ops: Signal disk-tier attach pending vs ready #45
Reference in New Issue
Block a user
Delete Branch "ops/disk-tier-attach-signal"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Why
Large disk caches can look memory-only/broken for seconds–minutes at startup while the disk tier attaches. That warm-up is intentional but opaque — operators should not need to read source to see attach pending vs ready.
What
disk_tier_readygauge (0pending /1ready, or1when no disk = N/A not waiting) in/metricsvia existing WriteText/lancache-heartbeatheaderX-SteamCache-Disk-Tier: pending|ready|disabledGetMetricsskips blockingdisk.Size()while attach pending so/metricsstays responsiveHow tested
go test ./steamcache/ -count=1(PASS)TestDiskOnlyDelayedAttach,TestDiskTierSignalMemoryOnly,TestDiskTierSignalMixedPendingReadyRisks
disk_cache_sizein metrics may stay at last/zero until ready (by design — avoids hang)Rollback
Revert this PR / merge revert commit.
Closes #33