Checkpoints on SQL Cluster

  • Anyone have any experience with a two node sql cluster running in front of a NAS? We've set one up and just put it into production. We're seeing some problems with checkpoints running the traffic on our gigabyte cards (running between SQL boxes and the NAS) way up, which in turn triggers blocking on the SQL server, which then runs up wait times on our IIS servers, maxing out the available work threads on those machines. Once a checkpoint finishes, the sql server screams through the queued up transactions and the IIS boxes clear out, but during the checkpoints, the IIS boxes are turning customers away.

    My Recovery setting is the SQL default (0) which is a sixty second checkpoint. Traffic on this site is not that heavy - 98 transactions per second. We cannot figure out what I've configured improperly to make something as basic as a checkpoint so painful. Any insights welcome.

    If this is better posted to another topic, point me the way...since checkpoints are basic admin...I figured I'd start here.

    Thanks much,

    Marcella

  • Are you getting a high number of packet losses on the NIC? If so are you running the NIC Full Duplex and have you tried Half-Duplex? I have not done this with a clustered server but this sounds more network related than SQL.

    "Don't roll your eyes at me. I will tape them in place." (Teacher on Boston Public)

Viewing 2 posts - 1 through 2 (of 2 total)

You must be logged in to reply to this topic. Login to reply