Our diagnostic analysis has identified faulty RAM modules as the root cause of the recent server crashes. Now that the issue has been isolated, we have scheduled a maintenance intervention to repair and replace these 2 defective RAM sticks. To minimize disruption and ensure service continuity, we will urgently deploy an additional node this evening to migrate certain servers as a precaution.