From 7955be70ea7751c68d6a9b2ca3683ee697a38726 Mon Sep 17 00:00:00 2001 From: Craig Jennings Date: Tue, 7 Jul 2026 18:24:12 -0500 Subject: docs: add maintenance console design + interactive prototypes The design doc replaces the install-health framing in system-monitor-design-ideas.org with a maintenance console: dense metric subpanels behind a persistent selector, a doctor that streams into a results wall, determinate remedies only (guard-armed updates, curation lifecycles for noise, listeners, containers, and orphans), rotary band selectors on the tall subpanels, and a CLI-first build with a four-layer test strategy. Every decision is dated in the doc. Prototypes A-E5 keep the exploration path: five divergent layouts converging into E5, a fully interactive simulation whose GOOD/BAD snapshots become the build's test fixtures. todo.org gains the spec-create task, the panel keybinding-family task, and scope notes on the net/bt doctor retrofit. --- .../2026-07-07-maint-console-A-two-column.html | 335 ++++ .../2026-07-07-maint-console-B-domain-tiles.html | 382 +++++ .../2026-07-07-maint-console-C-severity-feed.html | 410 +++++ .../2026-07-07-maint-console-D-doctor-first.html | 364 +++++ ...07-07-maint-console-E-instrument-dashboard.html | 475 ++++++ ...6-07-07-maint-console-E2-grouped-dashboard.html | 443 +++++ ...2026-07-07-maint-console-E3-hifi-drilldown.html | 488 ++++++ .../2026-07-07-maint-console-E4-interactive.html | 546 +++++++ ...6-07-07-maint-console-E5-selector-subpanel.html | 1696 ++++++++++++++++++++ 9 files changed, 5139 insertions(+) create mode 100644 docs/prototypes/2026-07-07-maint-console-A-two-column.html create mode 100644 docs/prototypes/2026-07-07-maint-console-B-domain-tiles.html create mode 100644 docs/prototypes/2026-07-07-maint-console-C-severity-feed.html create mode 100644 docs/prototypes/2026-07-07-maint-console-D-doctor-first.html create mode 100644 docs/prototypes/2026-07-07-maint-console-E-instrument-dashboard.html create mode 100644 docs/prototypes/2026-07-07-maint-console-E2-grouped-dashboard.html create mode 100644 docs/prototypes/2026-07-07-maint-console-E3-hifi-drilldown.html create mode 100644 docs/prototypes/2026-07-07-maint-console-E4-interactive.html create mode 100644 docs/prototypes/2026-07-07-maint-console-E5-selector-subpanel.html (limited to 'docs/prototypes') diff --git a/docs/prototypes/2026-07-07-maint-console-A-two-column.html b/docs/prototypes/2026-07-07-maint-console-A-two-column.html new file mode 100644 index 0000000..23d0e60 --- /dev/null +++ b/docs/prototypes/2026-07-07-maint-console-A-two-column.html @@ -0,0 +1,335 @@ + + + + + +Maintenance console — Approach A · two-column board + + + + + +
+
+
MAINT · ratio
+
+ ● + healthy +
+
+
Approach A — Two-column board
+ +
+ + +
+
FIXactionable — auto & confirm
+ +
+ + +
+ + +
+
Storage
+
●cache sizeAuto
+
●btrfs scrub ageConf
+
●fstrim.timerConf
+
●deep trim (keep-1)Conf
+
+ + +
+
Snapshots
+
●snapper count/retentionConf
+
●snapshot auto-timerConfactive
+
+ + +
+
Packages
+
●orphaned packagesConf
+
●.pacnew filesConf
+
●keyring freshnessConf
+
+ + +
+
Logs & coredumps
+
●coredumpsAuto
+
●journald disk usageAuto
+
●app-log cleanupAuto
+
+ + +
+
systemd
+
●maintenance timers firingConf
+
+ + +
+
Network
+
●firewall activeConfactive
+
●tailscale peersConf4/4
+
●fail2banConfactive
+
●NTP syncConfsynced
+
+ + +
+
Services
+
●docker reclaimableConf
+
●cronConfok
+
+
+ + +
+
WATCHread-only — none, human & forensic
+ + +
+
Hardware
+
●SMART healthNonePASSED
+
●SMART wearNone0%
+
●btrfs device errorsNone0/0
+
●CPU/GPU tempsNone61°/54°
+
●thermal throttlingNoneno
+
●kernel/hardware eventsNoneclean
+
+ + +
+
System & disk
+
●is-system-runningNonerunning
+
●disk usageNone69%
+
●btrfs unallocatedNone118 GiB
+
●kernel taint flagNone0
+
●pacman file integrityWflowclean
+
●reboot requiredHumanno
+
+ + +
+
Forensic (workflow)
+
●failed unitsWflow1
+
●unclean-shutdown rateWflow75%
+
●journal error countWflow12 real
+
●memory free / OOMWflow104 GB free
+
●rsyncshot backupWflow3h ago
+
●DNS / NetworkManagerWflowok
+
+ + +
+
Memory & power
+
●swap / zramNone16 GiB
+
○battery healthNonen/a
+
+ + +
+
Network posture
+
●unexpected listenersNone3
+
+ + +
+
Services & virt
+
●docker stopped containersNone2
+
○libvirt VMsNoneoff
+
+
+ +
+ + +
+ UPDATES + 47 pending · mesa, linux-lts, systemd · 3 CVEs + +
+
+ + +
+
+
MAINT · ratio · CLEANING
+
+ ◐ + working +
+
+
Doctor has the surface — Auto set streaming
+ +
+
+ ◐ + cache trim + running +
+
+ ● + journal vacuum + done · reclaimed 0.9 GB +
+
+ ● + coredump clear + done · 12 cleared +
+
+ ○ + app-log cleanup + queued +
+
+ +
+ 2 of 4 done · 1 running · 1 queued + +
+
+ + +
+ ● Auto — doctor "Clean up", fires unattended + ● Confirm — doctor "Review & fix", preview then one click + ● Human — panel nudges, you act + ● Workflow — escalate to the agent health-check + ● None — diagnostic only, no fix +
+
+ ● ok/pass + ● warn + ● fail + ◐ running + ○ off / n/a +
+ + + diff --git a/docs/prototypes/2026-07-07-maint-console-B-domain-tiles.html b/docs/prototypes/2026-07-07-maint-console-B-domain-tiles.html new file mode 100644 index 0000000..ebc3acf --- /dev/null +++ b/docs/prototypes/2026-07-07-maint-console-B-domain-tiles.html @@ -0,0 +1,382 @@ + + + + + +MAINT · ratio — Approach B + + + + +
+
+
+
MAINT · ratio
+
+
● healthy
+
+
Approach B — Domain-tile grid
+ + +
+
+ + +
+
+
+ ● + 47 pending · 3 CVE · mesa,linux-lts + +
+
+ + +
+ + +
+
+ ● +
+
Storage
+
fstrim off · scrub 34d
+
+
+ + 4 fix · 6 watch +
+
+
+
10 checks — lever shown where the doctor can act
+ +
●cacheauto8.8 GB
+
●deep-trimconfirmkeep-3
+
●disk69%—
+
●btrfs unalloc118 GiB—
+
●btrfs scrubconfirm34 d
+
●btrfs device-err0 / 0—
+
●SMARTPASSED—
+
●SMART wear0%—
+
●fstrim.timerconfirmoff
+
●pacman integrityworkflowclean—
+
+
+ + +
+
+ ● +
+
Snapshots
+
42 snaps · timer active
+
+
+
+ 1 fix · 1 watch + ▸ detail +
+ +
+ + +
+
+ ● +
+
Packages & Security
+
3 CVE · 47 updates
+
+
+
+ 3 fix · 4 watch + ▸ detail +
+ +
+ + +
+
+ ● +
+
systemd & Boot
+
1 failed · timers 4/5
+
+
+
+ 1 fix · 3 watch + ▸ detail +
+ +
+ + +
+
+ ● +
+
Logs & Coredumps
+
18 coredumps / 7d
+
+
+ +
+
+
+ 3 fix · 2 watch + ▸ detail +
+ +
+ + +
+
+ ● +
+
Memory / Thermal / Power
+
unclean-shutdown 75%
+
+
+
+ 0 fix · 6 watch + ▸ detail +
+ +
+ + +
+
+ ● +
+
Network & Posture
+
all posture green
+
+
+
+ 4 fix · 2 watch + ▸ detail +
+ +
+ + +
+
+ ● +
+
Services & Backups
+
docker reclaim 3 GB
+
+
+
+ 2 fix · 3 watch + ▸ detail +
+ +
+ +
+
+ + +
+
+
MAINT · ratio · CLEANING ◐
+ 4 Auto actions +
+
+
◐cache trimrunning…
+
●journal vacuumdone · reclaimed 0.9 GB
+
●coredump cleardone · 12 cleared
+
○app-log cleanupqueued
+
+
+
+ +
+ Automation — Auto doctor cleans unattended · Confirm doctor previews then you click · Human panel nudges you to act · Workflow escalates to an agent workflow · None diagnostic only. +
+ + + + diff --git a/docs/prototypes/2026-07-07-maint-console-C-severity-feed.html b/docs/prototypes/2026-07-07-maint-console-C-severity-feed.html new file mode 100644 index 0000000..90182d6 --- /dev/null +++ b/docs/prototypes/2026-07-07-maint-console-C-severity-feed.html @@ -0,0 +1,410 @@ + + + + + +MAINT · ratio — Approach C · Severity feed + + + + +
+
+
+
MAINT · ratio
+
Approach C — Severity feed
+
+
● healthy
+
+ + +
+ + +
+
+ 47 pending · 3 CVE · mesa, linux-lts + +
+
+ +
+ + +
Fail — act now · 2
+
+
+ ● + fstrim.timer + + off + Confirm + +
+
+ ● + arch-audit + + 3 CVE + Workflow + → workflow +
+
+ + +
Warn — review · 13
+
+
+ ● + cache + + 8.8 GB + Auto + +
+
+ ● + btrfs scrub + + 34d ago + Confirm + +
+
+ ● + orphans + + 13 + Confirm + +
+
+ ● + .pacnew files + + 2 + Confirm + +
+
+ ● + pending updates + + 47 + Workflow + → workflow +
+
+ ● + failed units + + 1 + Workflow + → workflow +
+
+ ● + maintenance timers + + 4/5 + Confirm + +
+
+ ● + coredumps + + 18 / 7d + Auto + +
+
+ ● + unclean-shutdown + + 75% + Workflow + → workflow +
+
+ ● + docker reclaim + + 3 GB + Confirm + +
+
+ + +
+
+ ▾ + ● + 29 OK — clean, no action + + collapse +
+
+
●deep-trim keep-3availableConfirm
+
●disk usage69%None—
+
●btrfs unalloc118 GiBNone—
+
●btrfs device-err0 / 0None—
+
●SMART statusPASSEDNone—
+
●SMART wear0%None—
+
●pacman integritycleanWorkflow—
+
●snapper count42Confirm
+
●snapshot timeractiveConfirm—
+
●keyring12dConfirm
+
●AUR staleness5Workflow→ workflow
+
●reboot-requirednoHuman—
+
●is-system-runningrunningNone—
+
●kernel taint0None—
+
●journald size1.2 GBAuto
+
●app-log cleanupokAuto
+
●journal errors12 realWorkflow—
+
●kernel/hw eventscleanNone—
+
●memory104 GB freeWorkflow—
+
●swap/zram16 GiBNone—
+
●temps61° / 54°None—
+
●thermal throttlenoNone—
+
●DNS / NMokWorkflow—
+
●firewallactiveConfirm—
+
●listeners3None—
+
●tailscale4/4Confirm—
+
●fail2banactiveConfirm—
+
●NTPsyncedConfirm—
+
●rsyncshot3h agoWorkflow—
+
●docker stopped2None—
+
●cronokConfirm—
+
+
+ + +
+
+
○batteryn/aNone—
+
○libvirtoffNone—
+
+
+ +
+
+ + +
+
+ MAINT · ratio · CLEANING ◐ + 4 Auto actions +
+
+
+ ◐ + cache trim running… +
+
+ ● + journal vacuum — done + reclaimed 0.9 GB +
+
+ ● + coredump clear — done + 12 cleared +
+
+ ○ + app-log cleanup queued +
+
+
+ +
+
+ +
+ Automation classes: + Auto — doctor "Clean up" runs unattended · + Confirm — doctor "Review & fix" preview then click · + Human — you act · + Workflow — escalate to an agent workflow · + None — diagnostic only. +   Lamps: ● ok / warn / fail · ◐ in-progress · ○ off-or-n/a. ZFS rows omitted (btrfs host). +
+ + + diff --git a/docs/prototypes/2026-07-07-maint-console-D-doctor-first.html b/docs/prototypes/2026-07-07-maint-console-D-doctor-first.html new file mode 100644 index 0000000..7f2a86e --- /dev/null +++ b/docs/prototypes/2026-07-07-maint-console-D-doctor-first.html @@ -0,0 +1,364 @@ + +MAINT · ratio — Approach D Doctor-first + + +
+ + +
+
MAINT · ratio + Approach D — Doctor-first +
+
● healthy
+
+
+ +
— resting state · the doctor is what you see first —
+ + +
+
✚ ✚ ✚
+
+ ratio · 1 fail · 8 warnings · 34 ok +
+ + +
+ + ●●● + ●●● + ●●● + ●●● + ●●● + ●●● + ●●● + ●●● + ●●● + ●●● + ●○● + ●●● + ●●● + ●●○ + ●● +
+ + +
+ + +
+ +
+ "Should I run the doctor?" — press Clean up for the 4 unattended tasks, or Review & fix to preview the 9 confirm items. +
+
+ + +
+
+ ▾ 44 metrics · the fine print +
+
+ + +
+
Storage & filesystem
+ + + + + + + + + + +
●cache8.8 GBauto
●deep-trim keep-3readyconfirm
●disk69%—
●btrfs unalloc118 GiB—
●btrfs scrub34 d agoconfirm
●btrfs device-err0 / 0—
●SMARTPASSED—
●SMART wear0%—
●fstrim.timeroffconfirm
+
+ + +
+
Packages & snapshots
+ + + + + + + + + + +
●pacman integritycleanworkflow—
●snapper count42confirm
●snapshot timeractiveconfirm—
●orphans13confirm
●.pacnew2confirm
●keyring12 dconfirm
●pending updates47workflowupdates
●arch-audit3 CVEworkflowupdates
●AUR staleness5workflowupdates
+
+ + +
+
System state
+ + + + + + +
●reboot-requirednohuman—
●is-system-runningrunning—
●failed units1workflow
●maintenance timers4 / 5confirm
●kernel taint0—
+
+ + +
+
Logs & coredumps
+ + + + + + +
●coredumps18 / 7 dauto
●journald1.2 GBauto
●app-log cleanupokauto
●journal errors12 realworkflow—
●kernel/hw eventsclean—
+
+ + +
+
Resources & thermals
+ + + + + + + +
●memory104 GB freeworkflow—
●swap / zram16 GiB—
●temps61° / 54°—
●thermal throttleno—
○batteryn/a—
●unclean-shutdown75%workflow—
+
+ + +
+
Network & security
+ + + + + + + +
●DNS / NMokworkflow—
●firewallactiveconfirm—
●listeners3—
●tailscale4 / 4confirm—
●fail2banactiveconfirm—
●NTPsyncedconfirm—
+
+ + +
+
Backup & services
+ + + + + + +
●rsyncshot3 h agoworkflow—
●docker reclaim3 GBconfirm
●docker stopped2—
○libvirtoff—
●cronokconfirm—
+
+ +
+
+ + +
+
Updates — quarantined from the doctor
+
+
+ 47 pending · 3 CVE · mesa, linux-lts, systemd +
+ +
+
Updates run through the workflow, never the doctor. Clean up will not touch these.
+
+ +
— running state · press CLEAN UP and the hero becomes the output wall —
+ + +
+
+ ✚ DOCTOR · CLEAN UP + 4 tasks · unattended · started 14:22:07 +
+
Running the 4 safe (auto) tasks. Confirm items and updates are untouched.
+
+ +
+
+ ◐ + cache trim + running… 8.8 GB → clearing +
+
+ ● + journald vacuum + done · reclaimed 0.9 GB +
+
+ ● + coredump clear + done · 12 cleared +
+
+ ○ + app-log cleanup + queued +
+
+ +
+ 3 of 4 complete · reclaimed 0.9 GB so far · ~1.1 GB pending + +
+
+ +
diff --git a/docs/prototypes/2026-07-07-maint-console-E-instrument-dashboard.html b/docs/prototypes/2026-07-07-maint-console-E-instrument-dashboard.html new file mode 100644 index 0000000..52ecfad --- /dev/null +++ b/docs/prototypes/2026-07-07-maint-console-E-instrument-dashboard.html @@ -0,0 +1,475 @@ + +MAINT · ratio — Approach E · Instrument Dashboard + + +
+ + +
+
+
MAINT · ratio
+
Approach E — Instrument dashboard
+
+
● healthy
+
+ + +
+ Doctor + + + + + Auto unattended · Confirm preview+click · Human you act · Workflow escalate · None diagnostic + +
+ + +
+ + +
Storage & Filesystem
+ +
+
Disk usage
+
69%
+
root · btrfs
+
+
+
+
+
+
+
08090100%
+
+
+ +
+ +
Package cache ●
+
8.8 GB
+
Auto · threshold ~10 GB
+
+
+
+
+
+
0threshold 10 GB
+
+
+ +
+ +
Deep-trim keep-3 ●
+
keep 3
+
Confirm · retain 3 versions
+
+
+
staged & ready
+
+
+ +
+ +
btrfs scrub
+
+ + + + + 34 + days + +
+
● past 30d cadence
+
+ +
+
btrfs unalloc ●
+
118 GiB
+
unallocated headroom
+
+
+
ample slack
+
+
+ +
+
btrfs device-err ●
+
0 / 0
+
read / write errors
+
+ +
+ +
fstrim.timer ●
+
OFF
+
Confirm · SSD discard disabled
+
+ + +
Thermal & Power
+ +
+
Temperatures ●
+
+
+ + + + + + + + + +
61°C
CPU package
+
+
+ + + + + + + + + +
54°C
NVMe
+
+
+
+ +
+
Thermal throttle ●
+
NO
+
no throttle events
+
+ +
+
Battery ○
+
n/a
+
desktop
+
+ +
+
Swap / zram ●
+
16 GiB
+
zram active
+
+ + +
+
Memory free ●
+
104 GB free
+
of 125 GB total · Workflow
+
+
+
0free 104 / 125 GB
+
+
+ +
+
Unclean-shutdown ratio ●
+
75%
+
clean boots · Workflow — investigate
+
+
+
+
+
+
0clean-boot fraction100%
+
+
+ + +
Updates — quarantined
+
+
Updates
+
47pending
+
3CVE · arch-audit
+
5AUR stale
+
+
+ ● Security advisories present. + Applying updates escalates to the update workflow — not an unattended clean. +
+ +
+ + +
Packages · Logs · Counters
+ +
+ +
Orphans
+
●13
+
Confirm · unowned pkgs
+
+ +
+ +
.pacnew files
+
●2
+
Confirm · config merges
+
+ +
+ +
Coredumps 7d
+
●18
+
Auto · clearable
+
+ +
+ +
Failed units
+
●1
+
Workflow · systemd
+
+ +
+ +
Snapper snapshots ●
+
42
+
Confirm · retention ok
+
+
+
within budget
+
+
+ +
+ +
Keyring age
+
+ + + + 12 + days + +
+
● Confirm · fresh
+
+ +
+
Journal errors
+
●12
+
Workflow · real errors
+
+ +
+
Listeners
+
●3
+
None · open sockets
+
+ + +
Status board — binary & diagnostic
+
+
●SMARTPASSED
+
●SMART wear0%
+
●pacman integrityclean
+
●snapshot timeractive
+ +
●is-system-runningrunning
+
●reboot-requiredno
+
●kernel taint0
+
●kernel/hw eventsclean
+ +
●firewallactive
+
●fail2banactive
+
●tailscale4/4
+
●NTPsynced
+ +
●DNS / NMok
+
●rsyncshot3h ago
+
●cronok
+
●app-log cleanupok
+ +
●journald size1.2 GB
+
●docker stopped2
+
●maintenance timers4/5
+
●docker reclaim3 GB · Prune
+ +
○libvirtoff
+
+ +
+ + +
+
+ MAINT · ratio · CLEANING ◐ + 4 Auto actions +
+
+
◐cache trimrunning…
+
●journal vacuumdone · reclaimed 0.9 GB
+
●coredump cleardone · 12 cleared
+
○app-log cleanupqueued
+
+
+
+ +
diff --git a/docs/prototypes/2026-07-07-maint-console-E2-grouped-dashboard.html b/docs/prototypes/2026-07-07-maint-console-E2-grouped-dashboard.html new file mode 100644 index 0000000..b818006 --- /dev/null +++ b/docs/prototypes/2026-07-07-maint-console-E2-grouped-dashboard.html @@ -0,0 +1,443 @@ + +MAINT · ratio — Approach E2 · Grouped Instrument Dashboard + + +
+ + +
+
+
MAINT · ratio
+
Approach E2 — Grouped instrument dashboard (categories collapse when clean)
+
+
● healthy
+
+ + +
+ Doctor + + + + + Auto unattended · Confirm preview+click · Human you act · Workflow escalate · None diagnostic + +
+ + +
+
Updates
+
47pending
+
3CVE · arch-audit
+
5AUR stale
+
+
+ ● Security advisories present. Updates run through the workflow — never an unattended clean. +
+ +
+ + + + +
+
+ ▾● + Storage & Filesystem + 3 need attention · 6 ok +
+
+
+
+
Disk usage
+
69%
+
root · btrfs
+
+
+
+
+
08090100
+
+
+
+ +
Cache ●
+
8.8 GB
+
Auto · thr ~10 GB
+
+
+
010 GB
+
+
+
+ +
btrfs scrub
+
+ + + + 34 + days + +
+
● past 30d
+
+
+ +
fstrim.timer ●
+
OFF
+
Confirm · SSD discard off
+
+
+
+
●SMARTPASSED
+
●SMART wear0%
+
●device errors0 / 0
+
●btrfs unalloc118 GiB
+
●pacman integrityclean
+
●deep-trimkeep 3
+
+
+
+ + +
+
+ ▾● + Packages & Security + 2 need attention · 2 ok · updates above +
+
+
+
+ +
Orphans
+
●13
+
Confirm · unowned pkgs
+
+
+ +
.pacnew files
+
●2
+
Confirm · config merges
+
+
+
+
●keyring age12d
+
●reboot requiredno
+
+
+
+ + +
+
+ ▾● + systemd & Boot + 2 need attention · 2 ok +
+
+
+
+ +
Failed units
+
●1
+
Workflow · systemd
+
+
+ +
Maintenance timers
+
●4/5
+
Confirm · one not firing
+
+
+
+
●is-system-runningrunning
+
●kernel taint0
+
+
+
+ + +
+
+ ▾● + Logs & Coredumps + 1 needs attention · 4 ok +
+
+
+
+ +
Coredumps 7d
+
●18
+
Auto · clearable
+
+
+
+
●journald size1.2 GB
+
●app-log cleanupok
+
●journal errors12 real
+
●kernel/hw eventsclean
+
+
+
+ + +
+
+ ▾● + Memory · Thermal · Power + 1 needs attention · 4 ok +
+
+
+
+
Temperatures ●
+
+
+ + + + + + + + +
61°C
CPU
+
+
+ + + + + + + + +
54°C
NVMe
+
+
+
+
+
Memory free ●
+
104 GB
+
of 125 GB · Workflow
+
+
0free 104 / 125
+
+
+
Unclean boot ●
+
75%
+
Workflow · investigate
+
+
+
0100
+
+
+
+
●swap / zram16 GiB
+
●thermal throttleno
+
○batteryn/a · desktop
+
+
+
+ + +
+
+ ▾● + Services & Backups + 1 needs attention · 3 ok +
+
+
+
+ +
Docker reclaim
+
●3 GB
+
Confirm · reclaimable
+
+
+
+
●rsyncshot backup3h ago
+
●docker stopped2
+
●cronok
+
○libvirt VMsoff
+
+
+
+ + + +
+
+ ▸● + Snapshots + all clear · 2 ✓ +
+
+ +
+
+ ▸● + Network & Posture + all clear · 6 ✓ +
+
+ +
+ Collapsed panels are all-green — one header line each (chevron ▸ to expand). On a fully-healthy day every + panel collapses to this and the console is ~4 header rows plus the updates strip. The board's height tracks + how much needs you, not how many metrics exist. Verdict lamp (top-right) reads the worst diagnostic + state only — the fstrim fail and the 3 CVEs are actionable/updates, so the box still reads "healthy." +
+ + +
+
+ MAINT · ratio · CLEANING ◐ + 4 Auto actions +
+
+
◐cache trimrunning…
+
●journal vacuumdone · reclaimed 0.9 GB
+
●coredump cleardone · 12 cleared
+
○app-log cleanupqueued
+
+
+
+ +
diff --git a/docs/prototypes/2026-07-07-maint-console-E3-hifi-drilldown.html b/docs/prototypes/2026-07-07-maint-console-E3-hifi-drilldown.html new file mode 100644 index 0000000..fbfb03d --- /dev/null +++ b/docs/prototypes/2026-07-07-maint-console-E3-hifi-drilldown.html @@ -0,0 +1,488 @@ + + + + + +MAINT · ratio — Approach E3 · Hi-fi drill-down console + + + +
+ + +
+
+
archsetup · maintenance console
+
MAINT · ratio
+
+
HEALTHY
+
+
+ Doctor + + +
+
+ + +
+ Updates +
47pending
+
3CVE
+
5AUR stale
+ Security advisories present — updates run through the workflow, never the doctor. + + +
+ + +
Overview· 8 systems
+
+ + +
+ + + click any category header — or its overview tile — to drill in +
+ + +
+
+ ▸ + Storage & Filesystem + 3 attention · 6 ok +
+
+
+
+
Disk usage
+
69%
+
+
root · btrfs · thr 80 / 90
+
+
+ +
Package cache
+
8.8 GB
+
+
Auto · threshold ~10 GB
+
+
+ +
btrfs scrub
+
+ 34days +
+
past 30d cadence
+
+
+ +
fstrim.timer
+
OFF
+
Confirm · SSD discard disabled
+
+
+
+
SMARTPASSED
+
SMART wear0%
+
device errors0 / 0
+
btrfs unalloc118 GiB
+
pacman integrityclean
+
deep-trim keep 3
+
+
+
+ + +
+
+ ▸ + Packages & Security + 2 attention · 2 ok · updates above +
+
+
+
+ +
Orphans
+
13
+
Confirm · unowned pkgs
+
+
+ +
.pacnew files
+
2
+
Confirm · config merges
+
+
+
+
keyring age 12d
+
reboot requiredno
+
+
+
+ + +
+
+ ▸ + systemd & Boot + 2 attention · 2 ok +
+
+
+
+ +
Failed units
+
1
+
Workflow · systemd
+
+
+ +
Maint timers
+
4/5
+
Confirm · one not firing
+
+
+
+
is-system-runningrunning
+
kernel taint0
+
+
+
+ + +
+
+ ▸ + Logs & Coredumps + 1 attention · 4 ok +
+
+
+
+ +
Coredumps 7d
+
18
+
Auto · clearable
+
+
+
+
journald size 1.2 GB
+
app-log cleanup ok
+
journal errors12 real
+
kernel/hw eventsclean
+
+
+
+ + +
+
+ ▸ + Memory · Thermal · Power + 1 attention · 4 ok +
+
+
+
+
Temperatures
+
+
61°CPU
+
54°NVMe
+
+
+
+
Memory free
+
104 GB
+
+
of 125 GB · 0 OOM
+
+
+
Unclean boot
+
75%
+
+
Workflow · investigate
+
+
+
+
swap / zram16 GiB
+
thermal throttleno
+
batteryn/a · desktop
+
+
+
+ + +
+
+ ▸ + Services & Backups + 1 attention · 3 ok +
+
+
+
+ +
Docker reclaim
+
3 GB
+
Confirm · reclaimable
+
+
+
+
rsyncshot backup3h ago
+
docker stopped2
+
cronok
+
libvirt VMsoff
+
+
+
+ + +
+
+ ▸ + Snapshots + all clear · 2 ✓ +
+
+
+
snapper count 42 · retention ok
+
snapshot timeractive
+
+
+
+ + +
+
+ ▸ + Network & Posture + all clear · 6 ✓ +
+
+
+
firewallactive
+
unexpected listeners3
+
tailscale4/4
+
fail2banactive
+
NTP syncsynced
+
DNS / NMok
+
+
+
+ + + + +
+ Two levels: the Overview grid is every system at a glance — click a tile to jump to and open its section. + Each category below is a drill-down you expand or collapse (click the header, or use Expand/Collapse all). + All-green categories (Snapshots, Network) start collapsed — on a fully-healthy day every section collapses and the + console is the masthead + overview. Verdict lamp reads worst diagnostic state only — the fstrim fail and the + 3 CVEs are actionable/updates, so the box still reads HEALTHY. +
+ +
+ + + + diff --git a/docs/prototypes/2026-07-07-maint-console-E4-interactive.html b/docs/prototypes/2026-07-07-maint-console-E4-interactive.html new file mode 100644 index 0000000..c678a8d --- /dev/null +++ b/docs/prototypes/2026-07-07-maint-console-E4-interactive.html @@ -0,0 +1,546 @@ + + + + + +MAINT · ratio — Approach E4 · Interactive view-swap console + + + +
+ + +
+
+
archsetup · maintenance console · E4 interactive
+
MAINT · ratio
+
+
HEALTHY
+
+
+ + Doctor + + +
+
+ + +
+ +
+ Fully interactive: views replace each other (no inline collapse). HOME = overview of all systems; + click a station to drill in; PREV/NEXT walk the categories; CLEAN UP streams the output well and + actually updates the metrics; REVIEW & FIX lists every Confirm item — destructive ones arm on + first press, fire on second. Workflow items always escalate (toast) — the console never runs them. +
+
+
+ + + + diff --git a/docs/prototypes/2026-07-07-maint-console-E5-selector-subpanel.html b/docs/prototypes/2026-07-07-maint-console-E5-selector-subpanel.html new file mode 100644 index 0000000..95ec4c9 --- /dev/null +++ b/docs/prototypes/2026-07-07-maint-console-E5-selector-subpanel.html @@ -0,0 +1,1696 @@ + + + + + +MAINT · ratio — E5 · Selector + dense subpanel + + + +
+ + +
+ prototype controls — + + + + +
+ +
+ + +
+ + MAINTENANCE + 3 CVE + 9 ATTN + + + MNT·01 + +
+ + +
+ Doctor + + + + last scan 07:41 · 44 checks · 0.9s +
+ + +
+ Updates + 47 pending + 3 CVE + 5 AUR stale + 0 firmware + mesa · linux-lts · systemd — guard-checked before any apply + + + + +
+ + +
+ + +
+ + + + +
+
+ +
MNT·01 closed — click to reopen (waybar: click the maint tag)
+ +
+ Prototype notes (not part of the panel): selector stays put — the gold/slate tile is the selection; the + subpanel below swaps. ✕ or Esc closes (chip reopens; reopening replays the hydrate sequence). CLEAN UP + streams into the subpanel and mutates real state; REVIEW & FIX lists every Confirm lever, destructive + ones arm-then-fire. Metrics without a determinate lever are read-only telemetry (agent/workflow assistance + is vLater). SIM FAIL makes every action fail (red lamp, error line, RETRY key, FAILED badge); SIM BAD DAY + loads a degraded-machine snapshot so every metric shows its warn/fail face, doctor or not; SIM RELOAD + replays the open: tiles pulse amber per tier (local reads land first, process probes next, package scans + last) and settle individually — cached values would fill the board instantly in the real build, with lamps + confirming freshness as probes return. +
+
+ + + + -- cgit v1.2.3