zkSync: The reason for the downtime is that the block sequence database is faulty, and the database monitoring agent has been set

zkSync tweeted that the specific reason for the downtime of zkSync Era was that the database of the block sequence failed, causing the production of blocks to stop, and the database health alarm was not triggered because it could not connect to the database to collect metrics. But the server API remains unaffected, transactions continue to be added to the mempool, and queries are served normally. zkSync stated that it has comprehensive monitoring, logs and alerts in all components, but due to API functions, none of them are triggered, and the situation of "temporarily no blocks" may be normal. According to zkSync, the outage repair took 5 minutes. Thereafter, zkSync sets up the database monitoring agent to connect to the database and continuously collect metrics even when the database is experiencing problems. Also, alerts are raised if the database monitoring agent fails or fails to establish a connection to the database to collect metrics. According to zkSync, zkSync Era is still in Alpha stage. Currently, the team has disabled the automatic resolution of all alerts related to the database "golden indicator" in the platform, ensuring that even if an important alert temporarily resolves itself, the engineer on duty still needs to investigate the issue. In the end, the zkSync team stated that decentralized sequencers are still the long-term solution.