Chatee ahora con Soporte
Chat con el soporte

Foglight for SQL Server (Cartridge) 5.7.5.30 - User Guide

Managing SQL Server Database Systems Introduction to this Guide Using Foglight for SQL Server Monitoring Business Intelligence Services Administering Foglight for SQL Server Glossary Reference

Tracking the Status of the Mirroring Operation

If one or more databases within the monitored instance take part in a mirroring operation, either as a principal database whose exact copy is mirrored on a different instance, or as a mirror database, the Mirroring panel allows viewing the status of the mirroring operation.
If no database within the monitored instance takes part in a mirroring operation, the Mirroring panel is left blank and displays the note This instance has no database configured for database mirroring.
If the monitored instance supports the Always On feature (supported only for SQL Server version 2012 Enterprise edition), Always on cannot function at the same time with Mirroring operation: enabling Mirroring requires disabling Always on and conversely. Therefore, if the monitored instance contains databases that have Always On enabled, the Mirroring panel displays the note This instance has no database configured for database mirroring.
The Mirroring panel includes the following sections:
•
Mirroring table — see Mirroring Table.
•
Role and data flow of the selected database — see Viewing the Role and Data Flow of the Selected Database.
In addition, this panel allows further investigation of the selected database, by drilling down to other locations.
•
Investigating using the Databases drilldown — carried out by clicking the instance name under the Database column.
•
Viewing the database’s mirroring performance history — carried out by clicking the instance’s mirroring role (MIRROR or PRINCIPAL) under the Mirroring Role column. For details, see Viewing the Selected Database’s Mirroring Performance History.
•
Investigating using the partner’s mirroring page — carried out by clicking the link at the upper right side of the screen or the Mirror database icon under the section Role and Data Flow of the Selected Database. For details, see Viewing the Partner’s Mirroring Page.
Mirroring Table
The Mirroring table displays the columns listed below.
The table below provides data about the columns that appear by default in the table. For a list of additional metrics, which are hidden by default, see Additional metrics in the Mirroring table.
 
Table 56. Mirroring table, default columns:
Column
Description
DBID
A number (ID) uniquely identifying a database within a SQL Server.
Database
The database name.
NOTE: The database name appears as a link which, when clicked, displays the Databases drilldown in Overview mode, with the selected database highlighted in the Databases table and its details displayed in the panes below.
Mirroring Role
The role the database takes in the mirroring process; either principal or mirror.
NOTE: The mirroring role is displayed as a link which, when clicked, displays the mirroring performance history of the selected database. For details, see Viewing the Selected Database’s Mirroring Performance History.
Principal
The name of the instance whose role in the process is principal.
Mirror
The name of the instance whose role in the mirroring process is mirror.
Mirroring Status
Indicates the severity determined based on the database state: Normal, Warning, or Critical. Each of these severity level values can correspond to several mirroring states.
Normal — can correspond to one of the following mirroring states:
•
SYNCHRONIZED (with a Witness)
•
SYNCHRONIZED (without a Witness)
NOTE: To view whether the mirroring operation is configured to use a Witness, click the Customizer button at the end of the table and turn on the display of the Witness Name and Witness State columns. For details, see Additional metrics in the Mirroring table.
•
SYNCHRONIZING
Warning — can correspond to one of the following mirroring states:
•
DISCONNECTED
•
PENDING_FAILOVER
Critical — can correspond to one of the following mirroring states:
•
SUSPENDED
•
UNSYNCHRONIZED
Mirroring State
A state indicating the mirroring session condition. This field can have one of the following values:
•
DISCONNECTED
•
SYNCHRONIZED (with a Witness configured)
•
SYNCHRONIZING
•
PENDING_FAILOVER
•
SUSPENDED
•
UNSYNCHRONIZED
•
SYNCHRONIZED (without a Witness configured)
Safety Level
The safety level at which the mirroring session is configured to work. This column indicates the level of synchronization between the two servers that take part in the mirroring process.
This field can have one of the following values:
•
UNKNOWN — unknown state
•
OFF — high availability (asynchronous). When an instance is configured with this safety level, the main consideration is that it remains available, regardless of the availability of its mirroring partner.
•
FULL — high safety (synchronous). The instance is configured to fail over when its mirroring partner becomes unavailable. Failover can be carried out automatically (if a Witness is configured) or manually.
Redo Queue
The redo queue size. This size can be either left Unlimited or defined in megabytes (MB).
Alarms
The number of warning, critical, and fatal alarms for the SQL Server database instance. When holding the cursor over one of the alarm counts, the dwell displays the most recent alarms raised against this database, sorted by severity.
Clicking this field displays the Alarms list, which is listed by severity order. See the Foglight Online Help, Monitoring System-Wide Alarms for details on the alarm information.
Additional metrics in the Mirroring table
The metrics listed below, which are also part of the Database Mirroring collection, do not appear by default in the Mirroring table. For details on these metrics, see the Database Mirroring Collection section in Foglight for SQL Server Reference Guide.
To display the metrics listed below:
1
Click the Customizer button () at the end of the table.
2
Select the check box near the requested metric.
Additional metrics list
•
Failover LSN
•
End Log LSN
•
Replication LSN
•
Witness State
•
Connection Timeout
•
Partner Name
•
Witness Name
•
Role Sequence
•
Log Scanned For Undo
•
Log Remaining For Undo
•
Mirroring Roundtrip - Latency
•
Commit Acknowledgement Delay
•
Write Commit
•
Log Sent
•
Log Send Flow Buffer Wait
•
Log Sent From Cache
•
Send Queue
•
Roll Forward Queue
•
Log Received
•
Log Rolled Forward
•
Log Cache Redone
•
Log Harden Wait Time
Viewing the Role and Data Flow of the Selected Database
The Role and Data Flow of Database section displays the mirroring operation of the database selected in the Mirroring table and its mirroring partner. The monitored instance is always displayed on the left, and the data is shown as flowing in this instance’s direction, that is: Log Received if the database is Mirror, and Log Sent if the database is Principal.
Figure 23. The Role and Data Flow of Database displays the mirrored databases.
To investigate the data flow by displaying the selected database’s mirroring partner, use one of the following methods:
•
Click the cylinder icon that represents the mirroring partner
•
Click the link Investigate using the Partner’s Mirroring Page on the upper right side of the panel, above the table.
If the partner is currently monitored, its mirroring page will now be displayed. For details, see Viewing the Partner’s Mirroring Page. Otherwise, an error message is displayed, notifying that the partner server is currently not monitored.
Viewing the Selected Database’s Mirroring Performance History
Clicking the Mirroring Role column of the Mirroring table displays the database’s Mirroring Performance History page, which allows carrying out the tasks described in the following sections:
•
Tracking the mirroring role over time
•
Tracking data transfer between the principal and the mirror databases
•
Tracking the mirroring roundtrip during the selected time range
Tracking the mirroring role over time
The Mirroring Role pane displays the role the database played in the mirroring operation during the selected time range: Principal, Mirror, or Not Mirroring.
The mirroring role is displayed as a gauge, with each of the mirroring roles indicated by another color. Positioning the cursor over the gauge displays the percentage, within the specified time range, the database spent in each role.
Tracking data transfer between the principal and the mirror databases
The middle section of the Mirroring Performance History page includes the following panes:
•
Principal Counters — this pane displays a chart of the following values, which can be selected from the list on the upper left:
•
Commit Acknowledgement Delay — indicates a delay in waiting for acknowledgement of unterminated transaction commit. This metric is specific to the principal database, and holds values only when a Full safety level is configured.
•
Write Commit — the number of transactions in the principal database that had to wait for a write commit in the mirror database's transaction log. Low values of this metric indicate a bottleneck.
•
Log Sent — indicates the rate, in megabytes, of log sent from the principal to the mirror database.
•
Log Send Flow Buffer Wait — the amount of time the mirroring session had to wait to use the mirroring flow control buffer. This value is specific to the principal database.
•
Log Sent from Cache — indicates the aggregated size, in megabytes, of the log records sent from the principal database cache rather than straight from the transaction log.
•
Send Queue — total size, in megabytes, of data waiting to be sent to the mirroring database.
•
Mirror Counters — this pane displays a chart of the following values, which can be selected from the list on the upper left:
•
Roll Forward Queue — total size, in megabytes, that remains to roll forward to the mirroring database.
•
Log Received — indicates the aggregated size, in megabytes, of log records received from the principal database.
•
Log Rolled Forward — the aggregated size, in megabytes, of log records that were rolled forward on the mirror database.
•
Log Cache Redone — the aggregated size, in megabytes, of log records that are being read from redo cache rather than from the transaction log. Constantly low values of this metric indicate that the transaction logs are arriving faster than they can be read by the redo task. This metric is specific to the mirror database.
•
Log Harden Wait Time — the amount of time spent waiting for the log to be written to the mirroring database. High values of this metric can indicate that the disk of the mirroring database is loaded.
Tracking the mirroring roundtrip during the selected time range
The Mirroring Roundtrip section displays a chart that indicates the latency of the mirroring session during the selected time range.
Viewing the Partner’s Mirroring Page
When viewing the mirroring performance of a database that takes part in the mirroring operation, either as a principal or as a mirror database, it is possible to investigate the selected database’s mirroring partner, using one of the following methods:
•
Clicking the cylinder icon that represents the mirroring partner
•
Clicking the link Investigate using the Partner’s Mirroring Page on the upper right side of the table.
In so doing, the mirroring operation can be investigated from the point of view of the other database, thereby determining the source of the performance issue.

Reviewing the Always On Availability Groups

If one or more databases within the monitored instance is of version SQL Server 2012 Enterprise Edition, is part of an OS cluster and has the Always On feature enabled, the Always On panel allows viewing the status of the Always On availability groups that reside on the cluster.
* 
NOTE: Always on cannot function at the same time with Mirroring operation: enabling Mirroring requires disabling Always on and conversely. Therefore, if the monitored instance contains databases that support Always On but have Mirroring enabled, the Always On panel displays the note This instance has no database configured for Always On.
The Always On panel includes the following sections:
•
OS Cluster information — see OS Cluster Information.
•
High Availability groups table — displays all High Availability groups that are configured on the instance. For details, see High Availability groups table.
•
Availability group map — displays the dependency (primary/secondary relations) between the databases in the groups selected in the High Availability groups table, or between all groups in the table. For details, see Availability group map.
OS Cluster Information
The OS cluster information section displays the following details:
•
OS Cluster Name — the name of the OS cluster hosting the SQL Server instances that are enabled for AlwaysOn Availability Groups.
•
Quorum Type — the type of quorum used by the selected OS cluster; can have one of the following values:
•
0 — Node Majority.
The cluster is defined as healthy only if indicated so by more than 50% of the voting nodes in the cluster. For example, on a nine node cluster this quorum configuration can sustain up to four node failures.
•
1 — Node and Disk Majority.
This mode is similar to Node Majority quorum mode, with the addition of a shared disk cluster resource as a voting witness. More than 50% of the cluster’s possible votes must be affirmative in order to determine the cluster’s state as healthy, and connectivity from any node to that shared disk is also counted as an affirmative vote.
•
2 — Node and File Share Majority.
This mode is similar to Node Majority quorum mode, with the addition of a remote file share configured as a voting witness. More than 50% of the cluster’s possible votes must be affirmative in order to determine the cluster’s state as healthy, and connectivity from any node to that remote file share is also counted as an affirmative vote.
•
3 — No Majority: Disk Only.
A shared disk cluster resource is designated as a witness, and connectivity by any node to that shared disk is counted as an affirmative vote.
•
Quorum State — the state of the quorum used by the selected OS cluster; can have one of the following values:
•
0 — Unknown quorum state
•
1 — Normal quorum
•
2 — Forced quorum
High Availability groups table
The High Availability group table displays information about the availability and health of all availability groups configured in the instance and the databases they serve.
IMPORTANT: It is possible that one or more the availability replicas that take part in a group are not currently monitored. In such a case an indication appears near the replica in the availability group map. Click this indication to launch the instance monitoring wizard. For details, see Availability group map.
The High Availability group table includes the following columns:
•
HA Group Status — displays an icon that indicates the worst severity reported within the group from the group alarms.
•
Group Name — displays the High Availability group name.
•
Role — the current state of the database that runs on the selected instance and belongs to the selected group; either Secondary or Primary. When there is no value for Secondary/Primary, this column should indicate the status Off line.
•
Alarms — displays alarms related to the health of the availability group or its associated databases.
•
Max Estimated Recovery Time — displays the maximum value of the estimated recovery time parameter within all databases that belong to the selected group. If all databases are fully synchronized, this column displays a zero value.
•
Max Estimated Data Loss — displays the maximum value of the estimated data loss parameter within all databases that belong to the selected group. If all databases are fully synchronized, this column displays a zero value.
•
Explore — contains a shortcut icon which, when clicked, displays the Availability Group details pop-up.
Figure 24. Availability Group details pop-up.
This pop-up contains the Group Info, Availability Replica, and Databases tabs.
The Group Info tab includes the following fields:
•
Primary replica — the primary replica’s name.
•
Primary recovery health — indicates the primary replica’s recovery health, with the following possible values: In progress, Online or Null.
•
Synchronization health — summarizes the synchronization status of all replicas in the availability group:
•
Healthy — if all replicas are fully synchronized.
•
Partially healthy — if only some of the replicas are fully synchronized.
•
Not healthy — if no replica is fully synchronized.
•
Automated backup preference — indicates whether backups should run on the primary replica. The possible values are:
•
0 - Primary — backups should always be carried out on the primary replica.
•
1 - Secondary_only — performing backups on a secondary replica is preferable.
•
2 - Secondary — performing backups on a secondary replica preferable, but performing backups on the primary replica is acceptable if no secondary replica is available for backup operations. This is the default behavior.
•
3 - None — no preference about whether backups are performed on the primary replica or on a secondary replica.
The Availability Replica tab lists all replicas associated with the group. This tab includes the following fields:
•
Server Name — the name of the server.
•
Join State — can have the following values:
•
Not joined
•
Joined standalone — joined as a standalone instance
•
Joined Failover — joined as failover cluster instance
•
Role — either Primary or Secondary.
•
Operational State — indicates whether the replica is online, offline or in a pending stage. The possible values are as follows:
•
Pending failover
•
Pending
•
Online
•
Offline
•
Failed, no quorum
•
Null — replica is not local
•
Connected State — indicates whether a secondary replica is currently connected to the primary replica. Possible values are either Connected or Disconnected.
•
Recovery Health — indicates whether all databases joined to the availability group on the availability replica, are online, or are being recovered after a failover. Possible values are:
•
Online
•
In progress — recovering from a failover
•
Synchronization Health — summarizes the synchronization status of all replicas in the availability group:
•
Healthy — if all replicas are fully synchronized.
•
Partially healthy — if only some of the replicas are fully synchronized.
•
Not healthy — if no replica is fully synchronized.
•
Secondary Role Allow Connections — indicates whether an availability replica that is performing the secondary role (that is, a secondary replica) can accept connections from clients:
•
No — no connections are allowed to the databases in the secondary replica, and the databases are not available for read access. This is the default setting.
•
Read only — read-only connections have exclusive access to the databases in the secondary replica. All databases in the replica are available for read access.
•
All — all connections are allowed to the databases in the secondary replica for read-only access.
•
Availability Mode — can have either of the following values:
•
Asynchronous commit — the primary replica can commit transactions without waiting for the secondary to write the log to disk.
•
Synchronous commit — the primary replica waits to commit a given transaction until the secondary replica has written the transaction to disk.
•
Failover Mode — either Manual or Automatic.
The Databases tab displays all databases associated with the group. This tab includes the following fields:
•
Replica Server Name — the name of the server.
•
Database Name — the name of the database.
•
Database State — indicates whether the database is online, offline or in a pending stage. The possible values are as follows:
•
Online
•
Restoring
•
Recovering
•
Recovery pending
•
Suspect
•
Emergency
•
Offline
•
Synchronization State — indicates the data movement state, and can have one of the following values:
•
Not Synchronizing
•
Synchronizing
•
Synchronized
•
Reverting pages from undo log
•
Initializing Undo log
•
Synchronization Health — summarizes the synchronization status of all replicas in the availability group:
•
Healthy — if all replicas are fully synchronized.
•
Partially healthy — if only some of the replicas are fully synchronized.
•
Not healthy — if no replica is fully synchronized.
•
Estimated Recovery Time (sec) — the estimated time that the secondary replica will require in order to catch up with the primary replica.
•
Estimated Data Loss (sec) — the time difference of the last transaction log record between the primary and secondary replicas. Failure of the primary replica will result in the loss of all transaction log records within the time window.
•
Is Failover Ready — indicates whether the secondary database is synchronized in the cluster with the corresponding primary database. Possible values are Yes or No.
Availability group map
The Always On configuration supports one primary replica and up to four secondary replicas for each availability group. The Availability Group Map section allows viewing the dependencies between the various replicas that form the availability group selected in the table. In addition to indicating the name of the OS cluster where the availability replicas reside, the map displays the connections between the primary replica and the secondary replicas, showing on the left the replica that resides on the currently monitored instance.
* 
IMPORTANT: On each replica there is an indication of the highest severity level encountered on the replica. A replica that is not currently monitored is indicated by the icon. Click this icon to launch the wizard used for configuring an instance for monitoring.
Clicking the monitored replica displays a pop-up with the following tabs:
•
Group Info — identical to the Group Info tab on the pop-up that appears upon clicking the Explore column on the High Availability groups table. For details, see High Availability groups table.
•
HA Summary — displays the selected replica's availability-related alarms, including the severity and the error message.
•
Databases Health — a table displaying all alarms sent during the specified time range, including the severity and the error message.
•
Listeners — if one or more availability group listeners are configured, displays the listeners' names and their IP status, either Online or Offline.
Clicking any other replica displays a pop-up with the following tabs:
•
HA Summary — displays the selected replica's availability related alarms, including the severity and the error message.
•
Databases Health — a table displaying all alarms sent during the specified time range, including the severity and the error message.
To view the dependencies between all replicas in all availability groups, click View map for all availability groups. To return from this map to the Always On panel, click Back.

Using the Logs Drilldown

The Logs drilldown provides access to SQL Server Error Log and SQL Server Agent Error Log.
To configure which error logs generated by the SQL Server database are to be displayed in the Logs drilldown, use the Error Log Scanning view in the Databases Administration dashboard. For details, see Defining Error Log Filtering.

Reviewing the SQL Server Error Logs

The SQL Server Error Logs panel displays errors that are defined in the match list and can be viewed by a paging resolution of 1 hour. This panel contains the following items:
•
List of current and archived error logs
Use this list to choose the error log whose details are to be displayed in the Error Log table below.
•
Error Log table — displays the following parameters.
 
Table 57. Error log table parameters:
Column
Description
Date/Time
The date and time when the specified error was logged.
Process Info
Information about the process to which the entry in the error log refers.
Message
A brief description of the error.
The SQL Server Error Logs table displays the contents of the error log selected in the list above the table. This table is a snapshot of the selected error log, and is therefore not updated automatically.
To define and edit the alert rules for which Foglight for SQL Server scans the SQL Server Error Log, go to the Error Log Scanning view in the Databases Administration dashboard. To access this view, click the In-context actions button at the upper right side of the screen and then select Agent settings. For details, see Defining Error Log Filtering.
Documentos relacionados

The document was helpful.

Seleccionar calificación

I easily found the information I needed.

Seleccionar calificación