4 Commits

Include renames

Author	SHA1	Message	Date
bvandeusenandClaude Opus 5.5	8b9b7fc11d	chore(docker): the history move logs which services its old keys belong to (#6718) ... #6718 "A swarm container's metrics, normal and anomalies are keyed by its task name, so they start over whenever the task moves". The move found 391,091 hourly-only task names against 51 services and about 400 keys with raw samples. Log the services with the most keys and a few names, to read before those keys start moving. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-30 20:13:20 -04:00
bvandeusenandClaude Opus 5.5	acc487f9c3	fix(docker): the move of task-named history rests between passes (#6718) ... #6718 "A swarm container's metrics, normal and anomalies are keyed by its task name, so they start over whenever the task moves". Live, the move took VDNT-Docker02's I/O pressure (psi_io_full_avg60) from about 1% to about 23%, and passes of 5000 raw samples outran their 15 s slot. Each moved sample rewrites its row and every index on it. Batches are 1000 samples, and after a pass the task waits four times as long as it took, so it holds the disk for about a fifth of the time and the move slows if the disk does. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-30 20:01:10 -04:00
bvandeusenandClaude Opus 5.5	4689386b69	fix(docker): the move of task-named history finishes many small keys in one pass (#6718) ... #6718 "A swarm container's metrics, normal and anomalies are keyed by its task name, so they start over whenever the task moves". Live, the move found 391,440 keys, about 2,500 of them with raw samples, and took one key per 15 s pass: about 68 days, in sorted order, so recent history could wait months while charts and normals read thin. A pass now runs for up to 2 s, and keys with raw samples go first. Old hourly keys take a few statements each. A key with no anomalies no longer rewrites the service's timeline entries. The start logs how many keys each table holds, as that count is far higher than expected and not explained yet. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-30 19:53:43 -04:00
bvandeusenandClaude Opus 5.5	db39fc2604	feat(docker): a swarm service's metrics stay under one key across its tasks (#6718) ... #6718 "A swarm container's metrics, normal and anomalies are keyed by its task name, so they start over whenever the task moves". The operator: "the metrics should be keys to the service name not the task name... as that's what's consistent between updates." CPU, memory, restart count and health now go under "swarm/<service>", so a normal, its anomalies and an alert rule carry across a redeploy or a move between nodes. Network rates stay "<host>/<service>", the task id gone, as they answer who on a host is sending. Tasks of a multi-task service share the key, so its level reads as an average across them (of 51 live services, only authentik-worker and portainer_agent run more than one task). History under task names moves in a background task, 5000 raw samples every 15 s, not a migration: the raw table holds about 2M such rows and the app would wait on it at boot. Hourly rows combine where two tasks share an hour, and rows still to review merge to one per metric. A service's anomalies no longer reach a host's page, as it has no one host; get_anomaly reads what happened around one from the service's own events. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-30 19:45:16 -04:00