
may I know why hardenlsn is higher than last commitedLSN ?

may I know why hardenlsn is higher than last commitedLSN ?
Please read the docs you often do not, reading show me on very starting that
Depending on the action and higher-level states, database-state information may be unavailable or out of date. Furthermore, the values have only local relevance. For example, on the primary replica, the value of the last_hardened_lsn column reflects the information about a given secondary database that is currently available to the primary replica, not the actual hardened LSN value that the secondary replica might have currently.
So you can see its not reflecting actual hardened values and it would only be more confusing to derive any sense out of it
Hi @sakuraime,
As stated in the article posted by Shashank, the values of the end_of_log_lsn, last_hardened_lsn, last_received_lsn, last_sent_lsn, recovery_lsn, and truncation_lsn columns are not actual log sequence numbers (LSNs). Of the LSN values returned by sys.dm_hadr_database_replica_states, only last_redone_lsn is a real LSN.
Best Regards,
Amelia
"only last_redone_lsn is a real LSN."
what do you mean by "real"?
I am not concerning whether it's real or not. but why hardenlsn is higher than last commitedLSN
harden > for my understanding is the log block that hardened on 2nd replica.
committted > for my understanding is the log block flushed to Tlog on Primary replica .
3 people are following this question.