harmony: Fix task reclaim on restart

[magik6k]

Related Issues

Proposed Changes

On restart harmony would fail to reclaim tasks claimed because the owner_id was already non-null. With this change we just take ownership of all tasks which are recovered

2023-12-07T11:26:05.521+0100	DEBUG	harmonydb	harmonydb/harmonydb.go:133	SQL run	{"query": "\n\t\t\tWITH upsert AS (\n\t\t\t\tUPDATE harmony_machines\n\t\t\t\tSET cpu = $2, ram = $3, gpu = $4, last_contact = CURRENT_TIMESTAMP\n\t\t\t\tWHERE host_and_port = $1\n\t\t\t\tRETURNING id\n\t\t\t),\n\t\t\tinserted AS (\n\t\t\t\tINSERT INTO harmony_machines (host_and_port, cpu, ram, gpu, last_contact)\n\t\t\t\tSELECT $1, $2, $3, $4, CURRENT_TIMESTAMP\n\t\t\t\tWHERE NOT EXISTS (SELECT id FROM upsert)\n\t\t\t\tRETURNING id\n\t\t\t)\n\t\t\tSELECT id FROM upsert\n\t\t\tUNION ALL\n\t\t\tSELECT id FROM inserted;\n\t\t", "err": null, "rowCt": 1, "milliseconds": 51}
2023-12-07T11:26:05.532+0100	DEBUG	harmonydb	harmonydb/harmonydb.go:133	SQL run	{"query": "DELETE FROM harmony_machines WHERE last_contact < CURRENT_TIMESTAMP - INTERVAL '1 MILLISECOND' * $1 ", "err": null, "rowCt": 0, "milliseconds": 11}
2023-12-07T11:26:05.532+0100	INFO	harmonytask	resources/resources.go:77	Cleaned up machines	{"count": 0}
2023-12-07T11:26:05.544+0100	DEBUG	harmonydb	harmonydb/harmonydb.go:133	SQL run	{"query": "SELECT id, name from harmony_task WHERE owner_id=$1", "err": null, "rowCt": 8, "milliseconds": 11}
2023-12-07T11:26:05.545+0100	DEBUG	harmonydb	harmonydb/harmonydb.go:133	SQL run	{"query": "UPDATE harmony_task SET owner_id=$1 WHERE id=$2 AND owner_id IS NULL", "err": null, "rowCt": 0, "milliseconds": 1}
2023-12-07T11:26:05.545+0100	INFO	harmonytask	harmonytask/task_type_handler.go:97	did not accept task	{"task_id": "20241", "reason": "already Taken", "name": "WdPostRecover"}
2023-12-07T11:26:05.557+0100	DEBUG	harmonydb	harmonydb/harmonydb.go:133	SQL run	{"query": "SELECT epoch FROM mining_tasks WHERE task_id = $1", "err": null, "rowCt": 1, "milliseconds": 11}
2023-12-07T11:26:05.558+0100	DEBUG	harmonydb	harmonydb/harmonydb.go:133	SQL run	{"query": "UPDATE harmony_task SET owner_id=$1 WHERE id=$2 AND owner_id IS NULL", "err": null, "rowCt": 0, "milliseconds": 0}
2023-12-07T11:26:05.558+0100	INFO	harmonytask	harmonytask/task_type_handler.go:97	did not accept task	{"task_id": "24308", "reason": "already Taken", "name": "WinPost"}

Additional Info

Checklist

Before you mark the PR ready for review, please make sure that:

• Commits have a clear commit message.
• PR title is in the form of of <PR type>: <area>: <change being made>
  • example: fix: mempool: Introduce a cache for valid signatures
  • PR type: fix, feat, build, chore, ci, docs, perf, refactor, revert, style, test
  • area, e.g. api, chain, state, market, mempool, multisig, networking, paych, proving, sealing, wallet, deps
• If the PR affects users (e.g., new feature, bug fix, system requirements change), update the CHANGELOG.md and add details to the UNRELEASED section.
• New features have usage guidelines and / or documentation updates in
  • Lotus Documentation
  • Discussion Tutorials
• Tests exist for new functionality or change in behavior
• CI is green