Used{{ used }}
Cap{% if report.cap_bytes %}{{ cap }}{% else %}none{% endif %}
Mode{% if report.mode == "counts_only" %}counting only (store full){% else %}normal{% endif %}
Last cleanup{% if report.last_cleanup_at %}{{ report.last_cleanup_at|timesince }} ago{% else %}never{% endif %}{{ last_at }}
Database{{ report.dialect }}{% if report.dialect == "sqlite" %}whole file {{ file }}{% else %}Foxray tables {{ file }}{% endif %}
Cleanup runs by itself at most every {{ interval_minutes }} minutes, after an event arrives (or from manage.py foxray_tick). Over 90% of the cap the oldest event bodies go until it is under 80%; issues and their counts stay. Still over the cap: counting only until there is room.
Last cleanup removed
| {{ k }} | {{ n }} |
| Nothing. |
Rows
| {{ t }} | {{ n }} |
Alerts waiting to send: {{ report.pending_alerts }}