Aller au contenu principal

Essentials crashed and GCA did not report an outage

Commentaires

2 commentaires

  • Permanently deleted user
    Ethan,

     

    You don't say if you are running multiple servers (hub + agents) or not but we are running into a similar problem.  When I spoke to support about it, we found out that the Essentials server with Analytics agent installed was running out of memory.  This stopped all the logging of data including the status of the server.  Basically, Analytics isn't monitoring itself to see if it's operating correctly.  It never checked to see if the agent node had stopped responding and therefore it shows everything as okay (no news is good news!).

     

    If you open the elastic search webpage (http://localhost:19200/_plugin/head/) http://localhost:19200/_plugin/head/  and you don't see the cluster health in green at the top of the screen, something is wrong.

     

    Peter
    0
  • Permanently deleted user
    Hi Ethan and Peter,

     

    Thanks for the reply Peter, you are correct, checking the Elastic Search page is a good troubleshooting step!

     

    Ethan, was the error you were seeing ("The callback request was not handled. This is likely a problem with your session") in Manager, or the viewers?  Analytics monitors the REST endpoints for the applications, so if there is a problem on the endpoint, it will show up as an error in Analytics, providing it exceeds the threshold for an alarm (I think the default is 30 seconds without a reply, or if it receives an invalid http status code).  I might be wrong, but I don't think Analytics monitors Manager, just the sites and viewers.  I'll ask the developers and confirm that though!  

     

    Thanks!

     

    Danny
    0

Vous devez vous connecter pour laisser un commentaire.