kshitijk4poor 9088707160 fix(monitoring): report cron store health across every served profile ...

The store gauges called degraded_record() in the exporter's (launch profile's) scope, so a
secondary profile's full or read-only store in a multiplexed gateway exported writable=1 and
skipped_runs=0 (Greptile r1 #4). Every served profile's ticker runs in this process, so the
gauges are now host-wide: writable is 0 while any store is degraded and skipped_runs sums them.

2026-10-07 10:20:18 +05:30

..

__init__.py

…

cron_health.py

fix(monitoring): report cron store health across every served profile

2026-10-07 10:20:18 +05:30

emitter.py

…

events.py

…

gateway_health_export.py

fix(monitoring): register the cron store gauges so OTLP exports them

2026-10-07 10:20:18 +05:30

gateway_health.py

…

otlp_exporter.py

…

policy.py

…

redaction.py

…