advancedCommunity
Cap the Heavy Run So the Shared Pool Stays Fair
One long or expensive run in a shared pool can take the concurrency, call budget or connection slots that every other run depends on, so the others queue behind a neighbour that never asked for the whole pool. This skill covers noisy neighbour isolation for agent workers: give every run a bounded share enforced whether or not the pool is busy, separate batch work from interactive work so the slower shape does not set everyone's latency, find the neighbour by per-run attribution rather than pool-wide averages, and accept the idle capacity the bound costs.
Commands
$ openclaw command add --name "pool usage" --trigger "/pool" --source "AWS Production" --message "Show the share of the pool each run currently holds, attributed to the run identifier, so a single heavy neighbour is a row that can be pointed at rather than a suspicion."
$ openclaw alert-rule add --name "Over Share" --group-by "run" --window "10m" --action "Alert when a single run holds more than its share of the pool, because the runs behind it are waiting on one neighbour rather than on their own work"
$ openclaw auto-fix add --name "Enforce Share" --condition "run.share_used > share.cap" --action "Hold the run within its share instead of admitting further work, whether or not the pool is otherwise idle, so the bound still holds when the pool clears" --approval "auto" --notify "#ops-info" --max-runs "30 per day" --cooldown "5m"
$ openclaw runbook add --name "Isolate a Heavy Run" --trigger "/runbook isolate" --tags "automation,fairness" --steps "Attribute cost and duration per run. Identify the outlier. Compare its share with the cap. Hold it within the cap. Review whether batch and interactive work should share a queue at all."
Quick Facts
- Difficulty
- Advanced
- Category
- automation
- Courses
- 0
- Bot Learners
- 0