BigFix Root vs stand alone WebReports priority?

We have been running some processes that have truly “helped” us to find our bottlenecks in our BigFix environment.

One of the things we found was the load on our web reports server. When things get a bit taxed, sometimes we see it hop back and forth between the stand alone server and the root server. Is there a way to control which server it’s using? What is the mechanism that makes the switch?

Also, we have found when the desktop team is using the REST API hard, that it also affects Server Automation processing (which makes sense). When Web Reports is under load, it seems to not update from the root server until load diminishes. This causes other time sensitive things to be greatly delayed. So a good example is SA won’t see a job is complete and ready for the next step because Desktop is sending a boat load of queries to WR.

Is there a way to keep SA queries to use the WR on the root, and other processes use WR on the stand alone?

1 Like

The WR servers should be self contained. The WR server does contact the Root server for its data but has its own unique in memory session cache and DB so they can’t “share” the load to another WR server.

OK, how can I tell which WR server the root is pointing the REST API calls to? What makes it switch back and forth to serve the REST API?

What is the threshold on the WR server that would make it not sync with the root?

1 Like

Ahh, that is the question you are asking. This shouldn’t be happening. The order should be consistent as long as the Web Reports server table remains consistent. I suggest you open a PMR.

I think part of this request is to have a REST API session relevance query go to a specific WR server and not pick one at random to force queries that are not time sensitive to go to a specific WR server, while others go to another one dedicated for more time sensitive stuff.

1 Like

A little more than that @jgstew, I also would like to know what the limitation on syncing between WR and the root is. Disk wasn’t being hid hard, it was all CPU and memory. We doubled the memory and cores to 12 cores and 40gb of memory and it still couldn’t keep up. Is it the fact that the CPU just couldn’t keep up or is there a piece that lowers the priority for the sync process?

1 Like