Screen turns off 30 seconds before it locks #43

New issue

Closed

opened 2026-09-26 07:29:06 +08:00 by Breadway · 2 comments

Breadway commented

2026-09-26 07:29:06 +08:00

Owner

Problem

iso/airootfs/etc/skel/.config/hypr/hypridle.conf: DPMS off at 480 s, lock at 510 s. For those 30 seconds, moving the mouse wakes an unlocked desktop, which feels like it should have been locked.

Fix

Lock at or before DPMS off (e.g. lock 470 s, DPMS off 480 s), or make the DPMS listener run loginctl lock-session first.

Acceptance

When the screen wakes from DPMS-off, it's always locked.

Source: BOS v0.6.2 source review (tag dated 2026-09-19). Read AGENTS.md first. One issue per branch/PR. Add a CHANGELOG [Unreleased] entry.

## Problem `iso/airootfs/etc/skel/.config/hypr/hypridle.conf`: DPMS off at 480 s, lock at 510 s. For those 30 seconds, moving the mouse wakes an unlocked desktop, which feels like it should have been locked. ## Fix Lock at or before DPMS off (e.g. lock 470 s, DPMS off 480 s), or make the DPMS listener run `loginctl lock-session` first. ## Acceptance When the screen wakes from DPMS-off, it's always locked. --- Source: BOS v0.6.2 source review (tag dated 2026-09-19). Read AGENTS.md first. One issue per branch/PR. Add a CHANGELOG `[Unreleased]` entry.

Breadway added the

security

idle

labels

2026-09-26 07:29:06 +08:00

Breadway referenced this issue from a pull request that will close it,

2026-09-30 03:50:07 +08:00

Screen turns off 30 seconds before it locks #62

Breadway commented

2026-09-30 03:50:14 +08:00

Author

Owner

PR: #62

Moved the idle lock timeout to 470 seconds, before DPMS turns the display off at 480 seconds, and added a changelog entry. git diff --check passes. I checked the installed Hypridle CLI; it has no config-only validation option, so I did not start it in the active session.

PR: https://git.breadway.dev/Breadway/bos/pulls/62 Moved the idle lock timeout to 470 seconds, before DPMS turns the display off at 480 seconds, and added a changelog entry. `git diff --check` passes. I checked the installed Hypridle CLI; it has no config-only validation option, so I did not start it in the active session.

Breadway commented

2026-09-30 03:52:36 +08:00

Author

Owner

Reviewed PR #62 at head d7c648322245555b6114773618fe9b4fea82cb7f.

Acceptance map:

When the display wakes after DPMS-off, the session is locked. iso/airootfs/etc/skel/.config/hypr/hypridle.conf:27-31 moves loginctl lock-session to 470 seconds, before the DPMS-off listener at :20-25 (480 seconds). The 10-second lead satisfies the requested ordering, and activity waking the display after the DPMS timeout still returns to the already locked session.
Record the behavior change in [Unreleased]. CHANGELOG.md:19-21 adds the corresponding entry.

There are no tests in the diff. This is a declarative timeout-ordering change; the PR reports git diff --check passed, but the idle daemon was not started for runtime validation. The latest head is based directly on current main; Forgejo reports it mergeable. No commit statuses are published, consistent with BOS AGENTS.md, which says ordinary PR pushes do not run CI.

Verdict: Reviewed, looks ready to merge. BOS is on the release train, so I am leaving the merge to the user.

Reviewed PR #62 at head `d7c648322245555b6114773618fe9b4fea82cb7f`. Acceptance map: 1. When the display wakes after DPMS-off, the session is locked. `iso/airootfs/etc/skel/.config/hypr/hypridle.conf:27-31` moves `loginctl lock-session` to 470 seconds, before the DPMS-off listener at `:20-25` (480 seconds). The 10-second lead satisfies the requested ordering, and activity waking the display after the DPMS timeout still returns to the already locked session. 2. Record the behavior change in `[Unreleased]`. `CHANGELOG.md:19-21` adds the corresponding entry. There are no tests in the diff. This is a declarative timeout-ordering change; the PR reports `git diff --check` passed, but the idle daemon was not started for runtime validation. The latest head is based directly on current `main`; Forgejo reports it mergeable. No commit statuses are published, consistent with BOS `AGENTS.md`, which says ordinary PR pushes do not run CI. Verdict: Reviewed, looks ready to merge. BOS is on the release train, so I am leaving the merge to the user.

Breadway closed this issue

2026-09-30 03:53:44 +08:00