Microsoft SQL Server Always On availability group performance metrics
Gain deeper visibility into the key metrics of your Microsoft SQL Server Always on Availability Group—including Active Hadr Threads, Replica information, Log Compression Cache Hits, Failover alerts, Synchronization health, Database state, and much more—all from a single, comprehensive console.
The Site24x7 server monitoring agent will send you instant alerts whenever there is a breach in any predefined threshold. You can also automate remediate actions and avoid outages proactively.
Supported metrics
| Log Transactions | ||
| Database Name | The name of a database that belongs to the availability group. | Text |
| Replica Server Name | The server name of the SQL Server hosting this replica. | Text |
| Log Apply Pending Queue | The number of log blocks received but not yet applied on the replica. | Count |
| Log Apply Ready Queue | The number of log blocks that are ready to be applied on the replica. | Count |
| Log Bytes Compressed/sec | The number of log bytes compressed per second before being sent to the secondary replica. | Bytes per Second |
| Log Bytes Decompressed/sec | The number of log bytes decompressed per second on the secondary replica. | Bytes per Second |
| Log Bytes Received/sec | The number of log bytes received per second from the primary replica. | Bytes per Second |
| Log Compression Cache Hits/sec | The number of times per second the compression cache was used for log blocks. | Count per Second |
| Log Compression Cache Misses/sec | The number of times per second the compression cache was missed during log compression. | Count per Second |
| Log Compressions/sec | The number of log compressions happening per second. | Count per Second |
| Log Decompressions/sec | The number of log decompressions happening per second. | Count per Second |
Replicas – Summary (Replication details)
| Replica Server Name | The server name of the SQL Server hosting this replica. | Text |
| Replication Role | Indicates the role of the replica—primary or secondary. | Text |
| Endpoint URL | The URL used by the availability replica for data transmission. | Text |
| Availability Mode | Specifies whether the replica operates in synchronous or asynchronous mode. Possible values: Synchronous Commit, Asynchronous Commit | Text |
| Failover Mode | Indicates whether failover is automatic or manual. | Text |
| Primary Role Allow Connections | States whether the primary replica allows connections. | Text |
| Secondary Role Allow Connections | States whether the secondary replica allows read-only connections. | Text |
| Creation Time | The timestamp of when the availability replica was created. | Text |
| Modified Time | The timestamp of the last configuration modification to the replica. | Text |
| Cluster Health | Overall health status of the WSFC (Windows Server Failover Cluster) hosting the replica. | Text |
| Operational State | The current operating state of the replica (for example, online, offline). | Text |
| Connection State | Indicates the connectivity status between replicas. | Text |
| Recovery Health | Indicates the health of the replica’s recovery process. | Text |
| Synchronization Health | Reflects the combination of the synchronization state of databases on the replica and the configured availability mode. Possible values: Healthy, Partially Healthy, Not Healthy, Unknown | Text |
| Seeding Mode | Mode used for automatic seeding of databases (manual or automatic). | Text |
| Last Connect Error | The last error message encountered when trying to connect to the replica. | Text |
| Last Connect Error Number | The numeric code of the last connection error. | Count |
| Last Connect Error Timestamp | The time when the last connection error occurred. | Text |
| Resent Messages/sec | The number of messages re-sent per second due to dropped connections or errors. | Count per Second |
Replicas – Performance metrics
| Replica | The name of the replica. | Text |
| Resent Messages/sec | The number of messages re-sent per second from this replica. | Count per Second |
| Flow Control Time (ms/sec) | The time messages waited in flow control during the last second. | Milliseconds |
| Flow Control/sec | The number of flow control initiations per second. | Count per Second |
| Bytes Sent | The total number of bytes sent from this replica. | Bytes |
| Bytes sent to Replica/sec | The rate at which bytes are sent from this replica to another per second. | Bytes per Second |
| Bytes Sent to Transport/sec | The rate at which bytes are sent from the replica to the transport layer per second. | Bytes per Second |
| Bytes Received | The total number of bytes received by this replica. | Bytes |
| Receives from Replica/sec | The rate at which this replica receives bytes from another replica per second. | Bytes per Second |
| Sends to Replica/sec | The rate at which this replica sends bytes to another replica per second. | Bytes per Second |
Databases – Performance metrics
| Database Name | The name of a database that belongs to the availability group. | Text |
| Database Flow Control Delay | Time delay introduced by flow-control throttling when synchronizing database transactions | Microseconds |
| Database Flow Control per Second | Number of flow-control events triggered at the database level each second. | Count per Second |
| File Bytes Received /sec | Rate at which database file bytes are received from the primary replica. | Bytes per Second |
| Group Commit Time | Time taken for a group-commit operation on the secondary replica. | Microseconds |
| Group Commit Time /sec | Number of group-commit operations performed per second. | Count per Second |
| Mirrored Write Transactions /sec | Number of write transactions mirrored to the secondary per second. | Count per Second |
| Recovery Queue | Amount of log data (in kilobytes) queued for recovery processing. | Kilobytes |
| Redo Blocked /sec | Number of times per second that redo operations were blocked. | Count per Second |
| Redo Bytes Remaining | Amount of log data still pending to be redone on the secondary. | Kilobytes |
| Redone Bytes /sec | Rate at which bytes are redone on the secondary replica. | Bytes per Second |
| Redones /sec | Number of redo operations occurring per second. | Count per Second |
| Transaction Delay | Time difference between commit on primary and hardened copy on secondary. | Milliseconds |
| Log Apply Queue | Number of log blocks waiting in queue to be applied. | Count |
| Log Compression Cache | Number of log blocks currently stored in the compression cache. | Count |
| Log Compression Cache Misses /sec | Number of times per second that the log compression cache did not contain the block needed. | Count per Second |
| Log Compressions | Total number of log compression operations performed. | Count |
| Log Bytes Compression | Total bytes processed through log compression. | Bytes |
Databases – Status metrics
| Database Name | The name of a database that belongs to the availability group. | Text |
| Replica Server Name | The server name hosting this database replica. | Text |
| Synchronization Health | Reflects the combined state of the database’s synchronization and the configured availability mode of its replica. Possible values: Healthy, Partially Healthy, Not Healthy, Unknown | Text |
| Synchronization State | Describes the current data-movement state of the database. Possible values: Synchronized, Synchronizing, Suspended, Not Synchronizing, Initializing | |
| Database State | Indicates the operational status of the database on its replica (for example, Online, Restoring, Recovery Pending). | Text |
| Suspended | Whether data movement for this database is currently suspended. | Text |
| Reason | If suspended, provides the reason why the database’s data movement was halted. | Text |
| Failover Cluster | Indicates if the secondary database remains in sync with the primary as part of the failover cluster. | Text |
| Database Joined State | Shows whether the database has been successfully joined to the availability group. | Text |
Related links
Database monitoring: Microsoft SQL | Microsoft SQL Insight | Microsoft SQL individual database
Server monitoring: Windows
