After some ongoing testing of KSDU Profile application using the Kaseya Policy Module, we made some changes and reprocessed policies which seemed to trigger heavy KSDU activity the following night at 2 AM, when our scans and deployments happen.
This dragged on well into the next morning and throughout the day. It took some investigation to determine it was in fact KSDU; all Kaseya traffic streams out on one port, so we weren't able to traffic-shape the KSDU deployments, but what we think caused the issue was reprocessing policies after some policy changes, which then ended up causing rolling outages for many of our servers, and on one account severe server performance issues that were resolved by disabling the Kaseya Agent altogether. We then confirmed it was KSDU via Process Monitor from SysInternals on our K-Server.
We opened a case because our attempts at stopping the updates from the GUI were without success. Deleting deployment schedules seemed to only work temporarily, as what I believe was queued kept recreating Scan and Deploy schedules. As an emergency measure, removed all Deployment Profiles from the KPM policies that were applied to our endpoints, hoping this would stop deployments. Checking Policy compliance showed that yes, the profiles were removed from the policies in the field... Latent scans appeared to continue to push updates however, and we've been actively monitoring and deleting schedules as they've been re-appearing to prevent further issue.
Right now our KSDU is effectively turned off as best as we know how due to it created instability for several of our networks and bogging down performance for all remote connections; a core function of our business.
In hindsight, we probably could have avoided this with local file sources on our networks, but it didn't appear to be a problem before yesterday. It's my believe that reprocessing policies was the cause, but right now we're not sure whether to put the KPM Profiles back in the Profile Folder that's applied on our endpoints, fearing it might cause major, major network issues.
I'm posting this hear because I ran into the forums trying to find some information on KSDU... I found an out of date TechJam, and I write this as I re-watch, but I don't think there's anything relevant to deleting in-progress deployments or managing KSDU deployment bandwidth.
Has anyone else run into this problem? Has anyone else had their network brought down? Does anyone know how to get support in a server outage situation?
These are things I'd like to know.
Kaseya phoned us yesterday, 9 hours after opening the ticket. By then, we had somewhat resolved the issue, but support was able clarify some configuration settings that contributed to the problem.
This is effectively resolved, but having no ability to get assistance in our time of need has left a less than desirable taste in our mouth...
We have run into the exact same issue. What did Kaseya suggest to do in this situation? I'm currently having bandwidth issues and CANNOT find a way to cancel the deployments that are alreay in progress.
Goose, get a ticket opened with support so we can start looking at what is going on.
We've resolved some issues with KSDU/Policy Management in the past that I think were likely the cause of diijikul's issues last year.
Could you elaborate on what some of the settings were that you had to adjust? We ran into a very similar issue two years ago and stopped using KSDU. We are now considering deploying it again, but I want to try and avoid those same problems.