D1 94 second Latency Spike

For Workers & Pages, what is the name of the domain?

What is the issue or error you’re encountering

Huge latency spikes to d1, up to 94 seconds. DB id: 3d4f24ec-fa8d-4920-82b2-179fad18b18c
See dashboard metrics: https://dash.cloudflare.com/a3f6e70849e6eb5ca1ebaf6083bc84ba/workers/services/view/trafficlight/production/metrics

What steps have you taken to resolve the issue?

None, it mitigated by itself. I’m looking for a root cause to build trust that this won’t happen again. No response on discord:

This is not a large db, 65 kB. Now got an error log:

PrismaClientUnknownRequestError: 
Invalid prisma.goal.findMany() invocation:

Error occurred during query execution:
ConnectorError(ConnectorError { user_facing_error: None, kind: QueryError(SqliteError { extended_code: 1, message: Some("D1_ERROR: Network connection lost.") }), transient: false })

Code:

const findActivityStartTime = performance.now();
    const activity = await prisma.activity.findFirst({
      where: {
        id: activityId,
        goalId: goalId,
        goal: {
          userId: user.id,
        },
      },
    });
    const findActivityEndTime = performance.now();
    console.log(
      `Time to find activity: ${findActivityEndTime - findActivityStartTime}ms`
    );

It happened again last night, 8:30pm PDT. This is a serious trust-buster