We found there was an accident high server load on SG02. We identify some background process causing this issue and already identify the account which causing it. Now the server is back up. In the meantime, we keep monitoring it in the next 24 hours