The obvious suggestion would be to check the memory utilization on your web server (generally this will be your primary polling engine). If it's a physical server, it'll be clear from the monitoring history (you are monitoring your SolarWinds servers, right? ) if you're having issues with memory usage. If it's a VM, check the memory assigned to it in your hypervisor, and check for memory contention. Whilst it's good to overcommit your compute resources to maximise your usage, it can cause problems when your VM thinks it has 8GB, when it's only really got a fraction of that and has to fight to get more .
If you've checked that, and it's all good, run a repair of the service using the SolarWinds Configuration Wizard (only check the box next to that service).
See if that clears things up for you.
Yes even we faced lot of issues in last month. randomly the Netflow service gets stopped, memory util goes to 90% on some day and then its back to normal.
Just few days back we had a situation wherein the monitoring went down, not sure for what reason(still need to analyse the logs). Only option was to run the config wizard and all came up fine post that.
I would suggest you check the Memory usage, space on DB server(if at all u can shrink any space, that would improve the performance).
And if u are seeing IIS crashing alerts more then i heard there is a specific hotfix available to resolve it.
Also check the performance of ur DB server.