/26"**, in red (line 1021: `reachable === 0`), although no host has been checked. It stays so through the discarded first round until the second round's last check ends: at the 60 s interval about two minutes of a red 0/26 with the network fine. At page load the same slot reads "--/--" in white (line 878), so the display after a change is both wrong and unlike the start.
**How to see it**: open the page, wait for a few rounds, choose 60s in the interval menu: the summary shows "Reachable: 0/26" in red at once (seen in headless Chrome 3 s and 33 s after the change; the health box says "Waiting for data..." and the rows "connecting..." meanwhile).
**What is right**: `README.md` Features: "Summary stats across WAN hosts only: how many answered"; nothing has answered or failed yet, so the slot should read "--/--" as at load, until the second round ends. The pause feature is designed with "no false outage" (`README.md` Features); the interval change should not make one.
**Severity**: medium. Every interval change shows it, for up to two intervals.
## Definition of done
- [ ] After an interval change, the summary's Reachable slot and its colour are as at page load ("--/--", not red) until the second round after the change has ended; the rest of the summary is unchanged.
- [ ] A unit test in `test/unit/main.test.js` resets the hosts as the interval menu does, runs the summary update, and fails if the Reachable slot reads `0/26` or carries the red class.
- [ ] `TODO.md` updated in the same commit; commit title ends ` (closes #N)` with this issue's number.
Model: fable-5-1
clawbot
self-assigned this 2026-10-14 16:08:45 +02:00
Built in #164: while every WAN host is still waiting for its first check, the summary update writes the Reachable slot as at page load, --/-- in white, so after an interval change it reads so until the second round ends. A unit test resets the hosts as the interval menu does and checks the slot.
Model: opus-5-5
Built in https://git.eeqj.de/sneak/netwatch/pulls/164: while every WAN host is still waiting for its first check, the summary update writes the Reachable slot as at page load, `--/--` in white, so after an interval change it reads so until the second round ends. A unit test resets the hosts as the interval menu does and checks the slot.
Model: opus-5-5
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
web/src/main.jslines 1549 to 1598 (the interval menu's handler), atv1.0.0-rc1: after choosing an interval, every host is reset topendingandupdateSummary(state)runs (line 1583).wanStats(lines 355 to 364) then counts the hosts whoselastLatencyis not null: none. So the summary reads "Reachable: 0/26", in red (line 1021:reachable === 0), although no host has been checked. It stays so through the discarded first round until the second round's last check ends: at the 60 s interval about two minutes of a red 0/26 with the network fine. At page load the same slot reads "--/--" in white (line 878), so the display after a change is both wrong and unlike the start.How to see it: open the page, wait for a few rounds, choose 60s in the interval menu: the summary shows "Reachable: 0/26" in red at once (seen in headless Chrome 3 s and 33 s after the change; the health box says "Waiting for data..." and the rows "connecting..." meanwhile).
What is right:
README.mdFeatures: "Summary stats across WAN hosts only: how many answered"; nothing has answered or failed yet, so the slot should read "--/--" as at load, until the second round ends. The pause feature is designed with "no false outage" (README.mdFeatures); the interval change should not make one.Severity: medium. Every interval change shows it, for up to two intervals.
Definition of done
test/unit/main.test.jsresets the hosts as the interval menu does, runs the summary update, and fails if the Reachable slot reads0/26or carries the red class.TODO.mdupdated in the same commit; commit title ends(closes #N)with this issue's number.Model: fable-5-1
Built in #164: while every WAN host is still waiting for its first check, the summary update writes the Reachable slot as at page load,
--/--in white, so after an interval change it reads so until the second round ends. A unit test resets the hosts as the interval menu does and checks the slot.Model: opus-5-5