Kaseya Community

VSA performance issue after upgrade from 5.1 to 6.2

This question is answered

I just upgraded our server from 5.1 to 6.2 last week. Since then the performance of VSA web page degrades over a few days until the server is rebooted.

What we see is extreme slowness in going from page to page or a page won't open and shows a Javascript exception "unable to connect to server". Once that happens, the only thing we can do is to reboot the server. However, even when the web page performance is degraded, the server itself is very healthy and I do not notice any slowness. 

Support says it is because of large numbers of event log entries that are captured from several agents. But that condition was there before the upgrade so I am not convinced. Support also confirmed that our server hardware spec is more than sufficient. (Win2008 R2 SP1 x64, 16GB RAM, SQL 2008 R2 SP1 x64, 10GB allocated for SQL server)

Since this happens every few days, we are having to restart the server very often. Otherwise, we can not use remote control to support our customers, or run reports, or run procedures.

Has anyone experienced this issue?

Thanks in advance.

Akira

Verified Answer
  • Thanks for the suggestions. I worked with support and updated kserver.exe to the latest one, installed hotfixes and we haven't had the issue since (a week).

All Replies
  • Have you checked max degrees of parallelism on the SQL server? Set it to 1.

  • "Support says it is because of large numbers of event log entries that are captured from several agents."

    If that's what they actually said then all I can say is good grief. I know it sounds goofy but I sometimes can't help but wonder if they don't have some kind of contest going for their support people that is something like, "whoever closes the most tickets during the week gets free beer on Friday". We hear over and over how Kaseya is supposed to scale so well and then they tell you something like that?

    So, obviously, did you try turning off the event log captures temporarily to see if it makes any difference? Have you done any performance monitoring on the KServer and the SQL Server? I would run a baseline on both servers and then monitor the counters as the performance decreases to see if I could determine what was causing the degradation. And, of course check out all of the relevant log files for suspicious entries including the IIS logs. And, although I'm sure you've probably already taken care of this - ensure that both systems are free of malware.



    [edited by: zippo at 9:04 AM (GMT -7) on 5-5-2012] correct spelling
  • This sounds more like an IIS issue than an SQL issue. Whats the ram usage like on the IIS process vs SQL? Have you tried the performance guide from Kaseya and maybe tweaked the IIS connections.  There are a lot of confiugration options in the Application Pools within IIS which is where i would be looking.

    You mentioned that a server restart fixes it, so I am interested to know what other services restarts might fix it. Surely an IISreset, Kaseya service restart, SQL service restart would have the same result.

    I have a feeling SQL is maybe not leaving enough RAM for IIS. I had a similar problem when I had a Kaseya box with too little ram and stuck at the 4GB maximum. Since then we have migrated to a different service.  

  • Thanks for the suggestions. I worked with support and updated kserver.exe to the latest one, installed hotfixes and we haven't had the issue since (a week).