monitor the world #286

New issue

Closed

opened 2026-06-10 21:48:58 +00:00 by coilysiren · 1 comment

coilysiren commented

2026-06-10 21:48:58 +00:00

Owner

pulse.uptime # seconds since boot, baseline liveness
pulse.reboots # unexpected restart detection
pulse.clock-drift # NTP offset, skew that breaks traces and certs
pulse.services # failed service or unit count
pulse.certs # TLS cert days-until-expiry on self-hosted services
strain.cpu # CPU utilization percent
strain.load # run-queue load average
strain.pressure-cpu # PSI cpu stall time
strain.pressure-memory # PSI memory stall time
strain.pressure-io # PSI io stall time
strain.fds # open file descriptors vs max
headroom.memory # free and available memory
headroom.swap # swap used and swap-in rate
headroom.disk # filesystem bytes free
headroom.inodes # inodes free
headroom.conntrack # conntrack table fill vs max
flow.net-bytes # interface throughput in and out
flow.net-errors # interface error count
flow.net-drops # dropped packet count
flow.net-retransmits # TCP retransmits, wifi degradation signal
flow.disk-throughput # read and write bytes per second
flow.disk-iops # read and write ops per second
flow.disk-latency # await, queue depth, percent util saturation
heat.temp # component temperatures
heat.throttle # thermal throttle event counters
heat.fans # fan speed RPM, doubles as fan-failure detector
aging.disk-wear # NVMe percentage-used and TBW endurance
aging.disk-sectors # SMART reallocated and pending sectors
aging.mce # CPU machine check exceptions
aging.edac # ECC memory error counts where present
aging.resets # dmesg disk and USB reset events
appetite.power # package and wall power draw via RAPL
appetite.gpu-power # GPU power draw in watts
brains.gpu-util # GPU utilization percent
brains.vram # GPU memory used
brains.gpu-clocks # core and memory clock speeds
brains.gpu-throttle # GPU throttle reason flags
brains.gpu-temp # GPU temperature, dual-homes to heat
residents.processes # per-process cpu and memory, the hard one

- pulse.uptime # seconds since boot, baseline liveness - pulse.reboots # unexpected restart detection - pulse.clock-drift # NTP offset, skew that breaks traces and certs - pulse.services # failed service or unit count - pulse.certs # TLS cert days-until-expiry on self-hosted services - strain.cpu # CPU utilization percent - strain.load # run-queue load average - strain.pressure-cpu # PSI cpu stall time - strain.pressure-memory # PSI memory stall time - strain.pressure-io # PSI io stall time - strain.fds # open file descriptors vs max - headroom.memory # free and available memory - headroom.swap # swap used and swap-in rate - headroom.disk # filesystem bytes free - headroom.inodes # inodes free - headroom.conntrack # conntrack table fill vs max - flow.net-bytes # interface throughput in and out - flow.net-errors # interface error count - flow.net-drops # dropped packet count - flow.net-retransmits # TCP retransmits, wifi degradation signal - flow.disk-throughput # read and write bytes per second - flow.disk-iops # read and write ops per second - flow.disk-latency # await, queue depth, percent util saturation - heat.temp # component temperatures - heat.throttle # thermal throttle event counters - heat.fans # fan speed RPM, doubles as fan-failure detector - aging.disk-wear # NVMe percentage-used and TBW endurance - aging.disk-sectors # SMART reallocated and pending sectors - aging.mce # CPU machine check exceptions - aging.edac # ECC memory error counts where present - aging.resets # dmesg disk and USB reset events - appetite.power # package and wall power draw via RAPL - appetite.gpu-power # GPU power draw in watts - brains.gpu-util # GPU utilization percent - brains.vram # GPU memory used - brains.gpu-clocks # core and memory clock speeds - brains.gpu-throttle # GPU throttle reason flags - brains.gpu-temp # GPU temperature, dual-homes to heat - residents.processes # per-process cpu and memory, the hard one

coilysiren added the

label

2026-06-17 07:06:05 +00:00

coilysiren commented

2026-06-17 08:22:33 +00:00

Author

Owner

Backlog burndown 2026-06-17: closing low-priority (P3/P4) to bring the open count to a manageable level. Nothing lost — reopen if this resurfaces. Batch tag: burndown-2026-06.

Backlog burndown 2026-06-17: closing low-priority (P3/P4) to bring the open count to a manageable level. Nothing lost — reopen if this resurfaces. Batch tag: `burndown-2026-06`.

coilysiren

2026-06-17 08:22:33 +00:00