Decrease in performace on Windows after ~2.5h of up time

[tysmith]

After about two and a half hours of up time we see consistent browser launch failures. There is a spike in logs related to Software Protection Platform Service, Service Control Manager, etc. It appears that disk defrag and trim is running as well. I will attach a dump of the logs from just before the trouble begins.

logs.txt

[petemoore]

From matrix...

On the same machine I also see Possible detection of CVE... earlier in the logs:

{
    "insertId": "16gwekgf3qdpfd",
    "jsonPayload": {
      "Qualifiers": 0,
      "EventType": "Information",
      "pool": "pool98",
      "ComputerName": "lvh4b8ocsz2gfqi",
      "host": "task-C3n7OUVqR8iKaxWRf0g-pA-run-0",
      "Data": "",
      "TimeGenerated": "2024-12-10 18:18:56 +0000",
      "SourceName": "Microsoft-Windows-Kernel-General",
      "Sid": "S-1-5-18",
      "EventID": 1,
      "EventCategory": 5,
      "Message": "Possible detection of CVE: 2024-12-10T18:18:56.7015922Z\r\nAdditional Information: 2024-12-10T18:18:56.7015922Z\r\n\r\nThis Event is generated when an attempt to exploit a known vulnerability (2024-12-10T18:18:56.7015922Z) is detected.\r\nThis Event is raised by a User mode process.\r\n\r\n",
      "RecordNumber": 1384,
      "StringInserts": [
        "2024-12-10T18:18:56.7015922Z",
        "2024-12-10T18:18:56.7015922Z",
        "3",
        "\\Device\\HarddiskVolume4\\Windows\\System32\\rundll32.exe",
        "1600"
      ],
      "TimeWritten": "2024-12-10 18:18:56 +0000",
      "Channel": "System"
    },
    "resource": {
      "type": "global",
      "labels": {
        "project_id": "fuzzmanager-cluster"
      }
    },
    "timestamp": "2024-12-10T18:19:56.997605900Z",
    "logName": "projects/fuzzmanager-cluster/logs/winlog.1",
    "receiveTimestamp": "2024-12-10T18:19:58.073081363Z"
  },