bto fleet · botmonitor

loading… Theme

Pipeline

Work waiting at each stage: integration → master (ship loop) → production (deploy)

Throughput & waste

Attempts started vs tasks merged; split tasks charged to their root

Capacity

Agents running vs each pod's ceiling, and time lost waiting on models

Supervisor & review (m1)

Numbers pushed from m1

Range

CPU utilisation by pod

Non-nice CPU, % of the pod's cores (heartbeat; Windows host measured by the collector)

Available memory by pod

GB free (MemAvailable on Linux)

Task queue

Tasks running, waiting on dependencies, and blocked

Time by state

Share of time over the selected range

Running now

Blocked & needing attention

Tasks the fleet will not retry on its own, plus the ship hold

What the queue is waiting on

Unfinished tasks ranked by how many pending tasks depend on them directly

All waiting tasks

Pods

One row per heartbeat session; stale sessions expire on their own

Release

agent-harness main vs tags vs the pin on bto_simulator master

Recent events

Status changes seen between snapshots

Windows ship pod log

Recently finished

Latest attempts to end, any outcome