)]}'
{
  "commit": "a24007566821978a0d3c70c4029df6509a1f50ea",
  "tree": "2a57fdaf7ab76d22aa82acfd748833dfa49dba67",
  "parents": [
    "999e1518982247c881f61daefcd05bb5d3107053"
  ],
  "author": {
    "name": "Nic",
    "email": "qianyong@api7.ai",
    "time": "Fri Apr 24 18:25:23 2026 +0800"
  },
  "committer": {
    "name": "Nic",
    "email": "qianyong@api7.ai",
    "time": "Fri Apr 24 18:25:23 2026 +0800"
  },
  "message": "test: tolerate shutdown flush errors in prometheus2.t TEST 25\n\nDuring worker shutdown, the batch processor now attempts to flush\npending entries. For zipkin, this means the reporter tries to send\nspans to an already-down server, generating an ERROR log from\nreporter.lua:133. Override the default no_error_log assertion to\nallow [error] during HUP-triggered shutdown while still catching\ncritical alerts.\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "cf303ce77945f1c7bec79c6bb9b63276bbe01c45",
      "old_mode": 33188,
      "old_path": "t/plugin/prometheus2.t",
      "new_id": "2730b1252ee48d595f5ea61e3c832aff1e082a45",
      "new_mode": 33188,
      "new_path": "t/plugin/prometheus2.t"
    }
  ]
}