Open for Voting

SQL 2012 AlwaysOn Availability Groups for AppInsight for SQL

SQL Server 2012 has a new feature AlwaysOn Availability Groups. For HA in SQL this is very powerful. It would be great to be able to monitor the Availability groups and how it is setup. SQL Server 2012 management Studio has a dashboard built in for this but being able to have this info in SolarWinds would be even better.

  • I had our Solarwinds database on an AG but found that its high level of activity was affecting the performance on both nodes.
    All transactions have to be applied to both/all replicas so there is a load to 1. apply a trans and 2. replicate it.

    Further, as the level of transactions is high the log grows to a monster in AG.

    I have moved the Solarwinds database to an FCI cluster and set the database to Simple Recovery mode.

    It works a treat now as there is only one action for a transaction (no replication) and the log is not required to store queued activity.

    Hope this information helps anyone planning an install.

  • I is interesting that we are each on our own in this monitoring situation - even from Microsoft.
    I went to failover a AG Group the other day to do some maintenance on one node and the failover FAILED.
    The failure was due to a database that had fallen out of synch on the secondary.

    I now have to examine the primary AG Dashboard for each failover - messy when there are 7+ AGs on a server.

    We need a reporting tool that indicates that there is an affected database so we can attend to it and avoid a disaster if the node failed.

    I am very supportive of this request and offer my time to achieve a useful outcome for both parties.

    I understand that SW promotes DPA but that is not on every SQL instance in our environment and we need AG warnings at SAM level not in a drill down.

    Especially if you have 100+ AGs.

  • Just a thought.

    Having configured and been monitoring 20 AGs in Solarwinds for 6 months now I think all that is needed is to include HADR items in the "Virtual server" that Solarwinds already shows.

    For example, in the SAM "AppInsight for SQL" instance page there are details on  Application, Databases, AppStack Environment etc. Why not have a frame with HADR values such as Synchronization State, Connection State.

    Better still is for when the "Application" is added (or modified) you tick the "HADR or AOHA" flag and only those databases involved in the Availability group are listed.
    Currently all, in come cases 30, databases are listed and if there are only 3 or 4 databases in an Availability Group then we have to manually "Unmanage" the other 26 databases (Quite annoying).

    Further to enhance the monitoring of the AG the current "Status" of the database could be updated with values such as the Failover Readiness values such  as No Data Loss / Not Joined / Suspended.  These are the statuses we need to know when monitoring HADR environments.

    Hope you can help.

    Yes always willing to offer assistance to achieve great outcomes for DBAs and Wintel Admins, consequently the client.
    Cheers, Greg.

  • Bumping this as SQL Server 2016 Standard edition now has Always On Availability Groups (AAG).  We are migrating hundreds of databases to SQL 2016 Standard w/AAG.  This is would be a big win for us.