SQL Monitor 8

MuradMurad Posts: 2 New member
edited April 1, 2023 5:48PM in SQL Monitor Previous Versions
Hi we have a old version of SQL monitor 8. Webserver: 8.0.21.17565 and Base monitor 8.0.21.17565.
We monitor the usual counters for SQL Server 2012. 
The Red-gate server has its own server and connects out to our production SQL2012 cluster.

During a 4 hour period we noticed that monitoring of CPU, Memory and DISK IO stopped for several hours and then restarted automatically. 
During that time the counter for Wait states was still connected.

Does anyone have any advice why these particular counters stopped and then restarted automatically.
I am not the person who has set up the system just inherited it


Answers

  • Alex BAlex B Posts: 1,131 Diamond 4
    Hi @Murad

    It's likely that the primary node of the Windows Server failover cluster (WSFC) changed for those 4 hours (and the original primary went down) and then was failed back afterwards.  In that version, the page will only show the machine level data for the currently active WSFC node that the SQL FCI is running on so the current primary node was down for those 4 hours (but the SQL FCI was still up as it had failed over to the other node).  If you were to fail the monitored entity (the SQL 2012 cluster) over to the other node, you would see different machine level data.

    In later versions, it shows the data for the WSFC node that was active at that time (so the gap would instead have machine level data from the other node).

    Kind regards,
    Alex
    Product Support Engineer | Redgate Software

    Have you visited our Help Center?
  • MuradMurad Posts: 2 New member
    Thanks for you comment.
    I have checked the event viewer from the failover cluster manager but we have no events of failover for the cluster.
    I can see viewed the roles section and owner nodes of the 2 servers as not have changed.
    Also i checked the SQL server failover last and that was last done in 2021.

    Can you suggest a better place to check if you still think it was this ?
    Or any other suggestions ?
  • Alex BAlex B Posts: 1,131 Diamond 4
    Hi @Murad,

    I'm not sure then, that is an explanation that fits the data and the version shown.  We'd probably need to look at the logs for that time period, but we only really support the last two major versions.  So I'd suggest opening a support ticket if you have a valid support and upgrades contract.

    If not, you could download the log files from Configuration > Retrieve all log files and look in the Base Monitor_Sampling.<date>.log files for any connection issues for that entity at that time period.  The log files are in UTC time.

    Kind regards,
    Alex
    Product Support Engineer | Redgate Software

    Have you visited our Help Center?
Sign In or Register to comment.